Nuke12.2v3 ReferenceGuide PDF

REFERENCE GUIDE
VERSION 12.2v3
Nuke™ 12.2v3 Copyright © 2020 The Foundry Visionmongers Ltd. All Rights Reserved. Use of this document and the Nuke software is subject to
an End User License Agreement (the "EULA"), the terms of which are incorporated herein by reference. This document and the Nuke software may
be used or copied only in accordance with the terms of the EULA. This document, the Nuke software and all intellectual property rights relating
thereto are and shall remain the sole property of The Foundry Visionmongers Ltd. ("Foundry") and/or Foundry's licensors.
The EULA is available here: https://www.foundry.com/eula
The Foundry assumes no responsibility or liability for any errors or inaccuracies that may appear in this document and this document is subject to
change without notice. The content of this document is furnished for informational use only.
Except as permitted by the EULA, no part of this document may be reproduced, stored in a retrieval system or transmitted, in any form or by any
means, electronic, mechanical, recording or otherwise, without the prior written permission of The Foundry. To the extent that the EULA
authorizes the making of copies of this Reference Guide, such copies shall be reproduced with all copyright, trademark and other proprietary
rights notices included herein. The EULA expressly prohibits any action that could adversely affect the property rights of The Foundry and/or The
Foundry's licensors, including, but not limited to, the removal of the following (or any other copyright, trademark or other proprietary rights
notice included herein):
Nuke™ compositing software © 2020 The Foundry Visionmongers Ltd. All Rights Reserved.
Nuke™ is a trademark of The Foundry Visionmongers Ltd.
Digital Domain ® is a registered trademark of Digital Domain, Inc.
Primatte™ keyer tool © 1997-2020 Photron USA, Inc. All Rights Reserved.
Primatte™ is a trademark of IMAGICA Corp.
Primatte™ patent is held by IMAGICA Corp.
In addition to those names set forth on this page, the names of other actual companies and products mentioned in this User Guide (including, but
not limited to, those set forth below) may be the trademarks or service marks, or registered trademarks or service marks, of their respective
owners in the United States and/or other countries. No association with any company or product is intended or inferred by the mention of its
name in this User Guide.
ACADEMY AWARD ® is a registered service mark of the Academy of Motion Picture Arts and Sciences.
Linux ® is a registered trademark of Linus Torvalds.
Windows ® is the registered trademark of Microsoft Corporation.
Mac, Mac OS X, macOS, High Sierra, Mojave, Catalina, Shake, Final Cut Pro, and QuickTime are trademarks of Apple, Inc., registered in the U.S. and
other countries.
Adobe ® and Photoshop ® are either registered trademarks or trademarks of Adobe Systems Incorporated in the United States and/or other
countries.
Maya ® is a registered trademark of Autodesk, Inc., in the USA and other countries.
Houdini ® is a registered trademark of Side Effects Software, Inc.
Boujou is a trademark of 2d3 Ltd.
3D-Equalizer is a trademark of Science.D.Visions.
RenderMan ® is a registered trademark of Pixar.
Cineon™ is a trademark of Eastman Kodak Company.
Stereoscopic images courtesy of Mr. Henry Chung, HKSC (http://www.stereoscopy.com/henry/). Images illustrating warping and morphing
courtesy of Ron Brinkmann (http://www.digitalcompositing.com). Images from “The Day After Tomorrow” ©2004 courtesy of and copyright by 20th
Century Fox. Images from "Stealth" courtesy of and copyright by Sony Pictures Inc. Images from "xXx" ©2002 courtesy of and copyright by
Columbia Pictures Industries. Images illustrating GridWarpTracker and EdgeExtend courtesy of Little Dragon Studios. All rights reserved by their
respective owners in the United States and/or other countries.
Thank you to Diogo Girondi for providing icons for the Nuke user interface and Tim Baier for proofreading.
The Foundry Visionmongers Ltd.

5 Golden Square
London
W1F 9HT
UK
Rev: Wednesday, September 16, 2020

Contents
Reference Guide 22
Image Nodes 26
CheckerBoard 26
ColorBars 27
ColorWheel 28
Constant 29
CurveTool 30
Profile 34
Read 35
UDIM Import 57
Viewer 58
Write 67
Draw Nodes 101
Dither 101
DustBust 103
Flare 105
Glint 111
Grain 114
Grid 116
LightWrap 120
MarkerRemoval 123
Noise 125
Radial 129
Ramp 133
Rectangle 136
Roto 139
RotoPaint 160
ScannedGrain 183
Sparkles 187
Text 191
Time Nodes 200
Add 3:2 pulldown 200
AppendClip 201
FrameBlend 203
FrameHold 205
FrameRange 206
Kronos 207
NoTimeBlur 216
OFlow 217
Remove 3:2 pulldown 224
Retime 225
SmartVector 229
TemporalMedian 233
TimeBlur 235
TimeClip 237
TimeEcho 240
TimeOffset 241
TimeWarp 243
VectorGenerator 244
VectorToMotion 249
Channel Nodes 250
AddChannels 250
ChannelMerge 251
Copy 254
Remove 258
Shuffle 260
ShuffleCopy 262
Color Nodes 264
Add 264
Clamp 266
ClipTest 269
CMSTestPattern 271
ColorCorrect 273
ColorLookup 277
ColorMatrix 280
Colorspace 282
ColorTransfer 286
Crosstalk 287
Exposure 289
Expression 293
Gamma 303
GenerateLUT 306
Grade 309
HistEQ 314
Histogram 315
HSVTool 317
HueCorrect 320
HueShift 323
Invert 325
Log2Lin 327
MatchGrade 330
MinColor 342
Multiply 344
OCIOCDLTransform 346
OCIOColorSpace 349
OCIODisplay 351
OCIOFileTransform 353
OCIOLogConvert 356
PLogLin 358
Posterize 361
RolloffContrast 363
Sampler 366
Saturation 367
SoftClip 369
Toe 371
Truelight 373
Vectorfield 373
Filter Nodes 378
Blur 378
Bilateral 380
BumpBoss 383
Convolve 386
Defocus 389
DegrainBlue 391
DegrainSimple 392
Denoise 393
DirBlur 403
DropShadow 407
EdgeBlur 408
EdgeDetect 411
EdgeExtend 412
Emboss 415
Erode (blur) 418
Erode (fast) 419

Erode (filter) 421
Glow 422
GodRays 425
Inpaint 427
Laplacian 431
LevelSet 433
Matrix 434
Median 436
MotionBlur 438
MotionBlur2D 442
MotionBlur3D 443
Sharpen 445
Soften 448
VectorBlur 450
VolumeRays 456
ZDefocus 460
ZSlice 471
Keyer Nodes 474
ChromaKeyer 474
Difference 478
HueKeyer 479
IBKColor 480
IBKGizmo 482
Keyer 487
Keylight 490
Primatte 498
Ultimatte 510
Merge Nodes 521
Absminus 521
AddMix 521
Blend 523
ContactSheet 524
CopyBBox 526
CopyRectangle 526
Dissolve 527
In 529
Keymix 529
LayerContactSheet 531
Matte 532
Max 532
Merge 532
MergeExpression 538
Min 544
Multiply 544
Out 544
Plus 545
Premult 545
Screen 546
Switch 546
TimeDissolve 547
Unpremult 548
ZMerge 549
Transform Nodes 551
AdjustBBox 551
BlackOutside 552
CameraShake 552
Card3D 556
CornerPin2D 562
Crop 568
GridWarp 570
GridWarpTracker 578
IDistort 585
LensDistortion 588
Mirror 607
PointsTo3D 608
Position 611
Reconcile3D 611
Reformat 617
SphericalTransform 621
SplineWarp 631
Stablilize 639
STMap 639
Tile 642
Tracker 644
Transform 658
TransformMasked 662
TVIScale 666
VectorCornerPin 668
VectorDistort 674
3D Nodes 678
AmbientOcclusion 678
ApplyMaterial 679
Axis 681
BasicMaterial 687
BlendMat 690
Camera 692
CameraTracker 698
Card 721
CrosstalkGeo 729
Cube 731
Cylinder 735
DepthGenerator 738
DepthToPoints 744
DepthToPosition 746
Diffuse 747
Direct 748
DisplaceGeo 754
Displacement 758
EditGeo 763
Emission 766
Environment 767
FillMat 772
GeoSelect 772
Light 774
LogGeo 783
LookupGeo 785
MergeGeo 787
MergeMat 788
ModelBuilder 789
ModifyRIB 796
Normals 799
Phong 802
Point 805
PointCloudGenerator 812
PoissonMesh 819
PositionToPoints 821
PrmanRender 823
ProceduralNoise 830
Project3D 832
RadialDistort 833
RayRender 835
ReadGeo 843
Reflection 853
Refraction 854
Relight 855
ScanlineRender 857
Scene 865
Specular 866
Sphere 868
Spot 872
TransformGeo 879
Transmission 883
Trilinear 885
UVProject 888
UVTile 890
Wireframe 891
WriteGeo 893
Particles Nodes 897
ParticleBlinkScript 897
ParticleBounce 903
ParticleCache 911
ParticleCurve 913
ParticleDirectionalForce 917
ParticleDrag 921
ParticleEmitter 926
ParticleExpression 937
ParticleGravity 944
ParticleLookAt 948
ParticleMerge 952
ParticleMotionAlign 953
ParticlePointForce 957
ParticleSettings 961
ParticleSpawn 962
ParticleSpeedLimit 970
ParticleToGeo 974
ParticleTurbulence 976
ParticleVortex 980
ParticleWind 985
Deep Nodes 991
DeepColorCorrect 991
DeepCrop 994
DeepExpression 996
DeepFromFrames 998
DeepFromImage 1000
DeepMerge 1001
DeepRead 1003
DeepRecolor 1009
DeepReformat 1010
DeepSample 1014
DeepToImage 1015
DeepToPoints 1016
DeepTransform 1017
DeepWrite 1018
Views Nodes 1023

Anaglyph 1023
JoinViews 1024
MixViews 1025
OneView 1026
ReConverge 1027
ShuffleViews 1028
SideBySide 1029
Split and Join 1030
Metadata Nodes 1032
AddTimeCode 1032
CompareMetaData 1033
CopyMetaData 1034
ModifyMetaData 1036
ViewMetaData 1037
ToolSets Nodes 1039
Create 1044
Delete 1044
Other Nodes 1045
Assert 1045
AudioRead 1045
Backdrop 1046
BlinkScript 1047
BurnIn 1050
DiskCache 1052
Dot 1054
Group 1055
Input 1056
LiveGroup 1056
LiveInput 1058
NoOp 1059
Output 1060
PostageStamp 1060
Precomp 1061
Root 1063
StickyNote 1071
All plugins 1071
FurnaceCore Nodes 1072
F_Align 1072
F_DeFlicker2 1075
F_ReGrain 1076
F_RigRemoval 1080
F_Steadiness 1084
F_WireRemoval 1088
Reference Guide
A reference guide for all the parameters within each node in Nuke. It does not give you any instructions on
using Nuke. For details on installing and using Nuke, refer to Nuke's online help.
Note: For the most up-to-date information, please see the Nuke product page and the latest
Nuke Online Help on our website at www.foundry.com.
Getting Help
Nuke features several forms of help, in a variety of locations.
Viewing Tooltips
Most controls offer concise instructions in the form of tooltips. To display the tooltips, move your mouse
pointer over an interface control or a node parameter.
To view a tooltip for a node, move your mouse pointer over the ? icon in the node's Properties panel.
REFERENCE GUIDE
22
|
Viewing the Context-Sensitive Online Help

The context-sensitive online help includes a comprehensive description of each node and its parameters.
To display these descriptions in your default web browser, click the ? icon in the node's properties panel.
Selecting the Help Source

In Nuke, the context-sensitive help content may differ depending on what you've selected as the help
source in the Preferences.
To select the help source:
1. Press Shift+S on the Node Graph to open the Preferences dialog.

2. Go to the Behaviors > Documentation tab.
3. Set documentation source to:
• local - Use Nuke's built-in help system. This is included in the Nuke installation and does not require
an Internet connection.
Note: When you click ? in a node's Properties panel, Nuke searches the following locations for
HTML files with the same name as the requested node (for example, blur.html):
1. Directories contained in the NUKE_PATH environment variable.

2. The /$HOME/.nuke and /$HOME/.nuke/Documentation directories.
3. Your local plugins directory.
• On Windows, this is either C:\Program Files\Common Files\Nuke\12.2\plugins\ or C:\Program
Files (x86)\Common Files\Nuke\12.2\plugins\
• On Mac, this is /Library/Application Support/Nuke/12.2/plugins/
• On Linux, this is /usr/local/Nuke/12.2/plugins/
4. The Nuke installation directory.
• foundry - Use the help system available on our website. This contains the most up-to-date
information, but requires an Internet connection.
REFERENCE GUIDE
23
|
Note: On Windows, you may have to add a firewall program or port exception to view the most
up-to-date help from our website. If the connection is blocked, Nuke falls back to the local copy.
• custom - Use your own custom help system.

4. If you set documentation source to local, you can either:
• enable auto port to automatically assign a free documentation server port, or
• use local port to specify a local documentation server port manually and port range to define a
range of ports to attempt. Typically, the local port value should be 1024 or higher. Setting the value
to 0 causes a port to be automatically assigned.
Note: To be able to use the online help from multiple sessions of Nuke running at the same
time, you must enable auto port.
5. If you set documentation source to foundry:

• the foundry host field displays the documentation server host address.
• use foundry port to specify the documentation server port.
6. If you set documentation source to custom:
• set custom host to your documentation server host address. This must be an Internet Protocol
version 4 (IPv4) address.
• use local port to specify a local documentation server port. Typically, the local port value should be
1024 or higher. Setting the value to 0 causes a port to be automatically assigned.
• use custom port to specify a custom documentation server port.
Using the Help Menu

You can click the main Help menu to access the following:
• Key Assignments - a list of hot keys.
• Documentation - the Nuke Online Help, the Nuke Developer Kit (NDK), and documentation for using
Python, TCL, and expressions in Nuke.
• Release Notes - important updates to features and bug fixes.
• Training and Tutorials - Nuke tutorial videos, and a list of other training resources.
• Nukepedia - an online resource containing useful information about all things Nuke.
• Mailing Lists - information on Nuke-related e-mail lists.
• Plug-in Installer - open the Nuke plug-ins page on our website, which gives you easy access to a large
selection of plug-ins for Nuke.
REFERENCE GUIDE
24
|
Contacting Customer Support

Should questions arise that the documentation fails to address, you can visit the Support Portal at
support.foundry.com.
REFERENCE GUIDE
25
Image Nodes
Image nodes deal with loading, viewing, and rendering image sequences as well as creating built-in Nuke
elements, such as checkerboards and color wheels.
CheckerBoard
Generates a checkerboard image that you can use as a placeholder for a texture or background.
Note: If you enable proxy mode the board may not match the full resolution version, because
the boxes are rounded to the nearest pixel.
Inputs and Controls

Control Knob Default Value Function
(UI) (Scripting)
CheckerBoard Tab
format format Dependent on Sets the checkerboard’s format. You can select a preset
Project format or create a custom size using the new or edit
Settings options.
size boxsize 64 Sets the size of the checkerboard boxes in pixels.
color 0 color0 N/A Sets the color of the boxes within the checkerboard
rgba starting from the bottom left-hand corner:
• color 0 - the box at 0,0 and then every alternate box
and row.
and row.
REFERENCE GUIDE
26
ColorBars |

(UI) (Scripting)

color 1 color1 N/A and row.
rgba
color 2 color2 N/A and row.
rgba
color 3 color3 N/A

rgba
line color linecolor 1 Sets the color of the delineating line between boxes on
the checkerboard.
line width linewidth 0 Sets the width, in pixels, of the delineating line between
boxes on the checkerboard.
centerline centerlinecolor N/A Sets the color of the center lines on the checkerboard.
color rgba
centerline centerlinewidth 3 Sets the width, in pixels, of the center lines on the
width checkerboard.
ColorBars
Generates a Society of Motion Picture and Television Engineers (SMPTE) color bars test pattern, useful for
color management.
Inputs and Controls

Control (UI) Knob Default Function
(Scripting) Value
ColorBars Tab
format format Dependent Sets the colorbar’s format. You can select a preset
on Project format or create a custom size using the new or edit
REFERENCE GUIDE
27
ColorWheel |

(Scripting) Value
Settings options.
bar intensity barintensity 0.5 Sets the color intensity of the bars.
PAL PAL disabled When enabled, NTSC specific colors are removed and
only PAL colors are displayed.
ColorWheel
Generates a color wheel image.
Inputs and Controls

(UI) (Scripting)
ColorWheel Tab
format format Dependent on Sets the color wheel’s format. You can select a preset
Project format or create a custom size using the new or edit
Settings options.
channels channels rgba The color wheel only contains these channels.
If you set this to something other than all or none, you

can use the checkboxes on the right to select individual
channels.
Center centerSaturation 0 Sets the HSV saturation level in the center of the color
Saturation wheel.
Edge edgeSaturation 1 Sets the HSV saturation level at the edges of the color
Saturation wheel.
Center centerValue 1 Sets the HSV value level in the center of the color wheel.
Value
REFERENCE GUIDE
28
Constant |

(UI) (Scripting)
Edge edgeValue 1 Sets the HSV value level at the edges of the color wheel.
Value
gamma gamma 0.45 Sets the overall gamma level of the color wheel.
rotate rotate 0 Sets the amount of rotation to apply to color position in

the color wheel. Negative values produce clockwise
rotation and vice-versa.
fill format fillFormat enabled When enabled, the color wheel fills the selected format
up to either the x or y axes, depending which is smaller -
the color wheel always retains a 1:1 aspect ratio.
When disabled, use the area controls to specify the color

wheel’s area.
area xyrt area N/A Sets the area of the color wheel when fill format is
(xywh) disabled.
Note: Click wh to toggle between xyrt and

xywh area controls.
Constant
Generates an image where every pixel is the same color. This includes pixels outside the image area as
well as inside it.
Inputs and Controls

(Scripting) Value
Constant Tab
REFERENCE GUIDE
29
CurveTool |

(Scripting) Value
channels channels rgba The constant only contains these channels.

channels.
color color 0 Sets the color of the constant.
format format Dependent Sets the constant’s format. You can select a preset
on Project format or create a custom size using the new or edit
Settings options.
frame range first 1 If you intend to animate the constant, sets the frame
range within which the animation is contained.
last 1
CurveTool
You can use this node to analyze and track the following aspects of the input sequence:
• the size and position of black areas in the sequence
• average pixel values in the sequence
• exposure changes in the sequence, and
• brightest and dimmest pixels in the sequence.
Based on the analysis, the node creates an animation curve. You can then use this curve data to drive
effects elsewhere. For instance, you can add matching flicker to a CG render.
Inputs and Controls

Connection Connection Function
Type Name
Input unnamed The image sequence to analyze.
REFERENCE GUIDE
30
CurveTool |
Control (UI) Knob (Scripting) Default Function

Value
CurveTool Tab
Curve Type operation Avg Sets the curve type to pass down the node tree:
Intensities • Auto Crop - locates black regions (or any color
you pick) around the edges of the frame sequence
and tracks their size and position over time. This is
useful for running a Crop node to remove
unnecessary outer pixels and speed up the
calculation.
• Avg Intensities - obtains the average pixel values
in a frame sequence and then matches that
intensity elsewhere. It takes the first value in the
frame range and the next value selected, adds
them together and divides by two, returning the
average between the two. You might want to use it
to match the background plate’s fire flicker in the
smoke in the foreground plate, for example.
• Exposure Difference - analyzes the exposure
changes in the frame sequence. It takes the first
value in the frame range and the next value
selected, and returns the difference between the
two. You can use the results to match the same
exposure elsewhere.
• Max Luma Pixel - tracks the brightest and
dimmest pixels in the frame sequence. This can be
useful in the following case, for example. Let’s say
you have a night-time sequence depicting a
person moving inside a dark house holding a
flashlight, and want to add lens flare on the
moving flashlight. Knowing where the brightest
pixel is located over time allows you to match-
move the lens flare and position it correctly
without having to manually animate it.
Go! go N/A Click to begin analysis of the selected Curve Type.
REFERENCE GUIDE
31
CurveTool |

Value
AutoCrop
color color 0 Sets the color detected during analysis.
Intensity ccrange 0 Sets how far the color can deviate from the selected
Range color and still be cropped.
Intensity
# frames for avgframes 2 Sets the range of frames that each frame analyzed is
base average compared against. For example, a value of 1 would
compare each frame to the frame following it,
whereas a value of 5 would compare each frame to
the following 5 frames.
Note: The higher frame range you use, the

more accurate and time-consuming the
calculation becomes.
All
channels channels rgba The selected analysis is only applied to these

channels.
If you set this to something other than none, you

can use the checkboxes on the right to select
individual channels.
ROI xyrt ROI Dependent Defines the area to analyze. By default, the ROI is set
(xywh) on input to the entire input image, but you can drag the ROI
format indicator in the Viewer or adjust the xyrt values
manually.
AutoCropData Tab
autocropdata autocropdata Dependent Displays the analysis results when Curve Type is set
xyrt (xywh) on input to AutoCrop. To see the animation curve, right-click
format on a parameter field and select Curve editor.
REFERENCE GUIDE
32
CurveTool |

Value
IntensityData Tab
intensitydata intensitydata 0 Displays the analysis results when Curve Type is set
to Avg Intensities, or Exposure Difference. To see
the animation curve, right-click on a parameter field
and select Curve editor.
MaxLumaData Tab
Maximum maxlumapixdata 0, 0 Displays the coordinates of the pixel with the

Luminence highest luminance value on the current frame.
Pixel xy
Maximum maxlumapixvalue 0 Displays the highest luminance value on the current

Luminence frame.
Pixel Value
Minimum minlumapixdata 0, 0 Displays the coordinates of the pixel with the lowest
Luminence luminance value on the current frame.
Pixel xy
Minimum minlumapixvalue 0 Displays the lowest luminance value on the current

Luminence frame.
Pixel Value
Python Tab (These controls are for Python callbacks and can be used to have Python functions
automatically called when various events happen in Nuke.)
before beforeRender none These functions run prior to starting rendering in

render execute(). If they throw an exception, the render
aborts.
before each beforeFrameRender none These functions run prior to starting rendering of
frame each individual frame. If they throw an exception,
the render aborts.
after each afterFrameRender none These functions run after each frame is finished
frame rendering. They are not called if the render aborts. If
they throw an exception, the render aborts.
after render afterRender none These functions run after rendering of all frames is
REFERENCE GUIDE
33
Profile |

Value
finished. If they throw an error, the render aborts.
render renderProgress none These functions run during rendering to determine

progress progress or failure.
Profile
The Profile node allows you to measure the performance of a script by inserting the node at any point in
the node tree and then using the Profile pane to determine the output. If you have more than one Profile
node in a script, use the profile node dropdown in the Profile pane to choose which point in the script to
calculate.
See Nuke's online help for more information.
Inputs and Controls

Value
Profile Tab
channels channels rgba The profiling data is only calculated for these
channels.
If you set this to something other than all or none,

you can use the checkboxes on the right to select
frame range frame_range Dependent Sets the frame range for which the profiling data is
on Project calculated.
Settings
open profile open_profile_panel N/A Click to open the Profile pane, which determines
panel what profiling data is collected.
REFERENCE GUIDE
34
Read |

Value

aborts.
the render aborts.

Read
This node loads images from disk, using the native resolution and the frame range for the sequence. It
converts all imported sequences to Nuke’s native 32-bit linear RGB colorspace.
The Read node supports multiple file formats, such as Cineon, TIFF, Alembic, PSD, OpenEXR, HDRI, DPX,
R3D, ARRIRAW, DNG, and RAW camera data. It uses the file name extension to determine the format.
You can also create a Read node by pressing R on the Node Graph.
See also DeepRead, ReadGeo, AudioRead, and UDIM Import.
REFERENCE GUIDE
35
Read |
Inputs and Controls

Value
Read Tab
File file dependent Sets the file path and name of the file to read. For
on file frame numbers, you can use #### for each digit or
the printf-style formatting %04d.
Localization localizationPolicy from auto- Sets the local file caching behavior. Copies of the files
Policy localize are stored in a specified local folder for faster access
path times:
• on - the files are cached, regardless of location, as
long as the limit to (GB) limit is not breached.
• from auto-localize path - the files are cached if
they reside in the auto-localize from directory, as
• on demand - only localize these source clips when
you manually update them. See Nuke's Online Help
for more information.
• off - the files are never cached, regardless of
location.
Update updateLocalization N/A When this Read node is set to Localization Policy >
on demand, click Update to manually localize the
files from the source files.
Format format dependent Sets the size of the file. Typically it is guessed
on file correctly from the file header, but you can change it
to set a different pixel aspect ratio if required.
Proxy proxy none Sets the file path and name of a relevant proxy image.
This proxy image is used if proxy mode is on and the
required resolution is less than or equal to the proxy
format.
Proxy Format proxy_format 1024x778 The file specified in proxy is read if the down-rez of
the format is smaller or equal to this value. Typically it
REFERENCE GUIDE
36
Read |

Value
is guessed correctly from the file header, but you can

change it to set a different pixel aspect ratio if
required.
Frame Range first dependent The first frame number of a sequence to display.
on file
before hold Sets the behavior of frames before the first frame
specified:
• hold - select to show a still picture of the first frame
of the frame range.
• loop - select to start over and keep looping the
span of the frame range outside the first frame of
the frame range.
• bounce - select to play the span of the frame range
backwards and forwards between the frame range
limits.
• black - select to display a black frame outside of the
first frame.
last dependent The last frame number of a sequence to display.

on file
after hold Sets the behavior of frames after the last frame
specified:
• hold - select to show a still picture of the last frame
of the frame range.
span of the frame range outside the last frame of
the frame range.
limits.
• black - select to display a black frame outside of the
last frame.
Frame frame_mode expression Sets the frame mode:
REFERENCE GUIDE
37
Read |

Value
• expression - Lets you enter an expression in the

field on the right. The expression changes the
relation between the current frame and the frame
read in. For example, if your clip begins from
image.0500.rgb and you want to place this first
frame at frame 1 rather than frame 500, you can use
the expression frame+499. This way, 499 frames are
added to the current frame to get the number of the
frame that’s read in. At frame 1, image.0500.rgb is
read in; at frame 2, image.0501.rgb is read in; and so
on. Another example of an expression is frame*2.
This expression multiplies the current frame by two
to get the number of the frame that’s read in. This
way, only every other frame in the clip is used. At
frame 1, image.0002.rgb is read in; at frame 2,
image.0004.rgb is read in; at frame 3,
image.0006.rgb is read in; and so on.
• startat - Lets you enter a start frame number in the
field on the right. This specifies the frame where the
first frame in the sequence is read in. In other words,
all frames are offset so that the clip starts at the
specified frame. For example, if your sequence
begins from image.0500.rgb and you enter 1 in the
field, image0500.rgb is read in at frame 1. Similarly,
if you enter 100 in the field, image0500.rgb is read
in at frame 100.
• offset - Lets you enter a constant offset in the field
on the right. This constant value is added to the
current frame to get the number of the frame that’s
read in. For example, if your clip begins from
image.0500.rgb and you want to place this first
499 as the constant offset. This way, 499 is added to
the current frame to get the frame that’s read in. At
REFERENCE GUIDE
38
Read |

Value
image.0501 is read in, and so on. You can also use

negative values as the constant offset. For example,
if you use the value -10, Nuke will subtract ten from
the current frame to get the frame that’s read in. At
frame none Depending on the frame mode, you can enter a start
frame, an offset, or an expression here.
Original origfirst dependent Displays the original first frame.

Range on file
origlast dependent Displays the original last frame.

on file
Missing on_error error Sets Nuke’s behavior when there is a problem with
Frames frames in the file:
• error - display an error in the Viewer.
• black - replace suspect frames with a black frame.
• checkerboard - replace suspect frames with a
checkerboard.
• nearestframe - replace suspect frames with the
nearest good frame.
reload reload N/A Click to re-read the image from disk.
Colorspace colorspace dependent Sets the lookup table (LUT) used to convert from this
on file file to the internal values used by Nuke.
default determines the LUT from the size and type of

data in the file, and other information from the file
header.
Premultiplied premultiplied disabled When enabled, Nuke corrects the color of the partially
transparent pixels produced by most renderers by
dividing color data by the alpha channel before
converting from the colorspace, and then multiplying
by the alpha channel afterwards.
REFERENCE GUIDE
39
Read |

Value
Raw Data raw disabled When enabled, Nuke does not convert the data.
For most file formats this is the same as linear

colorspace but, for some, it may disable other
processing such as conversion from YUV.
Auto Alpha auto_alpha disabled When enabled, if the Read produces RGB but no
alpha channel, then assume the alpha should be 1 if
it's requested later on.
OCIO Tab
Context
key1 key1 none OCIO Contexts allow you to apply specific LUTs to
individual shots.
value1 value1 none
key2 key2 none
value2 value2 none
key3 key3 none
value3 value3 none
key4 key4 none
value4 value4 none
Sequence Tab
import import sequence N/A Click to display a browser to locate an import

sequence sequence script which builds a frame sequence list
from the contents of an EDL file.
These are not standard EDL files, such as CMX. Each

line in the file simply defines a separate image
sequence to import in the form:
imagesequence.%04d.ext in out head tail
Where in and out are the start/end frame numbers of

the clip and head/tail are the number of frames to
REFERENCE GUIDE
40
Read |

Value
trim from the start and end.
Each sequence in the list is expanded into the

sequence control so that each line contains the name
of the file to read for that frame:
line 1 = frame 1, line 10 = frame 10.
Finally, the script replaces the Read > file text with:
[lindex [knob sequence] [frame]]
This looks up the file name from the sequence

control and sets the first frame to 1 through to the
last frame in the expanded list.
frame sequence none Most easily used in conjunction with the import
sequence sequence button. Each line contains the name of the
file to read for that frame:
line 1 = frame 1, line 10 = frame 10.
Set the Read > file control to [lindex [knob

sequence] [frame]] to look up the file name from this
control. This allows you to use a single Read to access
multiple sequences of files where the naming does
not follow any set numeric sequence, rather than
using multiple Read nodes with a single file sequence
per Read.
Metadata Tab
metadata shownmetadata N/A Displays the metadata of any file read in using a Read
viewer node. Once you know which keys exist in the file, you
can reference them in expressions.
The metadata displayed depends on the file type. For

example, a .jpg might only contain input/ keys,
whereas QuickTimes contain input/ and quicktime/
keys.
search metadatafilter N/A Filter the metadata by entering text in this field. By
REFERENCE GUIDE
41
Read |

Value
metadata for default, only metadata containing the keys and values
entered are displayed. You can change this behavior
using the within control.
File Type Specific Controls

These controls are context sensitive depending on which format you intend to render out.
ARI

Value
ARI Options (ARI and MXF)
Load arriReset N/A Click to reset the ARI Options to those from
settings the file metadata.
from
metadata
Local GPU arriLocalGPUName N/A Displays the GPU used for rendering when Use
GPU if available is enabled. Local GPU
displays Not available when:
• Use CPU is selected as the default blink
device in the Preferences.
• no suitable GPU was found on your system.
• it was not possible to create a context for
processing on the selected GPU, such as
when there is not enough free memory
available on the GPU.
You can select a different GPU, if available, by

navigating to the Preferences and selecting an
alternative from the default blink device
REFERENCE GUIDE
42
Read |

Value
dropdown.
Note: Selecting a different GPU

requires you to restart Nuke before
the change takes effect.
Use GPU if arriUseCuda disabled When enabled, rendering occurs on the Local
available GPU specified, if available, rather than the CPU.
Note: Enabling this option with no

local GPU allows the script to run on
the GPU whenever the script is
opened on a machine that does have
a GPU available.
You should also select this if you
wish to render from the command
line with the --gpu option.
See Nuke's Online Help Installation pages for

more information on the GPUs supported by
your operating system.
Clip Settings
Colorspace arriColorspace dependent Sets the colorspace used to convert from raw
on clip data into the Viewer colorspace.
Exposure arriExposureIndex Sets the overall sensitivity of pixels from the

Index (ISO) raw image.
White arriWhiteBalance Controls the relative RGB coolness and

Balance (K) warmth. Higher values increase the warmth.
Tint arriTint Adjusts the level of tints to compliment the

White Balance control.
REFERENCE GUIDE
43
Read |

Value
Lens arriLensSqueeze Sets the amount of anamorphic lens squeeze

Squeeze applied to the image.
Unsqueeze arriUnsqueezeAnamorphic disabled When enabled, unsqueezes the raw file while
Anamorphic converting it to the Viewer colorspace.
Note: This control is generally not

required because the file metadata
sets the pixel aspect ratio
automatically.
Processing
Debayer arriDebayerMode dependent Sets the debayering mode applied to reads. If

Mode on clip Proxy Playback is enabled, the debayer is
ignored.
ADA 5 Fine- arriAda5FinetuningRed Fine-tunes the ADA 5 debayering in the red,

tuning green, and blue channels.
arriAda5FinetuningGreen
arriAda5FinetuningBlue
Aspect Ratio arriAspectRatio Sets the output aspect ratio independent of

the pixel aspect ratio.
Resolution arriResolution Set the resolution and aspect ratio that the clip
is scaled to, cropping the image if required.
Sharpness arriSharpness Adjusts the sharpness of the image by

introducing contrast at edges.
Detail arriDetail Adjusts the amount of detail by affecting fine

structures.
Note: The Detail control can

introduce aliasing at higher values.
REFERENCE GUIDE
44
Read |

Value
Look Editor
Use Look arriUseLookEditor enabled When enabled, the Look Editor settings can
Editor be used to modify the image.
Note: The Look Editor settings

cannot be enabled when the
Colorspace control is set to LogC or
Scene Linear. Only Video color
spaces use Look Editor settings.
Saturation arriSaturation 1 Increases or decreases the overall saturation of

the image.
Slope/Offset/Power
Slope arriSlope 1,1,1 Adjusts overall the Slope, Offset, and Power.
Offset arriOffset 0,0,0
Power arriPower 1,1,1
Proxy arriProxyPlayback Off Setting proxy to anything other than Off

Playback speeds up read times, but the Debayer Mode
is ignored.
DNG

Value
DNG Options
color space dng_colorspace sRGB Sets the colorspace used during debayering. The
standard Read node colorspace control is changed to
match this setting automatically.
REFERENCE GUIDE
45
Read |

Value
Note: To view the data before debayering,

enable the raw data control.
force linear dng_force_linear disabled When enabled, overrides the specified color space
transfer transfer function to force it to use linear.
function
For some cases, such as sRGB, this results in slightly
faster read times and sets the colorspace on the Read
to linear as well.
As a result of forcing linear, all render settings are

applied in linear space, rather than the native space,
so there may be differences in how the controls
respond.
white dng_white_ As Shot Applies preset values to the temperature and tint
balance balance controls, depending on the setting chosen:
• As Shot - attempts to read the temperature and
tint values from the source file's metadata.
• Daylight to Flash - standard presets in common
usage.
• Custom - adjusting the temperature and tint
controls manually sets white balance to this value.
temperature dng_temperature dependent Defines the temperature, in Kelvin, and tint of the
on file image white point:
• temperature - in general, runs from blues at low
tint dng_tint values to yellows at high values.
• tint - in general, runs from greens at low values to
purples at high values.
exposure dng_exposure 0 Controls the amount of exposure data available on

top of the baseline exposure.
shadows dng_shadows 0 Controls the amount of low-level image data

available at render time.
REFERENCE GUIDE
46
Read |
EXR

Value
EXR Options
offset offset_negative_ enabled The .exr format allows the image’s lower left corner
negative display_window to start at any position, including negative x axis
display values. Nuke does not support this type of format, but
window compensates by offsetting the image or reducing the
format size on both sides:
• enabled - the image is offset so that the .exr
image’s left side start at x=0.
• disabled - the format is reduced on both sides by
the negative x amount, as if that area was overscan.
do not attach noprefix disabled When enabled, metadata keys are read as they are,
prefix without attaching a prefix to them.
When disabled, the prefix exr is attached to metadata

keys to make them distinct from other metadata in
the tree.
edge pixels edge_pixels plate detect Sets how to treat pixels at the edges of the data
window, or bounding box in Nuke terms, when
reading .exr files:
• plate detect - if the bounding box and format
match exactly, then repeat all edges. Otherwise, add
black at the edges.
• edge detect - for each matching edge, repeat the
edge pixels. Add black at mismatched edges.
• repeat - always repeat the edge pixels outside the
bounding box.
• black - always add black pixels outside the
bounding box.
REFERENCE GUIDE
47
Read |
MOV

Value
MOV Options
Note: The mov32Reader and Writer has been removed for macOS Catalina (10.15) and
deprecated under Windows, Linux, and macOS Mojave (10.14). The mov32: prefix is still
currently available on deprecated OS, but unsupported. We recommend using the standard
Nuke Read/Write controls, as mov32 is scheduled for removal in a future release.
YCbCr matrix mov64_ycbcr_ dependent Sets the way Y’CbCr is converted to RGB. Rec 601 and
matrix on the file Rec 709 follow the ITU.BC specifications.
Data Range mov64_decode_ default Sets the minimum and maximum color range values
video_levels for decoded DNxHD files. If your source file was
encoded with a reduced range (0-240 for 8-bit or 0-
960 for 10-bit), set source range to Video Range to
decode the full range (0-255 for 8-bit or 0-1023 for
10-bit).
first track mov64_first_track_ enabled When enabled, only the first view present in the file is
only only displayed, regardless of how many views the file
contains.
When disabled, and there is more than one view set

up in the project settings, you can switch between
views using the buttons above the Viewer.
[Metadata] mov64_no_prefix disabled When enabled, the metadata accompanying the file is
Do not saved without the quicktime prefix.
attach prefix
Note: QuickTime container-specific

metadata always carries the prefix.
When disabled, all metadata is prefixed with

quicktime for easy identification.
REFERENCE GUIDE
48
Read |

Value
[Metadata] mov64_match_ enabled When enabled, Nuke attempts to clean up meta keys
Match key meta_format into the standard Nuke metadata format.
format
When disabled, the metadata is passed down the
Node Graph unchanged.
Video Codec mov64_codec_ N/A Displays the codec used to encode the clip.
name
PSD

Value
PSD Options
Breakout breakout N/A Breaks out the .psd file into separate layers and
Layers recombines them with a number of PSDmerge nodes.
The blend modes/operations in the merge nodes are
approximated and do not match Photoshop® exactly.
It’s recommended that all masks and adjustment

layers are rasterized in Photoshop® before they’re
imported into Nuke.
R3D
Control (UI) Knob (Scripting) Default Value Function
R3D Options
Image r3dImagePipeline dependent on file Sets the image pipeline to use in the Read
Pipeline node, IPP2, IPP2 Custom, or legacy. The
selected Image Pipeline defines which
controls are available in the Properties
REFERENCE GUIDE
49
Read |
panel and is initially set by the file's

metadata.
See IPP2 Controls and Legacy Controls.
Load r3dLoadRMD N/A Click to reload control values from the

Settings .rmd file metadata.
from RMD
Local GPU r3dLocalGPUName N/A Displays the GPU used for rendering when
Use GPU if available is enabled. Local GPU
displays Not available when:
• no suitable GPU was found on your
system.
processing on the selected GPU, such as
when there is not enough free memory
available on the GPU.
You can select a different GPU, if available,

by navigating to the Preferences and
selecting an alternative from the default
blink device dropdown.

requires you to restart Nuke
before the change takes effect.
Use GPU if r3dUseCUDA disabled When enabled, rendering occurs on the

available Local GPU specified, if available, rather
than the CPU.
REFERENCE GUIDE
50
Read |
Note: Enabling this option with

no local GPU allows the script to
run on the GPU whenever the
script is opened on a machine
that does have a GPU available.
wish to render from the
command line with the --gpu
option.
See Nuke's Online Help Installation pages

for more information on the GPUs
supported by your operating system.
Decoding
Decode r3dDecodeResolution full high Sets the output resolution decoded from
Resolution the source file. High quality takes longer to
decode.
Rocket r3dRocketResolution dependent on When Decode Resolution is set to rocket

Resolution format custom, sets the resolution values if you're
using a Red Rocket card for decoding. The
default values are equal to the source
image.
Use ROCKET r3dUseROCKET disabled When enabled, an installed Red Rocket

card is used for decoding if available.
HDRx
HDR Mode r3dHDRMode A frame Sets how the high and low exposure tracks
are blended for HDR clips:
• A frame - only displays the normally
exposed track with standard motion blur.
• X frame - only displays the under
exposed, sharper reference track.
• simple blend - enables the Blend Bias
REFERENCE GUIDE
51
Read |
control allowing you to blend of high and

low exposure according to the bias.
• magic motion - a more sophisticated
automatic blend mode with a sharper
image and more natural motion blur.
Blend Bias r3dBlendBias 0 When HDR Mode is set to simple blend,

controls the bias between low and standard
exposure values, where -1 is fully the X
frame and +1 is fully the A frame.
IPP2 Controls
Demosaic
Denoise r3dIPP2Denoise 0 Controls the amount of noise reduction

applied to the image if Chroma Noise
Reduction is disabled.
Chroma r3dIPP2Cnr enabled When enabled, applies chroma noise

Noise reduction instead of the standard .r3d
Reduction Denoise.
Note: Chroma Noise Reduction

is only available with Decode
Resolution set to full high or
half high.
Primary
Color Space r3dIPP2Colorspace REDWideGamutRGB This control is display only for the IPP2
Image Pipeline.
Gamma r3dIPP2GammaCurve Log3G10 This control is display only for the IPP2
Curve Image Pipeline.
ISO r3dIPP2ISO 800 Sets the overall sensitivity of pixels from

the raw image.
REFERENCE GUIDE
52
Read |
Exposure r3dExposureAdjust 0 Increases or decreases the overall exposure.

Adjust This control is identical to FLUT Control in
the legacy Image Pipeline.
Kelvin r3dIPP2Kelvin 5600 Controls the relative RGB coolness and

warmth. Higher values increase the warmth.
Tint r3dIPP2Tint 0 Adjusts the level of yellow and green tints.

This control can help in environments
where the ambient light source contains
high levels of yellow and green, such as
fluorescent lights.
Grading
3D LUT r3d3DLUT N/A Specifies the path to a LUT file.
Enable 3D r3dEnable3DLUT disabled When enabled, the LUT file specified in the
LUT 3D LUT field is applied to the image.
CDL r3dCDL N/A Specifies the path to an ASC CDL file.
CC ID r3dCCID N/A Specifies which color correction should be

applied from CDL files containing more
than one correction.
Enable CDL r3dEnableCdl N/A When enabled, the CDL file specified in the
CDL field is applied to the image.
Slope r3dSlope 1,1,1,0 Adjusts the Slope, Offset, Power, and

Saturation manually.
Offset r3dOffset 0,0,0,0
If you have applied a CDL to the image,
Power r3dPower 1,1,1,0 these values are set from the selected CC
ID.
Saturation r3dCDLSaturation 1
Contrast r3dIPP2Contrast 0 Increases or decreases the overall contrast.
REFERENCE GUIDE
53
Read |
Note: Contrast is disabled when

Gamma Curve is set to Half Float
Linear.
Export r3dExportPipeline full graded IPP2 Controls whether or not creative color
Pipeline mode decisions are applied on export:
• primary raw development - all Grading
settings are ignored on export.
• full graded IPP2 mode - all Grading
settings are applied on export.
Legacy Controls
Look
Color r3dColorVersion v2 Sets the color version used by the decoder.

Version
Color Space r3dColorSpace dependent on clip Sets the colorspace used to convert from
raw data into the Viewer colorspace.
Note: Colorspaces listed in

brackets are deprecated in the
current R3D SDK and may be
removed in subsequent releases.
Gamma r3dGammaCurve dependent on clip Sets the gamma curve applied to the raw
Curve image data from the .r3d file.
Note: Gamma Curves listed in

brackets are deprecated in the
current R3D SDK and may be
removed in subsequent releases.
D.E.B. r3dDEB enabled When enabled, reduces the red grain that
can affect midtones, darks, and shadows
REFERENCE GUIDE
54
Read |
using DRAGON Enhanced Blacks.
Note: D.E.B. is only available for

clips shot with a DRAGON sensor,
or later, with Decode Resolution
set to full high or half high.
Kelvin r3dKelvin 5600 Controls the relative RGB coolness and

warmth. Higher values increase the warmth.
Tint r3dTint 0 Adjusts the level of yellow and green tints.

This control can help in environments
where the ambient light source contains
high levels of yellow and green, such as
fluorescent lights.
ISO r3dISO 800 Sets the overall sensitivity of pixels from

the raw image.
FLUT r3dFLUTControl 0 Allows you to adjust the ISO value without

Control clipping highlights or crushing shadows.
Note: FLUT Control is only

available with Color Version v2 or
higher.
Shadow r3dShadow 0 Sets the image black point.
Note: Shadow is only available

with Color Version v2 or higher.
Setting Gamma Curve to Half
Float Linear disables Shadow.
DRX r3dDRX 0 Dynamic Range Extension (DRX) can be

adjusted to recover clipped highlights.
REFERENCE GUIDE
55
Read |
Note: DRX is disabled for clips

shot with a DRAGON sensor.
Saturation r3dSaturation 1 Increases or decreases the overall

saturation, contrast, and brightness of the
image.
Contrast r3dContrast 0 Contrast
Brightness r3dBrightness 0 Brightness
Note: Saturation, Contrast, and Brightness are disabled when Gamma Curve is set to Half
Float Linear.
Exposure r3dExposure 0 Increases or decreases the overall exposure.
RGB r3dRGBGain 1,1,1 Increases or decreases the gain for the red,
green, and blue channels.
Note: RGB gain is disabled when

Gamma Curve is set to Half Float
Linear.
Sharpness r3dSharpness 0 Sharpens the image to compensate for the

OLPF low pass filter.
Note: Sharpness is only

applicable when Decode
Resolution is set to full high or
half high.
Denoise r3dDenoise 0 Controls the amount of noise reduction

applied to the image.
Detail r3dDetail 2 Sets the amount of detail extracted from

the image, separate from the Sharpness
REFERENCE GUIDE
56
UDIM Import |
control.
Note: Denoise and Detail are only applicable when Decode Resolution is set to full high.
Lift/Gamma/Gain
Lift r3dLift 0,0,0,0 Adjusts the overall black level (dark tones
and shadows).
Gamma r3dGamma 1,1,1,0 Adjusts the overall mid tones.
Gain r3dGain 1,1,1,0 Adjusts the overall highlights.
Note: Lift, Gamma, and Gain are only available with Color Version v2 or higher. Setting
Gamma Curve to Half Float Linear disables Lift, Gamma, and Gain.
UDIM Import
UDIM Import lets you import a set of texture patches that follow the UDIM numbering scheme and
quickly apply them to the surface of a 3D object. Clicking on UDIM Import opens the standard Read File
(s) file browser. Browsing to, and selecting, a sequence containing valid UDIM patches opens the UDIM
Import dialog. The UDIM Import dialog lists all the UDIM patches in a sequence and provides some
options for managing them. If you select a file that doesn’t contain a valid UDIM number, a dialog displays
explaining the problem.
UDIM Import creates a Read node for every patch in the sequence and appends a UVTile node (which
allows you to modify a patch’s coordinates in UV space) to each one. For more information on the UVTile
node see Nuke's online help.
The UDIM Import dialog lets you disable individual patches in a sequence, add additional files and set
how the Read nodes (for each individual patch) are displayed in the Node Graph. Enabling postage stamp
displays a thumbnail view of the patch in the Read node. Checking group nodes adds the nodes to a
single group node for clarity.
REFERENCE GUIDE
57
Viewer |
Viewer
The Viewer node displays the render output of any connected process nodes in the Viewer panel. It
doesn’t alter data in any way, but simply allows you to view the effect of any changes you make upstream.
You can place as many Viewer nodes in a script as you wish, which allows you to simultaneously view
multiple outputs. You can also connect the output from up to ten process nodes into single Viewer node,
and then cycle through the various displays. This allows you to easily compare an image before and after
processing by a given effect.
You can also create this node by:

• selecting Image > Viewer from the Nuke Toolbar, or
• pressing Ctrl/Cmd+I.
Inputs and Controls

Type Name
Input unnamed The process node upstream to output to the Viewer. You can
connect up to ten (0-9) nodes at once.

Value
Viewer Tab
frame_range frame_range none Sets the first and last frame in the playback
range. You can enter frame numbers here or
Cmd+drag the red triangles in the timeline.
frame_range_lock enabled Click to unlock or lock playback to the frame

numbers specified in frame_range.
<< frame_increment 10 Sets the number of frames to step through

>>increment using the Viewer << and >> buttons.
channels channels rgba Sets the channels to display in the viewer.
REFERENCE GUIDE
58
Viewer |

Value
You can also select a channel from a different

layer to display in the alpha channel.
clip_warning clip_warning no warnings When set to exposure, any pixels that are
outside the 0-1 range are Zebra-striped in
the Viewer.
gain gain 1 Sets the amount of gain to multiply the

image by before display.
gamma gamma 1 Sets the amount of gamma correction to

apply to the image.
Gamma correction is applied after the gain,

viewerprocess, and any custom
inputprocess.
apply LUT to rgb_only disabled When enabled, the LUT is applied to color
color channels only.
channels only
downrez downrez 1 Sets the amount of down scaling applied

before display. factor to speed up rendering.
You can use downrez to speed up rendering,

but bear in mind that it is multiplied into any
global proxy scale that may also be in effect.
ignore pixel ignore_pixel_aspect disabled When enabled, output to the Viewer ignores
aspect the file’s pixel aspect ratio and always draws
using square pixels.
prevent auto zoom_lock disabled When enabled, the Viewer is prevented from
zoom automatically adjusting the zoom level to fit
new images into the Viewer when swapping
between different sized images.
show show_overscan disabled When enabled, any overscan present is

overscan displayed in the Viewer.
enable full full_frame_processing disabled When enabled, force Nuke to render every
REFERENCE GUIDE
59
Viewer |

Value
frame scanline in the image. Compared to the

processing default mode, this may take slightly longer
initially and requires more space in the
Viewer cache, but once Nuke has cached the
frames you require, you can pan around and
change the zoom level without breaking the
cache or affecting playback.
overscan overscan 100 Sets the amount of overscan to show in the

Viewer when show overscan ins enabled.
safe zone safe_zone no guides Sets the safe zone overlay to apply to the
image:
• no guides
• title safe
• action safe
• format center
format format_center disabled When enabled, the center of the format is

center marked with a cross in the Viewer.
mask region masking_ratio square Sets the aspect ratio of the mask specified in
outside ratio the mask mode dropdown.
mask mode masking_mode no mask Sets the mask overlay to apply to the image:
• no mask
• lines
• half
• full
center f-stop center_fstop 8 Sets the f-stop to display when gain is set to
1. You can use zero to show stops as +/-n.
REFERENCE GUIDE
60
Viewer |

Value
Note: If you want a non-integer

value, enter a formula to get an
accurate value, otherwise the
display does not round to normal f-
stop names. For instance, instead of
5.6 you would use pow(2,2.5) to get
5.65685
shutter angle shutter_angle 180 Sets the camera shutter angle from 0-360
degrees.
This is used in combination with the Viewer's

playback fps setting to compute exposure
values.
gl buffer gl_buffer_depth float Sets the OpenGL buffer depth and enables
depth the use of the GPU for the Viewer process
and input process:
• byte - converts to 8-bit with error diffusion.
• half-float - converts to 16-bit (half) float.
• float - uses a full 32-bit floating point
texture (may be very slow on some cards).
In half and full float modes, the GPU can be

used to apply Viewer effects like gamma and
the LUT in a Viewer process.
Note: You can choose a default

value for this setting in the
Preferences or by using
knobDefault() in a startup script.
use GPU for useGPUForViewer enabled When enabled, Viewer effects (like gain,
viewer when gamma, and viewer process node) are
possible applied using the GPU when possible.
REFERENCE GUIDE
61
Viewer |

Value
The ’when possible’ clause is due to the fact

that in some cases, like when broadcast
monitor output is enabled or the GL buffer
depth is 8-bit, effects must be computed in
the CPU.
use GPU for useGPUForInputs enabled When enabled, compute nodes connected to
input when the Viewer input on the GPU when possible,
possible though not all nodes have a GPU
implementation.
Normally, the Viewer only attempts to run its

own effects (like gain, gamma, and viewer
process node) on the GPU.
Note: If nodes are computed on

the GPU, the color values displayed
in the Viewer are inaccurate since
they show the color from the last
node computed in the CPU, prior to
transferring the image into the
graphics card.
disabled GPU disableGPUDitherForViewer disabled When enabled, GPU dithering is allowed in

dithering the Viewer (which affects half-float and float
modes).
Note: When the Viewer AB mode is changed to wipe or stack, the state of the GPU
acceleration controls is stored, GPU acceleration is turned off, and GPU acceleration is disabled.
When the Viewer AB mode is changed back to default, GPU acceleration is re-enabled and the
state of the GPU acceleration controls is restored.
viewer viewerProcess sRGB Sets the Viewer Process operation applied to

process the viewed image before it is displayed on
the screen. It is typically used to convert from
REFERENCE GUIDE
62
Viewer |

Value
linear to your monitor's colorspace.
input process input_process_node VIEWER_ Sets the name of a node to process through
INPUT for viewing when the IP button is activated. If
the node does not exist, the IP button has no
effect.
input_process disabled Applies the input process node before

displaying the image.
To define the node to use, enter its name in

the input process field or, from the Node
Graph right-click menu, select Edit > Node >
Use as Input Process.
input process viewerInputOrder before Sets when the input process node specified
order viewer is applied:
process • before viewer process.
• after viewer process.
enable MonitorOutEnable disabled When enabled, the Viewer is output to the

monitor connected monitor device.
output
Note: You cannot enable this

control if no monitor output
device is specified.
use video MonitorOutLegalRange disabled When enabled, transform the monitor output
legal range to the legal video range.
monitor MonitorOutDevice None Sets the monitor output device to receive

output the Viewer output:
device • None
• Digital Cinema Desktop
• Firewire
monitor MonitorOutMode - Sets the display mode for the specified
REFERENCE GUIDE
63
Viewer |

Value
output mode monitor output device:

• none - no selections are available.
• D - select a Digital output device from the
dropdown.
• A - select a Firewire output device from the
dropdown.
monitor MinorOutPixelFormat - Set the pixel format output for the specified
output pixel monitor output device, dependent on the
format card's manufacturer.
viewer viewerSelectorMode rectangle Sets the selection mode for the current
selector selection Viewer:
mode • rectangle selection
• ellipse selection
• lasso selection
3D Tab
grid position grid_origin 0, 0, 0 Sets the position of the grid in the 3D Viewer
xyz on the x, y, and z axes.
display display unchanged Adjust the display characteristics of 3D

objects. These settings don’t affect the
render output of the scene; these are for
display purposes only in the 3D Viewer.
• off - hides all 3D objects.
• wireframe - displays only the outlines of
object geometry.
• solid - displays all object geometry with a
solid color.
• solid+lines - displays the geometry as solid
color with individual object’s geometry
outlines.
• textured - displays only the surface texture
of objects.
REFERENCE GUIDE
64
Viewer |

Value
• textured+lines - displays the wireframe

plus the surface texture.
• unchanged - doesn't change object display
modes.
Note: This setting can override

individual object settings.
transparency use_alpha enabled When enabled, objects appear transparent

where their alpha is < 1.
headlamp gl_lighting disabled When enabled, objects are shaded with a

spotlight attached to the viewpoint to make
their shape visible.
alpha gl_alpha_clip 1 Sets the alpha value used by the

clipping transparency control which determines if an
object is opaque or not.
polygon gl_poly_offset 1 Sets the visibility of lines in hidden-line

offset mode.
show show_points disabled When enabled, all points are displayed in the
3D Viewer.
show_point_num disabled When enabled, all points are numbered in

the 3D Viewer.
show_point_info disabled When enabled, all points are labeled with xyz
coordinates in the 3D Viewer.
show_point_normals disabled When enabled, all available point normals are

displayed in the 3D Viewer.
show_point_uvs disabled When enabled, all available point uvs are

show_vertex_num disabled When enabled, all vertices are labeled in the

3D Viewer.
REFERENCE GUIDE
65
Viewer |

Value
show_vertex_normals disabled When enabled, all available vertex normals

are displayed in the 3D Viewer.
show_vertex_uvs disabled When enabled, all available vertex uvs are

show_prim_num disabled When enabled, all primitives are labeled in

the 3D Viewer.
show_prim_normals disabled When enabled, all primitive normals are

labeled in the 3D Viewer.
show_prim_bbox disabled When enabled, all primitive bounding boxes

are labeled in the 3D Viewer.
grid_display enabled When enabled, a grid displays in the 3D

Viewer.
default lens default_lens 1 Sets the default lens (focal length divided by
film width) to use in 3D mode when no
camera is selected.
near near 0.1 Sets the limit of the OpenGL near clipping
plane. Objects closer to the camera that this
value are not rendered the 3D Viewer.
far far 1000 Sets the limit of the OpenGL far clipping
plane. Objects farther away from the camera
that this value are not rendered the 3D
Viewer.
The far clipping plane can be moved farther

away if objects in the scene are farther away.
In this case, the near plane is also moved so
that the ratio between near and far planes is
preserved.
skew order skew_order XY Sets the order in which skew operations are
applied.
• XY - Skew on the X axis is applied before
skew on the Y axis.
REFERENCE GUIDE
66
Write |

Value
• YX - Skew on the Y axis is applied before

skew on the X axis.
Soft Selection
Soft Select softSelect disabled When enabled, soft selection is applied to

any selection made in the Viewer.
When disabled, selections are made in the

same way as legacy versions of Nuke.
Falloff Radius softSelectRadius 1 Controls the radius of the selection falloff

added to the initial selection. Increasing the
value includes selections from farther away
from the initial points or faces.
Falloff Curve softSelectActiveCurve S-shaped Sets the curve used to describe falloff farther
away from the initial points or faces.
curves softSelectCurves none Adjusts the preset curve selected in the left-
hand panel. You can also add your own curve
to control falloff using the + button.
Adding a custom curve adds it to the Falloff

Curve dropdown.
Write
This node renders the result of all upstream nodes and saves the result to disk. You would usually place
one Write node at the bottom of the compositing tree to render the final output. However, Write nodes
have both input and output connectors, so you can embed them anywhere in the compositing tree.
The Write node supports multiple file formats, such as Cineon, TIFF, QuickTime, Alembic, OpenEXR, HDRI,
and DPX.
Note that this node executes all renders at the currently active scale: either full- or proxy-resolution. To
toggle between these, press Ctrl/Cmd+P.
REFERENCE GUIDE
67
Write |
You can also create this node by pressing W on the Node Graph.
See also WriteGeo.
Inputs and Controls

Value
Write Tab
channels channels rgb Sets the channels to render.

you can use the controls on the right to select
file file none Sets the file path and name of the file to render. For
frame numbers, you can use #### for each digit or
the printf-style formatting %04d.
proxy proxy none Sets the file path and name of a relevant proxy
image. This proxy image is used if proxy mode is
on and the required resolution is less than or equal
to the proxy format.
frame frame_mode expression Sets the frame mode:

relation between the currently processed frame
and the numbering of the frame written out. The
resulting file name for the current frame is
displayed on the Write node in the Node Graph.
For example, if your clip begins from frame 500
and you want to name that frame image.0001.exr
rather than image.0500.exr, you can use the
expression frame-499. This way, 499 frames are
subtracted from the current frame to get the
number for the frame written out. Frame 500 is
written out as image.0001.exr, frame 501 is written
out as image.0002.exr, and so on. Another
REFERENCE GUIDE
68
Write |

Value
example of an expression is frame*2. This

expression multiplies the current frame by two to
get the number of the frame that’s written out. At
frame 1, image.0002.exr is written out; at frame 2,
image.0004.exr is written out; at frame 3,
image.0006.exr is written out; and so on.
• startat - Lets you enter a start frame number in
the field on the right. This specifies the frame
number given to the first frame in the sequence.
The numbering of the rest of the frames is offset
accordingly. For example, if your sequence begins
from frame 500 and you enter 1 in the field, frame
500 is written out as image.0001.exr, frame 501 as
image.0002.exr, and so on. Similarly, if you enter
100 in the field, frame 500 is written out as
image.0100.exr.
• offset - Lets you enter a constant offset in the
field on the right. This constant value is added to
the current frame to get the number for the frame
that’s written out. For example, if your clip begins
from frame 500 and you want to render this first
frame as image.0001.exr rather than
image.0500.exr, you can use -499 as the constant
offset. This way, 499 is subtracted from the current
frame to get the number for the frame that’s
written out. At frame 500, image.0001.exr is
written out; at frame 501, image.0002.exr is written
out, and so on.
frame none Depending on the frame mode, you can enter a

start frame, an offset, or an expression here.
colorspace colorspace dependent Sets the lookup table (LUT) used to convert from
on file type the internal values used by Nuke to the values
written to the file.
The default value is determined from the type of
REFERENCE GUIDE
69
Write |

Value
file and the size and type of data written to it.
premultiplied premultiplied disabled When enabled, Nuke corrects the color to

reproduce the partially transparent pixels created by
some renders by dividing color data by the alpha
channel before converting to the colorspace, and
then multiplying by the alpha channel afterwards.
raw data raw disabled When enabled, Nuke does not convert the data.
For most file formats this is the same as linear

colorspace but, for some, it may disable other
processing such as conversion from YUV.
views views dependent When you’re working with stereo footage, select
on the required view to render.
Compositing
environment
Project
Settings
file type file_type none Sets the rendered file format manually, enabling
type specific controls. See Nuke's online help for
more information.
Note: If file type is left blank, Nuke

attempts to guess the format and disables
any file type specific controls.
create create_directories disabled When enabled, the render operation creates the
directories directory structure specified in the file control
automatically, if it doesn't already exist.
When disabled, attempting to write to non-existent

directories displays an error.
render order render_order 1 When multiple nodes are rendered at once, they are
REFERENCE GUIDE
70
Write |

Value
sorted into increasing order by this number.
Render Render N/A Click to display the pre-Render setup window.
frame range first 1 Sets the first frame of a sequence to render.
last 1 Sets the last frame of a sequence to render.
limit to range use_limit disabled When enabled, only frames within the frame range
are rendered.
Note: If the specified frames are outside

the sequence range, the Write node
behaves as if it is disabled.
read file reading disabled When enabled, the newly written file is passed
down the node tree instead of the input.
missing on_error error Sets Nuke’s behavior when there is a problem with
frames frames in the file:
• error - display an error in the Viewer.
• black - render suspect frames with a black frame.
• checkerboard - render suspect frames with a
checkerboard.
• nearestframe - render suspect frames with the
nearest good frame.
Reload reload N/A Click to re-read the image from disk.
read all lines read_all_lines disabled When enabled, upstream Read nodes read whole
frames, rather than line by line. This is currently
limited to .dpx files, but can speed up scripts
containing I/O-heavy Reads.
OCIO Tab
Context
REFERENCE GUIDE
71
Write |

Value
key1 key1 none OCIO Contexts allow you to apply specific LUTs to
individual shots.
value1 value1 none
key2 key2 none
value2 value2 none
key3 key3 none
value3 value3 none
key4 key4 none
value4 value4 none

aborts.
the render aborts.

File Type Specific Controls

These controls are context sensitive depending on which format you intend to render.
REFERENCE GUIDE
72
Write |
CIN

(Scripting) Value
edge code edge_code none Sets the sequence’s edge code, carried in the metadata,
in the following format: 00 00 00 0000 0000 00.
DPX

(Scripting) Value
data type datatype 10 bit Sets the bit depth of the rendered .dpx files:
• 8-bit
• 10-bit
• 12-bit
• 16-bit
fill fill disabled When enabled, 10- and 12-bit data is compressed by
removing unused parts of the image.
big endian bigEndian enabled When enabled, the rendered file is big-endian, rather
than native-endian. Big-endian files take longer to
render, but some applications only accept big-endian
files.
transfer transfer (auto detect) Set the Transfer header in the rendered .dpx files.
By default, Nuke attempts to set the header according to

the LUT used, but the transfer control allows you to
override this.
REFERENCE GUIDE
73
Write |
EXR

Value
write ACES write_ACES_compliant_ disabled When enabled, the rendered .exr files are ACES
compliant EXR compliant.
EXR • The datatype and compression controls are
disabled. ACES compliant .exr files written from
Nuke are always 16-bit half float with no
compression.
• An extra metadata field, chromaticities is
written into the file.
autocrop autocrop disabled When enabled, the bounding box is reduced to

the none zero area of the image.
Note: Autocrop is slow to process and

generally not required, though some
applications are able to read
autocropped images more quickly.
datatype datatype 16 bit half Sets the bit depth of the rendered .exr files:
• 16-bit half
• 32-bit float
compression compression Zip (1 Sets the compression type to apply to the

scanline) rendered file.
Note: Selecting DWAA or DWAB as

the compression setting allows you to
specify the compression level using a
slider.
heroview heroview dependent Sets the view labeled as the main view in multi-
on view projects.
Compositing
REFERENCE GUIDE
74
Write |

Value
environment
Project
Settings
metadata metadata default Determines what metadata is included with the

metadata rendered file:
• no metadata
• default metadata
• default metadata and exr/*
• all metadata except input/*
• all metadata
do not attach noprefix disabled When enabled, unknown metadata keys are
prefix written into the file as they are.
When disabled, unknown metadata keys have the

prefix nuke attached to them when they are
written into the file.
interleave interleave channels, Sets the interleave strategy for channels, layers,
layers and and views with a rendered .exr file. A single- or
views multi-part .exr file is created as described by the
following options, with layers and parts sorted
alphabetically:
• channels, layers and views - creates a single-
part .exr file ensuring backward compatibility
with applications that use OpenEXR 1.x.
• channels and layers - creates a multi-part .exr
file with one part per view. This can improve
Read performance as Nuke only reads the part
associated with the specified view.
• channels - creates a multi-part .exr with one
part per layer.
first part first_part none When the channels control is set to all and
interleave is set to channels, this dropdown
specifies which channel is written first in multi-
REFERENCE GUIDE
75
Write |

Value
channel .exr files. All subsequent channels are

written in alphanumeric order.
standard standard layer name disabled When enabled, the rendered EXRs follow the
layer name format standard .exr format layer.view.channel
format
Note: Older versions of Nuke use

view.layer.channel for .exr files.
write full write_full_layer_names disabled When enabled, always write the layer name in the
layer names channel names, folling the EXR standard.
When disabled, follow the legacy Nuke method

of storing the layer name in the part name of
multi-part files.
truncate truncateChannelNames disabled When enabled, channel names are truncated to a

channel maximum of 31 characters for legacy Nuke
names compatiblity.
When disabled, channel names are not truncated.
JPEG

(Scripting) Value
quality _jpeg_quality 0.75 Sets the quality of the rendered JPGs.
sub-sampling _jpeg_sub_ 4:1:1 Sets the chroma sub-sampling to use, from lowest to
sampling highest quality:
• 4:1:1
• 4:2:2
• 4:4:4
REFERENCE GUIDE
76
Write |
MOV
Note: The mov32Reader and Writer has been removed for macOS Catalina (10.15) and
deprecated under Windows, Linux, and macOS Mojave (10.14). The mov32: prefix is still currently
available on deprecated OS, but unsupported. We recommend using the standard Nuke
Read/Write controls, as mov32 is scheduled for removal in a future release.

(Scripting) Value
Codec mov64_codec Apple ProRes Sets the codec to use during rendering. See the codec
subsections for specific controls.
FPS mov64_fps dependent on Sets the playback frames per second for the output file.
Compositing
environment
Project
Settings
Animation Codec
Pixel Format mov64_pixel_ RGB 8-bit Lists pixel formats supported by the current codec:
format • Pixel bit-depth - 8-bit, 16-bit, and so on. This sets the
encoding depth used when decompressing the frames.
A large bit depth gives higher accuracy at the cost of
speed and memory usage.
Audio File mov64_ N/A Allows you to specify a separate audio file to include in
audiofile the output.
Either enter the file path manually or click the browse

button to locate the audio file.
Audio Offset mov64_ 0 Sets the start time of any audio file specified in the
audio_offset Audio File control. The unit of measure is specified
using the Units control.
Negative values cause the audio to start before the

video and vice versa.
Units mov64_units Seconds Sets the unit of measure for the audio offset control:
REFERENCE GUIDE
77
Write |

(Scripting) Value
• Seconds
• Frames
Advanced
Data Range dataRange Video Range Sets the minimum and maximum color range values for
encoding DNx files:
• Full Range - encodes with the full data range (0-255
for 8-bit or 0-1023 for 10-bit).
• Video Range - encodes with a reduced, or video/legal,
data range (16-240 for 8-bit or 64-960 for 10-bit).
Fast Start mov64_fast_ enabled When enabled, MOVs are playable while still down
start loading during streaming.
Write mov64_write_ enabled When enabled, a time code track is added to the
Timecode timecode rendered .mov file.
You can also use the quicktime/reel metadata, if present,

to give the track its reel name. You can add this key
using the ModifyMetaData node if it doesn’t exist.
Note: The input/timecode key must be

present in the sequence metadata in order to
write a time code.
GOP Size mov64_gop_ 12 Sets how many frames can be placed together to form a
size compression GOP (group of pictures).
Note: Use caution with this control as large

alterations can stop other applications reading
the rendered file.
Apple ProRes Codec
Codec Profile mov_prores_ ProRes 4:2:2 Sets the codec profile to use when a ProRes codec is
codec_profile HQ 10-bit selected in the Codec control:
REFERENCE GUIDE
78
Write |

(Scripting) Value
• Pixel layout - 420, 422, 444, and so on. This defines

how the chroma channels in the buffer are arranged.
444 buffers have lower spatial chroma sampling than
422, so they are generally preferred when available.
• Pixel bit-depth - 8-bit, 16-bit, and so on. This sets the
YCbCr Matrix mov64_ycbcr_ Auto Sets the way RGB is converted to Y’CbCr. Rec 601 and
matrix_type Rec 709 follow the ITU.BC specifications.
Auto uses a set a heuristics to determine the best color

matrix to use, including Format-based, which sets the
color matrix to Rec 601 for formats with a width below
840 pixels and Rec 709 for formats with a width of 840
pixels or above.


• Seconds
• Frames
Advanced
encoding DNx files:
REFERENCE GUIDE
79
Write |

(Scripting) Value



write a time code.
Avid DNxHD Codec
Codec Profile mov64_ DNxHD 444 Sets the codec profile to use when a DNxHD codec is
dnxhd_codec_ 10-bit 440 selected in the Codec control:
profile Mbit • Pixel layout - 420, 422, 444, and so on. This defines

REFERENCE GUIDE
80
Write |

(Scripting) Value

• Seconds
• Frames
Advanced
encoding DNx files:


write a time code.
Avid DNxHR Codec
Codec Profile mov64_ 4:4:4 12-bit Sets the codec profile to use when a DNxHR codec is
dnxhr_codec_ selected in the Codec control:
REFERENCE GUIDE
81
Write |

(Scripting) Value
profile • Pixel layout - 420, 422, 444, and so on. This defines


• Seconds
• Frames
Advanced
encoding DNx files:
REFERENCE GUIDE
82
Write |

(Scripting) Value


write a time code.
H.264 Codec
Codec Profile mov64_h264_ High 4:2:0 8- Sets the codec profile to use when a DNxHR codec is
codec_profile bit selected in the Codec control:
Quality mov64_ High Selects a preset level that automatically adjusts the
quality Advanced controls. Higher values produce larger files
with better quality.
Select Custom to adjust the Advanced controls

manually.

REFERENCE GUIDE
83
Write |

(Scripting) Value
pixels or above.


• Seconds
• Frames
Advanced
encoding DNx files:

REFERENCE GUIDE
84
Write |

(Scripting) Value

write a time code.

the rendered file.
B Frames mov64_b_ 0 Sets the maximum number of B frames that can be

frames consecutive in the rendered file.
The default, 0, does not impose any maximum number

of B frames in the output.
Bitrate mov64_ 28000 Sets the target bitrate that the codec attempts to reach,
bitrate within the limits set by the Bitrate Tolerance and
Quantizer Min/Max controls.
Bitrate mov64_ 0 Sets the amount that the bitrate can vary from the
Tolerance bitrate_ Bitrate setting. Setting this tolerance too low can result
tolerance in renders failing.
Quantizer Min mov64_ 1 Sets the quality range within which the codec can vary
quailty_min the image to achieve the specified bitrate. Higher ranges
can introduce image degradation.
Quantizer mov64_ 3
Max quailty_max
Motion JPEG A Codec
Pixel Format mov64_pixel_ YCbCr 4:4:4 8- Lists pixel formats supported by the current codec:
format bit • Pixel layout - 420, 422, 444, and so on. This defines
REFERENCE GUIDE
85
Write |

(Scripting) Value


manually.

pixels or above.


• Seconds
• Frames
Advanced
REFERENCE GUIDE
86
Write |

(Scripting) Value
encoding DNx files:


write a time code.
Quantizer mov64_ 3
Max quailty_max
Motion JPEG B Codec
format bit
REFERENCE GUIDE
87
Write |

(Scripting) Value


manually.

pixels or above.


REFERENCE GUIDE
88
Write |

(Scripting) Value
• Seconds
• Frames
Advanced
encoding DNx files:


write a time code.
Quantizer mov64_ 3
REFERENCE GUIDE
89
Write |

(Scripting) Value
Max quailty_max
MPEG-4 Codec

manually.

pixels or above.

REFERENCE GUIDE
90
Write |

(Scripting) Value

• Seconds
• Frames
Advanced
encoding DNx files:


write a time code.

the rendered file.
REFERENCE GUIDE
91
Write |

(Scripting) Value
B Frames mov64_b_ 0 Sets the maximum number of B frames that can be

frames consecutive in the rendered file.
The default, 0, does not impose any maximum number

of B frames in the output.
Limit Bitrate mov64_limit_ disabled When enabled, the maximum Bitrate is capped and the
bitrate Bitrate Tolerance is set to 0.
This control is designed to reduce output file size, but

the render quality is generally lower.
Quantizer mov64_ 3
Max quailty_max
PHOTO JPEG Codec
REFERENCE GUIDE
92
Write |

(Scripting) Value

manually.

pixels or above.


• Seconds
• Frames
Advanced
encoding DNx files:
REFERENCE GUIDE
93
Write |

(Scripting) Value


write a time code.
Quantizer mov64_ 3 Quantizer Max

Max quailty_max
PNG Codec
Pixel Format mov64_pixel_ RGB 16-bit Lists pixel formats supported by the current codec:
format • Pixel bit-depth - 8-bit, 16-bit, and so on. This sets the
REFERENCE GUIDE
94
Write |

(Scripting) Value


• Seconds
• Frames
Advanced
encoding DNx files:


write a time code.
Uncompressed Codec
REFERENCE GUIDE
95
Write |

(Scripting) Value
Pixel Format mov64_pixel_ YCbCr 4:2:2 Lists pixel formats supported by the current codec:
format 10-bit • Pixel layout - 420, 422, 444, and so on. This defines

pixels or above.


• Seconds
• Frames
Advanced
REFERENCE GUIDE
96
Write |

(Scripting) Value
encoding DNx files:



write a time code.
MXF

(Scripting) Value
Codec mxf_video_ Avid DNxHR Sets the codec to use during rendering.
codec_knob
Pattern mxf_op_ OP-1a Sets the operational pattern of the rendered file:
pattern_knob • OP-1a - combines audio and video into a single file,
similar to Final Cut Pro X.
• OP-Atom - creates separate audio and video files,
similar to Avid Media Composer.
FPS mxf_edit_ dependent on Sets the playback frames per second for the output file.
rate_knob
REFERENCE GUIDE
97
Write |

(Scripting) Value
Compositing
environment
Project
Settings
Codec Profile mxf_codec_ 4:4:4 12-bit Lists codec profiles supported by the current codec:
profile_knob • Pixel layout - 420, 422, 444, and so on. This defines
Tape ID mxf_tape_id_ N/A When Pattern is set to OP-Atom, specifies a tape name
knob that can be read by Avid Media Composer.
Advanced
Data Range dataRange Full Range Sets the minimum and maximum color range values for
encoding DNx files:
PNG

(Scripting) Value
data type datatype 8 bit Sets the bit depth of the rendered .png files:
• 8-bit
• 16-bit
REFERENCE GUIDE
98
Write |
SGI

(Scripting) Value
data type datatype 8 bit Sets the bit depth of the rendered .sgi files:
• 8-bit
• 16-bit
big endian bigEndian enabled When enabled, the rendered file is big-endian, rather
than native-endian. Big-endian files take longer to
render, but some applications only accept big-endian
files.
compression compression RLE Sets the compression type to apply to the rendered file.
TARGA

(Scripting) Value
compression compression RLE Sets the compression type to apply to the rendered file.
TIFF

(Scripting) Value
data type datatype 8 bit Sets the bit depth of the rendered .tiff files:
• 8-bit
• 16-bit
• 32-bit float
compression compression Deflate Sets the compression type to apply to the rendered file.
REFERENCE GUIDE
99
Write |
YUV

(Scripting) Value
interlaced interlaced disabled When enabled, the file is rendered as interlaced rather
than progressive.
REFERENCE GUIDE
100
Draw Nodes
Draw nodes contain roto shapes, paint tools, film grain, fills, lens flares, sparkles, and other vector-based
image tools.
Dither
Dither adds random noise to an image to hide color quantization effects. Color quantization is a
reduction in the number of colors used in the image, and can occur, for example, if the image has been
stored into an 8-bit file without error diffusion. In a dithered image, colors that are not available are
approximated by a diffusion of colored pixels from within the available palette. The human eye perceives
the diffusion as a mixture of the colors within it. This produces an impression of a larger number of colors
and eliminates banding artifacts.
See also Noise, Grain, ScannedGrain, and F_ReGrain.
Inputs and Controls

Type Name
Input unnamed The image sequence with unwanted color quantization effects.
mask An optional image to use as a mask. By default, the dither effect is

limited to the non-black areas of the mask.
At first, the mask input appears as triangle on the right side of the
node, but when you drag it, it turns into an arrow labeled mask. If you
cannot see the mask input, ensure that the mask control is disabled
or set to none.
REFERENCE GUIDE
101
|

Value
Dither Tab
channels channels all The effect is only applied to these channels.

amount amount 0.1 Up to 0.5 of this amount is added or subtracted from

each pixel to create the effect. Higher values produce
more dither and vice versa.
monodither monodither disabled When enabled, the same dithering pattern is used for
all channels.
seed seed 1 The same pattern is usually produced for each Dither
node. Changing the seed number produces a
different pattern, for instance if you’re using multiple
Dither nodes.
static seed static_seed disabled When enabled, the dither pattern remains the same
for every frame producing a constant dither effect.
mask N/A disabled Enables the associated mask channel to the right.
Disabling this checkbox is the same as setting the
channel to none.
maskChannelInput none The channel to use as a mask. By default, the dither

effect is limited to the non-black areas of this channel.
inject inject disabled Copies the mask input to the predefined mask.a
channel. Injecting the mask allows you to use the
same mask further downstream.
invert invert_mask disabled Inverts the use of the mask channel so that the dither
effect is limited to the non-white areas of the mask.
fringe fringe disabled When enabled, only apply the effect to the edge of
the mask.
When disabled, the effect is applied to the entire

mask.
REFERENCE GUIDE
102
DustBust |

Value
(un)premult N/A disabled Enables the associated channel to the right. Disabling
by this checkbox is the same as setting the channel to
none.
unpremult none The image is divided by this channel before being

processed, and multiplied again afterwards.
If you are using premultiplied input images, you may

want to check (un)premult by and select rgba.alpha
here. This will simulate applying the dither before the
premultiplication was done. It is the same as adding
an Unpremult node before this node and a Premult
node after, but allows you to work faster if you’re only
using one Dither node.
If you are using unpremultiplied input images, you

should leave this set to none.
invert invert_unpremult disabled Inverts the use of the (un)premultiply channel.
mix mix 1 Dissolves between the original image at 0 and the full
dither effect at 1.
DustBust
DustBust allows you to remove dirt and scratches from scanned images by painting over them with pixels
from another frame or another part of the same frame. It clones multiple areas from a source to a
destination based on dust points you create by Ctrl/Cmd+clicking in the Viewer. The source position can
be different for each dust point.
See also RotoPaint.
REFERENCE GUIDE
103
DustBust |
Inputs and Controls

Type Name
Input unnamed The image with the dirt or scratches to remove.

(Scripting) Value
DustBust Tab
enable enable 1 Sets the effect intensity from 0-1, where 1 is equal to the
full effect.
view source view_source disabled When enabled, you can view the source frame for the
points currently in the Viewer.
frame offset frame_offset 1 The clone source can be on the same frame (value 0) or
on another frame determined by the frame offset.
Note: Frame offset is the same to all dust

points on a particular frame (but frame offset
can be animated over time if desired).
edge edge_ 0.5 Sets the edge softness for all dust points on a particular
hardness hardness frame.
Output Mask N/A disabled Enables the associated mask channel to the right.
channel to none.
maskchannel none The channel to use as a mask. By default, the effect is

limited to the non-black areas of this channel.
REFERENCE GUIDE
104
Flare |
Flare
This node simulates lens flares that are created by reflections between lenses inside a film or video camera
when the camera is pointed towards a bright light source, such as the sun. Lens flares are typically added
to compositions to create a sense of realism, implying that the image is an un-edited shot of a real life
scene.
Lens flares are generally round, oval, or roundish with five or six flattened sides. They tend to show chroma
shifting, generating rainbow-like outlines. Nuke flares are implemented as a donut-like shape. They have
three radii:
• The outer radius is the outside of the donut.
• The middle radius is the thickest part.
• The inner radius is where the donut hole would be.
You can define two colors for the flare: the color of the donut itself and the fill color for the donut hole. If
these two colors are the same, flares are rendered as disks.
Inputs and Controls

Type Name
Input unnamed The image sequence to which you want to add lens flares.

(Scripting) Value
Flare Tab
Position
position position 100, 300 Sets the center position of the flare on the x and y axes.
You can reposition the flare along an imaginary line

between position and the center of the image using the
offset control. In this case, set position to the
coordinates of the light source.
REFERENCE GUIDE
105
Flare |

(Scripting) Value
offset offset 0 Adjusts the flare along a line between the center of the
image at 0 and the position, or light source, at 1.
radius radius 80, 100, 120 Sets the size of the flare as described by three radii:
• inner - sets the size of the ’hole’ filled by the inner
color control.
• middle - sets the thickness of the donut filled by the
ring color control.
• outer - sets the outer limit of the donut.
size mult size_mult 1 Scales the flare globally, while retaining the ratio
described by the radius controls.
anamorph anamorph 1 Simulates anamorphic lens flares by altering the default

1:1 flare shape.
Colors
ring color ring_color 1 Sets the ring color of the flare donut. Flares fade from
black at the outer radius to the ring color on the middle
radius.
inner color inner_color 0.5 Sets the inner color of the flare donut. Flares fade from
the ring color on the middle radius to the inner color on
the inner radius.
outer falloff outer _falloff 1 Sets the blending or falloff between the outer radius
and the background image.
inner falloff inner_falloff 1 Sets the blending or falloff between the inner radius and
the background image.
Color Shifts
chroma chroma_ 0 Moves chroma spread to and from the center of the
spread spread flare. Positive values shift red outward and blue inward,
and vice versa.
Note: You can use the color inner and outerfalloff

controls to compensate for banding created by large
chroma spread values.
REFERENCE GUIDE
106
Flare |

(Scripting) Value
chroma shift chroma_shift 0 Moves color components in the direction specified in

the angle control.
angle chroma_shift_ 0 Sets the direction of chroma shift. Positive values rotate
angle the shift counter-clockwise and vice versa.
rel chroma_shift_ disabled When enabled, angle is relative to the light angle (the
angle_add angle between position and the center point of the
image).
Visibility
brightness brightness 1 Adjusts the overall brightness of the flare.
Setting this value to 0 makes the flare invisible.
in vis_in 1 Sets the relative distance from the light source to the
matte, at which the flare is fully visible.
out vis_out 1.1 Sets the relative distance from the light source to the
matte, at which the flare is fully invisible.
channel to none.
vis_mask none The channel to use as a visibility mask. The lens flare
Where the visibility mask is white, the full brightness

value is used. Where the mask is black, brightness is set
to 0 (invisible). Gray areas are attenuated accordingly.
mask blur vis_mask_size 4 When a visibility mask is used, the mask is blurred by this
radius.
Shape
corners corners 5 If edge flattening is set to a value greater than 0, this

controls the number of sides that make up the
polygonal flare.
REFERENCE GUIDE
107
Flare |

(Scripting) Value
edge flattening 0 Simulates polygonal flares that some physical lens flares
flattening create. The default value of 0 creates a circular flare.
Higher values push the sides toward the center of the
flare, creating a star shape at values past 1.
corner sharpness 0.5 Adjusts the sharpness of the angles created with
sharpness edgeflattening.
angle corners_angle 0 Sets the position of the first corner of a polygonal flare.
Positive values rotate the first corner counter-clockwise
and vice versa.
rel corners_ disabled When enabled, angle is relative to the light angle (the
angle_add angle between position and the center point of the
image).
Multi Tab
Asymmetry
repeat asym_repeat 1 Sets the number of darkened sections around the

circumference of the flare.
multiplier asym_ 1 Sets how dark the darkened sections are. The default
multiplier value of 1 creates no darkening, whereas a value of 0
sets the darkened sections to black.
falloff asym_falloff 1 Sets the falloff at the edges of the darkened sections.
The higher the value, the wider the darkened sections.
angle asym_angle 0 Rotates the darkened sections around the center of the
flare. Positive values rotate the sections counter-
clockwise and vice versa.
Multiple Flares
repeat mult_num 1 Creates multiple flares with similar attributes. These are
randomized according to the controls below. You can
affect multiple flares by adjusting the overall seed or by
setting seeds for individual controls.
overall seed mult_seed 0 Sets the master, or overall, seed for all the below
REFERENCE GUIDE
108
Flare |

(Scripting) Value
attributes on all repeat flares. This produces random

attributes for each flare. Change the value to randomly
vary the attributes.
random offset mult_offset 1 Randomly varies the offsets for all repeat flares along
the line from the image center to position.
mult_offset_ 48235 Randomly varies the offsets for all repeat flares. You can
seed tie Multiple Flare controls together by giving them the
same value here.
size mult_size 0 Randomly varies the sizes of all repeat flares.
Positive values scale the flares down and vice versa.
mult_size_ 94335 Randomly varies the sizes of all repeat flares. You can tie
seed Multiple Flare controls together by giving them the
same value here.
hue mult_color 0 Randomly varies the colors of all repeat flares.
Positive values move the hue towards blue and negative

values toward red.
mult_color_ 77088 Randomly varies the colors of all repeat flares. You can
seed tie Multiple Flare controls together by giving them the
same value here.
brightness mult_ 0 Randomly varies the brightness for all repeat flares.
brightness
Positive values increase the flare brightness and vice
versa.
mult_ 75021 Randomly varies the brightness for all repeat flares. You
brightness_ can tie Multiple Flare controls together by giving them
seed the same value here.
visibility mult_visibility 0 Randomly varies the visibility for all repeat flares.
Positive values increase the flare opacity and vice versa.
mult_ 68995 Randomly varies the visibility for all repeat flares. You
REFERENCE GUIDE
109
Flare |

(Scripting) Value
visibility_seed can tie Multiple Flare controls together by giving them

the same value here.
edge mult_ 0 Randomly varies the edge flattening for all repeat flares.
flattening flattening
Positive values push the sides toward the center of the
flare. Negative values push the sides out, creating a
more circular flare.
mult_ 45203 Randomly varies the edge flattening for all repeat flares.
flattening_ You can tie Multiple Flare controls together by giving
seed them the same value here.
rotation mult_rotate 0 Randomly varies the rotation for all repeat flares.
Positive values rotate the flare clockwise and vice versa.
mult_rotate_ 49429 Randomly varies the rotation for all repeat flares. You
seed can tie Multiple Flare controls together by giving them
the same value here.
Extra
speedup speedup 0 Decreases rendering time at the expense of horizontal

resolution. The default value of 0 renders at full
resolution.
Presets Tab
Flares
MultiPenta MultiPenta N/A Creates multiple preset polygonal flares along the line
from the image center to position.
Rings
LgRainbow LgRainbow N/A Creates multiple rainbow colored ring flares along the
line from the image center to position.
Glowballs
Bright Bright N/A Creates multiple preset bright flares along the line from
the image center to position.
REFERENCE GUIDE
110
Glint |

(Scripting) Value
Sparkles
ManyBright ManyBright N/A Creates multiple preset sparkle flares along the line from
the image center to position.
Symmetry4 Symmetry4 N/A
Resets
all all N/A Removes all preset flares and resets all Flare node
controls to their defaults.
Glint
With the Glint node, you can create star-shaped rays around all the bright points in an image.
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the star-shaped rays.
mask An optional image to use as a mask. By default, the star-shaped rays

are limited to the non-black areas of the mask.
At first, the mask input appears as a triangle on the right side of the
or set to none.
REFERENCE GUIDE
111
Glint |

Value
Glint Tab
channels channels all The star-shaped rays are only applied to these
channels.

no. of rays repeat 2 The number of rays coming out of the bright points in
your image. For example, if you want to create five-
pointed stars, enter 5 here.
length length 50 Sets the length of the rays.
aspect ratio aspect 1 Sets the aspect ratio of the star-shapes.
odd ray oddlen 1 Gives every other ray a different length and sets that
length length.
rotation rotation 0 Rotates the star-shapes by degrees. Negative values

produce anti-clockwise rotation.
from color from_color 1 Sets the color in the beginning of the rays near the
center point of the stars.
to color to_color 0 Sets the color in the end of the rays.
steps steps 5 Sets how many steps the rays are formed of. The
more steps you use and the shorter the rays are, the
smoother the rays become.
max max enabled When enabled, the brightest image forming the rays
is used.
When disabled, the images forming the rays are

added up.
effect only effect_only disabled When enabled, only outputs the effect without
merging it with the original input image.
w N/A disabled Enables the associated w channel to the right.
REFERENCE GUIDE
112
Glint |

Value

channel to none.
w_channel none The channel to use as a mask for what are considered
glint sources. Highlights in the non-black areas of this
channel cause the glint effect, whereas highlights in
the non-white areas do not.
This is different from the mask channel in that the

glint effect can extend outside the matte in this
channel as long as the highlights that cause the effect
are inside the matte.
tolerance tolerance 0.5 Sets the threshold for how bright the highlights in the
input image need to be to cause the glint effect. Only
pixels above the threshold will bloom with the effect.
Low values produce more glint than high values.
gamma gamma 1 Lets you perform a gamma correction on the

highlights that cause glint before the glint effect is
applied.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the effect is

invert invert_mask disabled Inverts the use of the mask channel, so that the
transform is limited to the non-white areas of the
mask.
the mask.
REFERENCE GUIDE
113
Grain |

Value

mask.
effect at 1.
Grain
The Grain node lets you add synthetic grain (rather than grain derived from actual film stock) to an image.
This helps you ensure that all of the elements in your composite, including those which were digitally
generated, look like they were shot on the same film stock.
The presets dropdown menu includes predefined types of grain, such as Kodak 5248 and Kodak 5218.
These are the correct size for 2K scans.
You can also adjust the Grain node’s controls to match a sample piece of grain:
1. Find a sample with a rather constant background.
2. Blur the sample to remove the grain.
3. Connect the blurred sample to the Grain node’s input.
4. In the Viewer, wipe between the Grain node and the original sample image, and adjust the Grain
node’s controls to match the grain. It helps to view and match each of the red, green, and blue
channels separately.
See also ScannedGrain and F_ReGrain.
Inputs and Controls

Type Name
Input 1 The image sequence to receive the effect.
REFERENCE GUIDE
114
Grain |

(Scripting) Value
User Tab
seed seed 134 A different grain pattern is produced for each frame.
Change this number to get a different initial pattern, for
instance if you are using multiple Grain nodes.
If you require an identical grain plate for every frame,

enter -frame in this field.
presets presets N/A Choose the film stock you want to match:
• Kodak 5248
• Kodak 5279
• Kodak FX214
• Kodak GT5274
• Kodak 5217
• Kodak 5218
• other - changes the label on the node to other. This
has no effect on the effect applied.
Size
red red_size 3.3 Shrinks or enlarges the effect in the red channel.
green green_size 2.9 Shrinks or enlarges the effect in the green channel.
blue blue_size 2.5 Shrinks or enlarges the effect in the blue channel.
Irregularity
red red_i 0.6 Increases or decreases the random quality of the effect
in the red channel.
green green_i 0.6 Increases or decreases the random quality of the effect
in the green channel.
blue blue_i 0.6 Increases or decreases the random quality of the effect
in the blue channel.
Intensity
red red_m 0.416 Increases or decreases the contrast of the effect in the
REFERENCE GUIDE
115
Grid |

(Scripting) Value
red channel against the original image.
green green_m 0.46 Increases or decreases the contrast of the effect in the
green channel against the original image.
blue blue_m 0.85 Increases or decreases the contrast of the effect in the
blue channel against the original image.
black black 0 The base amount of grain to add at the black point.
minimum rgb minimum 0, 0, 0 The minimum output rgb values. If the calculation results
in pixel values lower than these, these values are used
instead.
Mask
Apply only maskgrain enabled When enabled, effect is limited to the non-black areas
through alpha of the image’s alpha channel. This can be useful when
you want to apply grain to premultiplied CG images
without contaminating the background black.
When disabled, effect is applied to the entire image.
Invert alpha invert_mask disabled Inverts the use of the alpha channel, so that when Apply
only through alpha is enabled, the effect is limited to
the non-white areas of the alpha.
Grid
This node creates a grid of horizontal and vertical lines on top of the input image.
REFERENCE GUIDE
116
Grid |
Inputs and Controls

Type Name
Input unnamed The image on which the grid is superimposed.
mask An optional image to use as a mask. By default, the grid is limited to

the non-black areas of the mask.
or set to none.

Value
Grid Tab
output output rgba The grid is only applied to these channels.
You can use the checkboxes on the right to select the

individual channels to output.
premult premult none These channels are set to black outside the grid
shape.
clip to cliptype format Sets how to restrict the output image:

• no clip - use the entire image.
• bbox - restrict the output image to the incoming
bounding box.
• format - restrict the output image to the incoming
format area.
• unionbboxandformat - restrict the output image to
a combination of the incoming bounding box and
format area.
• intersectbboxandformat - restrict the output
image to an intersection of the bounding box and
REFERENCE GUIDE
117
Grid |

Value
incoming format area.
replace replace disabled When enabled, existing channels are cleared to black
before drawing into them. You might find replace
useful, for instance, if you’re creating a mask in the
alpha channel, but the incoming image already has an
alpha channel which you want to throw away.
invert invert disabled When enabled, inverts the grid (that is, fills in the
holes in the grid and leaves holes for the lines).
opacity opacity 1 Adjusts the grid opacity on a sliding scale, where 0 is

completely transparent and 1 is completely opaque.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the grid is

invert invert_mask disabled Inverts the use of the mask channel, so that grid is
limited to the non-white areas of the mask.
number number 10 Sets the number of grid squares on the x and y axes.
size size 1 Sets the grid line thickness in pixels.
translate translate 0, 0 Translates the entire grid on the x and y axes.
Color Tab
REFERENCE GUIDE
118
Grid |

Value
ramp ramp none Selects whether to create a color gradient across the
grid:
• none - don’t create a color gradient.
• linear - the ramp changes linearly from one color
into another.
• smooth0 - the ramp color gradually eases into the
point 0 end. This means colors in the point 0 end are
spread wider than colors in the point 1 end.
• smooth1 - the ramp color eases into the point 1
end. This means colors in the point 1 end are spread
wider than at the point 0 end.
• smooth - the ramp color gradually eases into both
ends. This means colors in the point 0 and point 1
ends are spread wider than colors in the center of
the ramp.
color color 1 Sets a color for the grid.
If ramp is set to anything other than none, this is the

color for the ramp at the point 1 end (by default, the
top end).
point 1 p1 100, 400 Sets the position of point 1. This allows you to adjust
the spread and angle of the ramp.
This control is only available if ramp is set to anything

other than none.
color 0 color0 0 Sets the color for the ramp at the point 0 end (by
default, the bottom end).

other than none.

other than none.
REFERENCE GUIDE
119
LightWrap |
LightWrap
LightWrap can help composite objects into a very bright background by ’wrapping’ or spilling light from
the background onto the foreground object. The node creates a reflection of light around the edges of
your foreground object by blending in whatever is in the background.
You should apply LightWrap on your foreground element before you composite the foreground over the
background with the Merge node.
Inputs and Controls

Type Name
Input A The foreground image.
B A background that will bleed light into the foreground image.

(Scripting) Value
LightWrap Tab
Diffuse size 15 Adjusts the spread of the reflections on the

foreground element. This and Intensity need to be
balanced out together. You may want to start by
bringing Diffuse all the way down to better see what
you are blending in from the background. Then, adjust
Intensity before going back to the Diffuse slider and,
if necessary, Intensity again until you are happy with
the result.
Intensity Intensity 0 Adjusts the brightness of the reflections on the

foreground element. This and Diffuse need to be
balanced out together. You may want to start by
bringing Diffuse all the way down to better see what
REFERENCE GUIDE
120
LightWrap |

(Scripting) Value
you are blending in from the background. Then, adjust

Intensity before going back to the Diffuse slider and,
if necessary, Intensity again until you are happy with
the result.
Generate wrap onlywrap disabled When enabled, generates the LightWrap effect only,
only without merging it with the foreground object.
Disable disableluma disabled When enabled, creates a uniform effect around the
luminance edges of the foreground, rather than have the effect
based wrap adjust itself according to the background.
Enable Glow enableglow disabled When enabled, extends the effect outside the
foreground object’s alpha, making the object seem to
glow.
FGBlur FGBlur 1 Sets how much the foreground matte is blurred. The
more blur, the more of the background is added to
the foreground.
BGBlur BGBlur 0 Sets how much the background is blurred before it is

merged with the foreground element.
Saturation saturation 1 Adjusts the saturation of the LightWrap effect.
LumaTolerance LumaTolerance 0 Adjusts the luminance values of the LightWrap effect.
Highlight highlightmerge plus Sets how the foreground object is merged with the
Merge background. The default merge operation, plus, adds
the elements together, producing a glow effect.
Use constant useconstant disabled When enabled, a constant color of your choice rather
highlight than the background is used in the LightWrap effect.
Select the color using the Constant control.
Constant color 1 Sets the color to use in the LightWrap effect when
Use constant highlight is enabled.
CCorrect Tab
white white 1 This color remains unchanged when you adjust

gamma.
REFERENCE GUIDE
121
LightWrap |

(Scripting) Value
midtone bias gray 0.18 Sets the center of midtones, remaining unchanged
when you adjust contrast.
master
contrast contrast 1 Adjusts the contrast in the entire LightWrap effect.
gamma gamma 1 Adjusts the gamma in the entire LightWrap effect.
gain gain 1 Adjusts the gain in the entire LightWrap effect.
offset offset 0 Adjusts the offset in the entire LightWrap effect.
shadows
gamma shadows_ 1 Adjusts the gamma in the darkest parts of the

gamma LightWrap effect.
gain shadows_gain 1 Adjusts the gain in the darkest parts of the LightWrap
effect.
offset shadows_offset 0 Adjusts the offset in the darkest parts of the

LightWrap effect.
midtones
gamma midtones_ 1 Adjusts the gamma in the middle range of the

gamma LightWrap effect.
gain midtones_gain 1 Adjusts the gain in the middle range of the LightWrap
effect.
offset midtones_ 0 Adjusts the offset in the middle range of the

offset LightWrap effect.
highlights
gamma highlights_ 1 Adjusts the gamma in the highlights of the LightWrap

gamma effect.
gain highlights_gain 1 Adjusts the gain in the highlights of the LightWrap

effect.
offset highlights_ 0 Adjusts the offset in the highlights of the LightWrap
REFERENCE GUIDE
122
MarkerRemoval |

(Scripting) Value
offset effect.
mix mix 1 Dissolves between the original LightWrap effect at 0

and the color corrected effect at 1.
MarkerRemoval
MarkerRemoval uses roto shapes to mask an area of the footage that you don’t want to appear in the final
result, typically small areas like tracking markers. Nuke automatically paints the masked area with suitable
pixels sampled from the background plate.
Note: Painted pixels with values of 0.5 mean the algorithm was unable to locate any pixel to fill
in.
Inputs and Controls

Type Name
Input unnamed The node supplying the mask information, for example Roto or
RotoPaint.

(Scripting) Value
MarkerRemoval Tab
paint type mode bilinear Sets the paint mode used to fill the roto shape(s):
• bilinear - extrapolates pixels on the xy axes to fill the
roto shape(s).
• radial - extrapolates pixels radially outward from a
REFERENCE GUIDE
123
MarkerRemoval |

(Scripting) Value
center point to fill the roto shape(s).
Note: The center point is averaged between

the number of roto shapes, so radial painting
with more than one roto shape is not
recommended. When using the radial mode,
you should only use one roto shape for each
MarkerRemoval node.
bilinear y linearsearch 100 Sets the height (in pixels) used to search for suitable
search pixels to fill the roto shape.
This value should always be higher than the height of the

roto mask surrounding the area you want to fill, and
should only be increased for large mask areas.
Note: This option is only available when the

painttype is set to bilinear.
control enabled Enables the associated channel to the right. Disabling

channel this checkbox is the same as setting the channel to none,
effectively disabling the node.
ctrlMask rgba.alpha The channel that contains the roto mask(s).
Pixel offset
x offset x_offset 0 If the roto shape mask you intend to use is close to the
left or right edges of the plate, you can offset the
sample area by a specified number of pixels to create a
more accurate fill for the roto shape.
y offset y_offset 0 If the roto shape mask you intend to use is close to the
top or bottom edges of the plate, you can offset the
sample area by a specified number of pixels to create a
more accurate fill for the roto shape.
REFERENCE GUIDE
124
Noise |
Noise
Adds various types of seemingly random noise to the input image, all based on the Perlin noise function.
See also Grain, ScannedGrain, and F_ReGrain.
Inputs and Controls

Type Name
Input unnamed The image sequence to which noise is added.
mask An optional image to use as a mask. By default, the noise is limited to

or set to none.

Value
Noise Tab
output output rgba The noise is rendered into these output channels.
If you set this to something other than none, you can

use the checkboxes on the right to select individual
channels.
premult premult none Premultiply multiplies the chosen input channels with
a mask representing the noise. For example, where
there is no noise (the matte is black or empty), the
input channels are set to black. Where the noise is
opaque (the matte is white or full), the input channels
keep their full value.
REFERENCE GUIDE
125
Noise |

Value
Note: Selecting rgba premultiplies the

alpha against itself (a*a). If you don’t want
this to happen, set premult to rgb instead.
clip to cliptype format Select how to restrict the output image:

bounding box.
format area.
format area.
invert invert disabled When enabled, inverts the new noise and non-noise
areas.
opacity opacity 1 Sets the noise opacity where 0 is equal to transparent,

and 1 is equal to opaque.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the noise is

REFERENCE GUIDE
126
Noise |

Value

invert invert_mask disabled Inverts the use of the mask channel so that the noise is
type type fBm Choose the type of noise to use:

• fBm (fractional Brownian motion) - runs noise()
octave times at different scales to get random
numbers between -1 and +1 and adds them up.
• turbulence - similar to fBm except it uses the
absolute value of the noise function so the values it
sums up are all between 0 and 1.
x/ysize size 350 Sets the lowest possible noise frequency on the x and
y axes. For example, low y values relative to x
produces horizontal noise.
z zoffset 0 Offsets the noise to create new patterns. This control

must be animated to produce noise changes over
time. For example, the expression frame/10 causes the
noise to change completely over 10 frames.
octaves octaves 10 Sets the number of Perlin noise functions to add to

the noise frequency.
clip at nyquist enabled When enabled, limits the number of octaves so that
Nyquist limit the highest frequency is larger than one pixel.
Note: You may need to disable this control

if you’re animating the size of the noise, as
changes can be visible.
lacunarity lacunarity 2 Sets the amount by which octaves multiply noise

frequency. Lower values produce smoother noise and
higher values produce more granular noise.
gain gain 0.5 Sets the amount by which octaves multiply noise
REFERENCE GUIDE
127
Noise |

Value
amplitude.
gamma gamma 0.5 Sets the gamma output for the noise.
Transform Tab
translate xy translate 0, 0 Translates the noise along the x and y axes. You can
also adjust translate values by dragging the transform
handle in the Viewer.
rotate rotate 0, 0 Rotates the noise around the center x y coordinates.

You can adjust rotate values by dragging the
transform handle in the Viewer.
scale scaling 1 Scales the noise width and height around the center x
y coordinates.
skew skew 0 Skews the noise on the x and y axes.
center xy center N/A Sets the center of rotation and scale on the x and y
axes.
xrotate xrotate 30 Rotation around the x axis in 3D noise space. Setting

this to zero reveals artifacts in the Perlin noise
generator.
yrotate yrotate 30 Rotation around the y axis in 3D noise space. Setting

this to zero reveals artifacts in the Perlin noise
generator.
Color Tab
noise:
into another.
REFERENCE GUIDE
128
Radial |

Value

wider than colors in the point 0 end.
the ramp.
color color 1 Sets a color for the noise.

top end).

other than none.

other than none.

other than none.
Radial
Generates a radial gradation ramp.
See also Ramp.
REFERENCE GUIDE
129
Radial |
Inputs and Controls

Type Name
Input unnamed The image sequence to which the effect is added.
mask An optional image to use as a mask. By default, the effect is limited to

or set to none.

Value
Radial Tab
output output rgba The effect is rendered into these output channels.

channels.
a mask representing the radial ramp. For example,
where there is no radial ramp (the matte is black or
empty), the input channels are set to black. Where the
radial ramp is opaque (the matte is white or full), the
input channels keep their full value.


REFERENCE GUIDE
130
Radial |

Value

bounding box.
format area.
format area.
invert invert disabled When enabled, inverts the radial ramp and non-radial
ramp areas.
opacity opacity 1 Sets the radial ramp opacity. 0 is equal to transparent

channel to none.
maskChannelInput none The channel to use as a mask. By default, the ramp is

invert invert_mask disabled Inverts the use of the mask channel so that the radial
ramp is limited to the non-white areas of the mask.
area xyrt area N/A Sets the area in which the radial ramp is contained.
(xywh) The center of the area control is the originating point
REFERENCE GUIDE
131
Radial |

Value
for the radial ramp.
softness softness 1 Sets the softness of the radial ramp edges, where 0 is
a hard edge.
perceptually plinear disabled When enabled, the radial ramp appears more linear to
linear the naked eye.
Color Tab
radial ramp:
into another.
the ramp.
color color 1 Sets a color for the ramp.

top end).
point 1 xy p1 100, 400 Sets the position of point 1. This allows you to adjust

other than none.
REFERENCE GUIDE
132
Ramp |

Value

other than none.

other than none.
Ramp
Generates a gradation between two defined edges. This node is included for backward compatibility -
other Draw nodes, such as Rectangle, have their own Ramp controls.
See also Radial.
Inputs and Controls

Type Name
Input unnamed The image sequence to which the ramp is added.
mask An optional image to use as a mask. By default, the ramp is limited to

or set to none.
REFERENCE GUIDE
133
Ramp |

Value
Ramp Tab
output output rgba The ramp is rendered into these output channels.

channels.
a mask representing the ramp. For example, where
there is no ramp (the matte is black or empty), the
input channels are set to black. Where the ramp is
opaque (the matte is white or full), the input channels
keep their full value.


bounding box.
format area.
• union bbox+format - restrict the output image to a
combination of the incoming bounding box and
format area.
• intersect bbox+format - restrict the output image
to an intersection of the bounding box and
REFERENCE GUIDE
134
Ramp |

Value
invert invert disabled When enabled, inverts the new ramp and non-ramp
areas.
opacity opacity 1 Sets the ramp opacity where 0 is equal to transparent,

channel to none.
maskChannelInput none The channel to use as a mask. By default, the ramp is

invert invert_mask disabled Inverts the use of the mask channel so that the ramp is

other than none.

other than none.
type type linear Sets the type of ramp to apply:

• linear - the ramp has a linear gradient.
• plinear - the ramp appears to have more linear
gradient to the naked eye.
• smooth0 - the ramp gradually eases into the point 0
REFERENCE GUIDE
135
Rectangle |

Value
end.
• smooth1 - the ramp eases into the point 1 end.
• smooth - the ramp gradually eases into both ends.
Color Tab
Rectangle
Generates a rectangle in a solid color on top of the input image.
Inputs and Controls

Type Name
Input unnamed The image sequence to which the rectangle is added.
mask An optional image to use as a mask. By default, the rectangle is

or set to none.

Value
Rectangle Tab
REFERENCE GUIDE
136
Rectangle |

Value
output output rgba The rectangle is rendered into these output channels.

channels.
a mask representing the rectangle. For example,
where there is no rectangle (the matte is black or
rectangle is opaque (the matte is white or full), the
input channels keep their full value.


bounding box.
format area.
format area.
REFERENCE GUIDE
137
Rectangle |

Value
invert invert disabled When enabled, inverts the new rectangle and non-
rectangle areas.
opacity opacity 1 Sets the rectangle opacity where 0 is equal to

transparent, and 1 is equal to opaque.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the rectangle

is limited to the non-black areas of this channel.
invert invert_mask disabled Inverts the use of the mask channel so that the
rectangle is limited to the non-white areas of the
mask.
area xyrt area N/A Sets the area of the rectangle.

(xywh)
softness softness 1 Sets the softness of the rectangle edges where 0 is a

hard edge.
Color Tab
rectangle:
into another.
REFERENCE GUIDE
138
Roto |

Value

the ramp.

top end).

other than none.

other than none.

other than none.
Roto
REFERENCE GUIDE
139
Roto |
The Roto node is an optimal choice if you’re doing rotoscoping only - it allows you to create and edit
Bezier and B-spline shapes.
The toolbar on the left side of the Viewer includes point selection and manipulation, and shape creation
tools. Click and hold or right-click on a toolbar button to open a sub-menu to select any of its available
tool types. Options related to the current tool appear in a toolbar along the top of the Viewer. Click on a
toolbar item to cycle through the available options for that class of tools.
As a general workflow guide:

1. Select your tool from the left-hand Viewer tool bar.
2. Use the Roto tool settings above the Viewer to adjust the tool’s properties.
3. Draw your shape or shapes:
• Click in the Viewer to place points. You can drag while clicking to pull out Bezier handles or adjust B-
Spline tension.
• Ctrl/Cmd+drag to sketch the shape freely.
• To close the shape, click the first point or press Return. To leave the shape open, press Esc.
• Ctrl/Cmd+Alt+click to add points to an existing shape.
• To increase the smoothness of a point, select the point and press Z.
• To cusp a point, select the point and press Shift+Z.
• To delete a point, select the point and press Delete.
• Beziers: Shift+drag on a tangent handle to snap the opposite handle to the same length.
Ctrl/Cmd+drag on a tangent handle to move it independently of its opposite handle.
• B-Splines: Ctrl/Cmd+Alt+drag on a point to adjust its tension.
4. Select a shape using the Select tools or the shape list in the control panel.
5. Use the control panel to adjust or fine-tune your shape(s).
You can also create this node by pressing O on the Node Graph.
See also RotoPaint.
Inputs and Controls

Type Name
Input bg Background input - adding a background automatically creates

another bg input, allowing you to connect up to four images.
REFERENCE GUIDE
140
Roto |

Type Name
mask An optional image to use as a mask. By default, the roto shapes are
node, but when you drag it, it turns into an arrow labeled mask. If
you cannot see the mask input, ensure that the mask control is
disabled or set to none.

Value
Roto Tab
output output alpha The roto shapes are rendered into these output
channels. The output channels are the same for all
shapes created using this node - you cannot create a
subset of shapes and output them to a different
channel.

premultiply premultiply none Premultiply multiplies the chosen input channels with
a mask representing the roto shapes. For example,
where there are no roto shapes (the matte is black or
roto shapes are opaque (the matte is white or full),
the input channels keep their full value.

this to happen, set premultiply to rgb
instead.

REFERENCE GUIDE
141
Roto |

Value

bounding box.
format area.
• union bbox+format - restrict the output image to a
combination of the incoming bounding box and
format area.
to an intersection of the bounding box and
format dropdown
format format root.format This is used if Roto has no input connected. It is the
format which the node should output in the absence
of any available input format. If an input is connected,
this control has no effect.
If the format does not yet exist, you can select new to
create a new format from scratch. The default setting,
root.format, resizes the image to the format
indicated on the Project Settings dialog box.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the roto

shapes are limited to the non-black areas of this
channel.
REFERENCE GUIDE
142
Roto |

Value
invert invert_mask disabled Inverts the use of the mask channel, so that output is
the mask.

mask.
output mask disabled Enables the associated output mask channel to the
right. Disabling this checkbox is the same as setting
the channel to none.
outputMask none Output mask channel. This is a channel where Roto

will output a mask for what it rendered.
opacity opacity 1 The opacity of a new or existing roto shape in the

shape list. 1 is equal to totally opaque, and 0 is equal
to transparent.
feather feather 0 Softens the edges of a roto shape by fading it from

more transparent to less transparent.
With positive feather values, your feather effect is

outward. If your feather values are negative, the
feather effect is inward.
You can also add feathering to individual points by

Ctrl/Cmd+dragging a point in the Viewer. Press E to
increase the feathering of selected points, or Shift+E
to remove the feathering.
If you check feather link in the Roto tool settings

above the Viewer, you can move the shape point and
the feather point together. Otherwise, they move
independently. You can also Ctrl/Cmd+drag a shape
point to temporarily disable feather link and move
the points independently.
REFERENCE GUIDE
143
Roto |

Value
on feather_on enabled When enabled, feathering is applied to the selected

shape(s).
Disabling this is the same as setting feather to 0.
feather falloff feather_falloff 1 Sets the rate of opacity falloff at a feathered edge for
a shape, in pixels.
[feather type] feather_type linear Controls the center point of the feather falloff:
• linear - the falloff changes linearly from the shape
edge to the feather edge.
• smooth0 - the center point of the feather falloff lies
closer to the shape edge than the feather edge.
closer to the feather edge than the shape edge.
• smooth - the center point of the feather falloff lies
halfway between the shape edge and the feather
edge.
These options may be helpful, for example, in

matching the soft edge to motion blurred image
content.
spline key key 0 Quick access to spline keyframes:

• Numeric fields - show whether the spline is keyed
at this frame, which key you’re viewing, and the total
number of spline keys.
• - move to the previous or next spline key.
• - set or delete spline keys.
visible visible enabled Sets whether the selected shape or group is visible
and rendered or not. You can still edit an invisible
shape and view its position in the Viewer.
locked locked disabled Locks the selected shape or group to prevent it from
being edited.
view view N/A The view(s) in which you want to draw your shape(s).
REFERENCE GUIDE
144
Roto |

Value
This is only displayed if you have set up more than

one view in your Project Settings (typically, if you’re
working on a stereoscopic project).
stereo offset stereo_offset 0 Moves the selected shape or group on the x and y
x,y axes. This is an extra transform that is applied after all
other transforms. Typically, you would position the
shape or group correctly in the hero view, then split
this control, and drag the shape to its correct location
in any other views. Note that you can also press Shift
while dragging to constrain the movement to x or y
axis only.
Stereo offset can be useful, for example, if you have

a shape or group that is correctly positioned in one
view and you want to move it to its correct location in
another view, but can’t use the translate control on
the Transform tab because that’s being driven by
Tracker data.
This control is only displayed if you have set up more

than one view in your Project Settings (typically, if
you’re working on a stereoscopic project).
shape list curves N/A Shows the hierarchy of shapes and groups, allowing
you to adjust them and how they’re displayed in the
Viewer:
• Name - double-click to edit the shape or group
Name.
• - whether the shape or group is visible and
rendered or not.
• - lock or unlock the shape or group.
• - set the shape outline color to appear in the

Viewer.
• - set the color in which you want to render your
shape.
REFERENCE GUIDE
145
Roto |

Value
• - invert the shape or group.
• - set the shape or group blending mode.
• - whether motion blur is applied to the shape.
Note: You can only apply motion blur to

shapes, not groups of shapes.
• View - displays the view(s) the shape appears in.

• Lifetime - the range of frames during which the
shape is visible.
• Source - the shape source (as set on the Shape tab).
Use the shape list to:

• Select multiple items to adjust attributes on all the
selected list items.
• Drag and drop to re-order or move items between
groups.
• Right-click for a context menu containing Add new
layer and shape editing functions such as Copy and
Paste.
Transform Tab
translate x,y translate 0 Moves the selected shape or group on the x and y
axes.
You can also adjust translate values by dragging the

rotate rotate 0 Spins the selected shape or group around a pivot

point. Use center x,y to position the pivot point.
REFERENCE GUIDE
146
Roto |

Value
You can also adjust rotate values by dragging the

scale scale 1 Resizes the selected shape or group on the x and y

axes.
You can also adjust scale values by dragging the

skew X skewX 0 Skews the selected shape or group along the X axis
from the pivot point. Use center x,y to position the
pivot point.
You can also adjust skew values by dragging the

skew Y skewY Skews the selected shape or group along the Y axis
from the pivot point. Use center x,y to position the
pivot point.

skew order skew_order XY Sets the order in which skew X and skew Y are
applied:
• XY - Skew X is applied before skew Y.
• YX - Skew Y is applied before skew X.
center x,y center 1024, 778 Adjusts the center of rotation and skew on the x and y
axes. This control does not translate roto shapes.
You can also adjust center values by Ctrl/Cmd+Shift

and dragging the transform handle in the Viewer.
extra matrix transform_matrix N/A Adds an extra matrix which is calculated after the
transforms defined by the other controls. This can be
useful, for example, if you are using a 3rd party
REFERENCE GUIDE
147
Roto |

Value
application, such as a tracker, and your tracking data

cannot be fully represented using the available
transform controls.
shape list curves N/A See the Roto tab for the shape list controls.
Motion Blur Tab
shape global_ enabled When enabled, motion blur is applied to shapes

motionblur_on selected in the shape list or Viewer.
shape motion blur determines the exposure for each

moving shape and blends the resulting blurred
shapes. This may be more efficient than the global
motion blur since each shape will only be blended
once.
Note: Shape motion blur may result in

artifacts when shapes blur over the same
region.
global disabled When enabled, motion blur is applied to all shapes in

the current Roto node.
Note: Global motion blur overrides the

per-shape motion blur settings.
global motion blur correctly accounts for interaction

between motion blurred shapes.This may be more
expensive than the shape motion blur since it may
blend each shape for every sample.
REFERENCE GUIDE
148
Roto |

Value
Note: Global motion blur requires that

shutter and sampling parameters are the
same for all shapes and has been optimized
for consecutive shapes with the same
properties using the over blend mode.
Shape Blur
motionblur motionblur 1 Sets the number of motion blur samples for the
selected shape(s). Increase this for better quality but
slower rendering, decrease it for faster rendering.
on motionblur_on disabled When enabled, motion blur is applied to the selected

shape(s).
shutter motionblur_ 0.5 Sets the number of frames the shutter stays open
shutter when motion blurring. For example, a value of 0.5
corresponds to half a frame. Increasing the value
produces higher quality motion blur, but increases
render time.
shutter offset motionblur_ start Controls how the shutter behaves with respect to the
shutter_offset_ current frame value. There are four options:
type • centred - center the shutter around the current
frame. For example, if you set the shutter value to 1
and your current frame is 30, the shutter stays open
from frame 29,5 to 30,5.
• start - open the shutter at the current frame. For
example, if you set the shutter value to 1 and your
current frame is 30, the shutter stays open from
frame 30 to 31.
• end - close the shutter at the current frame. For
frame 29 to 30.
• custom - open the shutter at the time you specify.
REFERENCE GUIDE
149
Roto |

Value
In the field next to the dropdown menu, enter a

value (in frames) you want to add to the current
frame. To open the shutter before the current frame,
enter a negative value. For example, a value of - 0.5
would open the shutter half a frame before the
current frame.
[offset] motionblur_ 0 If the shutter offset control is set to custom, this

shutter_offset control is used to set the time that the shutter opens
by adding it to the current frame.
Values are in frames, so -0.5 would open the shutter

half a frame before the current frame.
Global Blur
motionblur global_ 1 Sets the number of motion blur samples for all
motionblur shapes in the current Roto node. Increase this for
better quality but slower rendering, decrease it for
faster rendering.
shutter global_ 0.5 Sets the number of frames the shutter stays open
motionblur_ when motion blurring. For example, a value of 0.5
shutter corresponds to half a frame. Increasing the value
render time.
shutter offset global_ start Controls how the shutter behaves with respect to the
motionblur_ current frame value. There are four options:
shutter_offset_ • centred - center the shutter around the current
type frame. For example, if you set the shutter value to 1
frame 30 to 31.
REFERENCE GUIDE
150
Roto |

Value

frame 29 to 30.
frame. To open the shutter before the current frame,
enter a negative value. For example, a value of - 0.5
would open the shutter half a frame before the
current frame.
[offset] global_ 0 If the shutter offset control is set to custom, this

motionblur_ control is used to set the time that the shutter opens
shutter_offset by adding it to the current frame.

Shape Tab
source shape_source color Sets the color or source input for the selected shape:
• color - the color specified in the color field.
• foreground - pulls pixels from the bg input,
including any shapes drawn on it. This input is
mainly used with the Clone tab controls. Using it
may cause slower renders.
• background - pulls pixels from the bg input, not
including any shapes drawn on it.
• backgrounds 1 to 3 - pulls pixels from bg1, bg2,
and bg3 inputs.
blending blending_mode over Sets how the colors in the current shape (A) are
mode merged with the underlying image and shapes (B):
• color-burn (darken B towards A) - darkens B to
reflect A by increasing the contrast. No part of the
image becomes lighter.
REFERENCE GUIDE
151
Roto |

Value
• color-dodge (brighten B towards A) - brightens B to

reflect A by decreasing the contrast. No part of the
image is darkened.
• difference (abs(A-B)) - subtracts either B from A or
vice versa, depending on which is brighter. Blending
with white inverts A, while blending with black
produces no change. Similar colors will return black
pixels.
• exclusion (A+B-2AB) - creates a result similar to the
difference mode but lower in contrast. Like with
difference, blending with white inverts A. Blending
with black produces no change.
• from (B-A) - subtracts A from B.
• hard-light (multiply if A<.5, screen if A>.5) -
lightens highlights and darkens shadows. If B is
lighter than 50% gray, the result lightens as if it were
screened. If B is darker than 50% gray, the result is
darkened as if it were multiplied.
• max (max(A,B)) - selects the lighter of the two colors
as the resulting color. Only areas darker than B are
replaced, while areas lighter than B do not change.
• min (min(A,B)) - selects the darker of the two colors
as the resulting color. Any parts that are lighter than
B are substituted. Any parts of the image that are
darker than B don’t change.
blending • minus (A-B) - subtracts B from A.

mode • multiply (AB, A if A<0 and B<0) - multiplies A by B.
(continued) The result is always darker. Blending with black gives
black and with white returns the color unchanged.
• over (A over B) - the colors of the two images will
not interact in any way, and Nuke will display the full
value of the colors in A.
• overlay (multiply if B<.5, screen if B>.5) -
depending on A, multiplies or screens the colors. B
REFERENCE GUIDE
152
Roto |

Value
brightens A while preserving highlights and

shadows.
• plus (A+B) - the sum of the two colors. Increases
brightness to lighten A and reflect B. Plus is similar
to the screen blending mode, but produces a more
extreme result.
• screen (A+B-AB if A and B between 0-1, else A if
A>B else B) - this is a soft plus making everything
brighter but ramping off the whites. Light colors
have more of an effect than dark colors. The result is
always a lighter color. Blending with black leaves the
pixel unchanged, blending with white always returns
white. The result is similar to projecting multiple
slides on top of each other.
• soft-light - depending on A, darkens or lightens the
colors. Less extreme than the hard light blending
mode.
color color 1 Sets the color for the roto shape. This control only has
an effect when source is set to color.
overall width openspline_width 10 Sets the overall spline width of selected splines and
all new open splines.
start type openspline_start_ rounded Sets the style of the first point for the selected spline
end_type (s) and all new open splines:
• rounded - the starting point of the spline has a
rounded end.
• square - the starting point of the spline has a square
end.
end type openspline_last_ rounded Sets the style of the last point for the selected spline
• rounded - the finishing point of the spline has a
rounded end.
• square - the finishing point of the spline has a
square end.
REFERENCE GUIDE
153
Roto |

Value
invert inverted disabled Swaps the colors of the currently selected shape and
the underlying image.
Clone Tab
translate x,y source_transform_ 0,0 Moves the source image along the x and y axes.
translate
transform handles in the Viewer.
If source is set to color on the Shape tab, this control

is disabled.
round to source_translate_ disabled Rounds the translate x,y amount to the nearest
pixel round whole integer pixel. This can help you avoid softening
when using filtering.

is disabled.
rotate source_transform_ 0 Spins the source image around the pivot point. Use
rotate center x,y to position the pivot point.
You can adjust rotate values by dragging the


is disabled.
scale source_transform_ 1 Resizes the source image on the x and y axes.

scale
You can adjust scale values by dragging the transform
handles in the Viewer.

is disabled.
skew X source_transform_ 0 Skews the source image along the X axis from the
skewX pivot point. Use center x,y to position the pivot
point.
REFERENCE GUIDE
154
Roto |

Value
You can adjust skew values by dragging the transform


is disabled.
skew Y source_transform_ 0 Skews the source image along the Y axis from the
skewY pivot point. Use center x,y to position the pivot
point.
You can adjust skew values by dragging the transform


is disabled.
skew order source_transform_ XY Sets the order in which skew X and skew Y are
skew_order applied:
• XY - skew X is applied before skew Y.
• YX - skew Y is applied before skew X.
center x,y source_transform_ N/A Adjusts the center of rotation and skew on the x and y
center axes. This control does not translate the source
image.


is disabled.
filter source_filter Cubic Select the filtering algorithm to use when remapping
pixels from their original positions to new positions.
This allows you to avoid problems with image quality,
particularly in high contrast areas of the frame (where
highly aliased, or jaggy, edges may appear if pixels
are not filtered and retain their original values).
REFERENCE GUIDE
155
Roto |

Value
Note: If source is set to color on the Shape

tab, the filter dropdown is disabled.
• Impulse - remapped pixels carry their original

values.
• Cubic - remapped pixels receive some smoothing.
• Keys - remapped pixels receive some smoothing,
plus minor sharpening (as shown by the negative -y
portions of the curve).
• Simon - remapped pixels receive some smoothing,
plus medium sharpening (as shown by the negative
-y portions of the curve).
• Rifman - remapped pixels receive some smoothing,
plus significant sharpening (as shown by the
negative -y portions of the curve).
• Mitchell - remapped pixels receive some
smoothing, plus blurring to hide pixelation.
• Parzen - remapped pixels receive the greatest
smoothing of all filters.
• Notch - remapped pixels receive flat smoothing
(which tends to hide moire patterns).
• Lanczos4, Lanczos6, and Sinc4 - remapped pixels
receive sharpening which can be useful for scaling
down. Lanczos4 provides the least sharpening and
Sinc4 the most.
black outside source_black_ disabled When rotating or translating the source image, a part
outside of the image area may get cropped. To fill the
cropped portion with black, check black outside. To
fill the cropped portion by expanding the edges of
the image, uncheck black outside.
REFERENCE GUIDE
156
Roto |

Value

tab, this control is disabled.
time offset source_time_ 0 Allows you to clone or reveal pixels from a different
offset frame. Time offset is either relative to the current
frame (-1 is the frame previous to the current one) or
absolute (1 is the first frame in the clip).
source_time_ relative
offset_type If source is set to color on the Shape tab, this control
is disabled.
view source_view current Sets the view you want to clone from.

Lifetime Tab
lifetime type lifetime_type single frame Sets the range of frames during which the currently
selected shape or group is visible:
• all - the selection appears in all frames of the
composition.
• start to frame - the selection appears from the first
frame to the specified to frame.
• single - the selection appears on one frame only.
You can specify the frame using the from field.
• to end - the selection appears from the specified
from frame to the last frame.
• range - the selection appears from the specified
from frame up to the specified to frame.
from lifetime_start 1 Sets the first frame for single, to end, and range
lifetime types.
REFERENCE GUIDE
157
Roto |

Value
to lifetime_end 1 Sets the last frame for start to frame and range
lifetime types.
Tracking Tab
Settings
track channels rgb The color correction is only applied to these channels.
channels
You can use the checkboxes on the right to select
pre-track pretrack_filter none Before image patches are compared, the selected
filter filter is applied. You can select one of the following
options:
• none - This disables all pre-filtering, which allows
you to have full control of tuning the input image
for tracking.
• adjust contrast - This stretches the image contrast
to better suit the tracking algorithm. This option is
recommended.
adjust for adjust_for_ disabled Enabling this option performs extra pre-filtering to
luminance luminance_ help compensate for changes in brightness in the
changes changes image over time.
clamp super- clamp_footage enabled When enabled, the pixel values in the tracked area are
white, sub- clamped between 0-1.
zero footage
If you want to track the full dynamic range of your
footage, you should disable this control.
hide hide_progress_bar disabled When enabled, the tracking progress bar is hidden.
progress bar
Export
export menu export_menu CornerPin2D Sets the node type to export the track data:
(relative) • CornerPin2D (relative) - warp the image according
REFERENCE GUIDE
158
Roto |

Value
to the relative transform between the current frame

and the reference frame.
• CornerPin2D (absolute) - set the to positions to
the corners of the planar surface, and the from
positions to the dimensions of the currently
selected node. This option places the incoming
image entirely inside the planar surface.
• CornerPin2D (stabilize) - applies the inverse
transform of the track to stabilize the image. This
option locks down the image so that the tracked
plane effectively doesn't move - can be very useful
for applying drift corrections.
• Tracker - creates a four-corner Tracker node with
each of the trackers taking the positions of planar
surface corners. This allows you to use the Tracker's
transform functions to stabilize, reduce jitter, and so
on. This export option is always baked out in order
for the transform to work correctly.
create export_button N/A Click to create the selected node containing the track
data.
link output link_output enabled When enabled, create a link between Roto and the
exported node, so that when the track updates, the
points are updated too.
Correction
CornerPin points
point1 xy pt1 N/A The CornerPin points are populated automatically

when you track an object. When you draw a roto
point2 xy pt2 shape and convert it into a track object, Nuke
point3 xy pt3 automatically places four corner pins around the
shape. These are the points that are tracked.
point4 xy pt4
Offsets
REFERENCE GUIDE
159
RotoPaint |

Value
offset1 xy of1 N/A You can correct the four automatically placed points
by offsetting any or all of the four points. To offset a
offset2 xy of2 point, simply click and drag it in the Viewer to the
offset3 xy of3 correct position.
offset4 xy of4
reference reference_frame The first The frame used as a reference to compare all other
frame frame used frames containing tracking data.
to track from.
RotoPaint
RotoPaint is a vector-based node for help with tasks like rotoscoping, rig removal, garbage matting and
dustbusting. You can draw Bezier and B-Spline shapes with individual and layer group attributes, including
per-point and global feather, motion blur, blending modes and individual or hierarchical 2D
transformations.
The RotoPaint node gives you a broader scale of tools to use than Roto, though many of the controls are
shared across both nodes. As with the Roto node, you should use the Viewer tools to create shapes and
paint strokes, and then edit them using the control panel.
The toolbar on the left side of the Viewer has point selection and manipulation, and shape creation tools
identical to Roto nodes. In addition, RotoPaint nodes also have brush, clone/reveal, blur, and dodge/burn
tools. Click and hold or right-click on a toolbar button to open a sub-menu to select any of its available
tool types. Options related to the current tool appear in a toolbar along the top of the Viewer. Click on a
toolbar item to cycle through the available options for that class of tools.

1. Select your tool from the left-hand Viewer tool bar.
2. Use the RotoPaint tool settings above the Viewer to adjust the tool’s properties.
3. When painting, you can:
• Shift+drag on the Viewer to change brush size.
REFERENCE GUIDE
160
RotoPaint |
• Clone tool: Ctrl/Cmd+click and drag to set the clone offset. To reset the clone offset,
Ctrl/Cmd+Shift+click and drag.
4. When drawing your shapes, you can:
• Click in the Viewer to place points. You can drag while clicking to pull out Bezier handles or adjust B-
Spline tension.
• Ctrl/Cmd+drag to sketch the shape freely.
• Click the first point or press Return to close the shape. To leave the shape open, press Esc.
• Ctrl/Cmd+Alt+click to add points to an existing shape.
• Select a point and press Z to increase the smoothness of the point.
• Select the point and press Shift+Z to cusp the point.
• Select the point and press Delete to delete a point.
• Beziers: Shift+drag on a tangent handle to snap the opposite handle to the same length.
Ctrl/Cmd+drag on a tangent handle to move it independently of its opposite handle.
• B-Splines: Ctrl/Cmd+Alt+drag on a point to adjust its tension.
5. Select a shape or stroke using the Select tools or the shape/stroke list in the control panel.
6. Use the control panel to adjust or fine-tune your shape(s) or stroke(s).
You can also create this node by pressing P on the Node Graph.
See also Roto, Project3D, F_RigRemoval, and F_WireRemoval.
Inputs and Controls

Type Name
Input bg Background input - adding a background automatically creates

another bg input allowing you to connect up to four images.
mask An optional image to use as a mask. By default, the roto shapes and
paint strokes are limited to the non-black areas of the mask.
or set to none.
REFERENCE GUIDE
161
RotoPaint |

Value
RotoPaint Tab
output output rgba The roto shapes and paint strokes are rendered into
these output channels. The output channels are the
same for all shapes and paint strokes created using
this node - you cannot create a subset of shapes or
strokes and output them to a different channel.

channels.
premultiply premultiply none Premultiply multiplies the chosen input channels with
a mask representing the roto shapes and paint strokes.
For example, where there are no shapes/strokes (the
matte is black or empty) the input channels will be set
to black. Where the shapes/strokes are opaque (the
matte is white or full) the input channels keep their full
value.
Note that selecting rgba premultiplies the alpha

against itself (a*a). If you don’t want this to happen, set
premultiply to rgb.
clip to cliptype format Sets how to restrict the output image:

bounding box.
format area.
• union bbox and format - restrict the output image
to a combination of the incoming bounding box and
format area.
• intersect bbox and format - restrict the output
REFERENCE GUIDE
162
RotoPaint |

Value

alpha channel that you want to throw away.
format dropdown
format format root.format This is used if RotoPaint has no input connected. It is

the format which the node should output in the
absence of any available input format. If an input is
connected, this control has no effect.
If the format does not yet exist, you can select new to
create a new format from scratch. The default setting,
root.format, resizes the image to the format
indicated on the Project Settings dialog box.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the output is

channel. Injecting the mask allows you to use the same
mask further downstream.
invert invert_mask disabled Inverts the use of the mask channel, so that output is
the mask.

mask.
output mask N/A disabled Enables the associated output mask channel to the
right. Disabling this checkbox is the same as setting
the channel to none.
REFERENCE GUIDE
163
RotoPaint |

Value
outputMask none Output mask channel. This is a channel where

RotoPaint will output a mask for what it rendered.
The mask can be useful, for example, if you need to

apply grain to the areas you’ve painted, but you don’t
want to double up the gain in other areas.
color color 1 Sets the color for the roto shape or paint stroke. This
control only has an effect when source is set to color
on the Shape or Stroke tab.
opacity opacity 1 Sets the opacity of a new or existing shape or stroke in

the shape/stroke list. 1 is equal to totally opaque and
0 is equal to transparent.
source paint_source color Sets the source color for the selected shape/stroke:
including any shapes and strokes drawn on it. This
input is mainly used with the Clone tab controls.
Using it may cause slower renders.
including any shapes and strokes drawn on it.
and bg3 inputs.
blending blending_mode over Sets how the colors in the current shape (A) are
mode merged with the underlying image and shapes (B):
• color-burn (darken B towards A) - darkens B to
reflect A by increasing the contrast. No part of the
image becomes lighter.
• color-dodge (brighten B towards A) - brightens B to
reflect A by decreasing the contrast. No part of the
image is darkened.
• difference (abs(A-B)) - subtracts either B from A or
vice versa, depending on which is brighter. Blending
with white inverts A, while blending with black
REFERENCE GUIDE
164
RotoPaint |

Value
produces no change. Similar colors will return black

pixels.
• exclusion (A+B-2AB) - creates a result similar to the
difference mode but lower in contrast. Like with
difference, blending with white inverts A. Blending
with black produces no change.
• hard-light (multiply if A<.5, screen if A>.5) - lightens
highlights and darkens shadows. If B is lighter than
50% gray, the result lightens as if it were screened. If
B is darker than 50% gray, the result is darkened as if
it were multiplied.
blending • minus (A-B) - subtracts B from A.

mode • multiply (AB, A if A<0 and B<0) - multiplies A by B.
(continued) The result is always darker. Blending with black gives
• over (A over B) - the colors of the two images will
not interact in any way, and Nuke will display the full
value of the colors in A.
• overlay (multiply if B<.5, screen if B>.5) - depending
on A, multiplies or screens the colors. B brightens A
while preserving highlights and shadows.
• plus (A+B) - the sum of the two colors. Increases
brightness to lighten A and reflect B. Plus is similar to
the screen blending mode, but produces a more
extreme result.
REFERENCE GUIDE
165
RotoPaint |

Value
• screen (A+B-AB if A and B between 0-1, else A if A>B

else B) - this is a soft plus making everything brighter
but ramping off the whites. Light colors have more
of an effect than dark colors. The result is always a
lighter color. Blending with black leaves the pixel
unchanged, blending with white always returns
white. The result is similar to projecting multiple
slides on top of each other.
• soft-light - depending on A, darkens or lightens the
colors. Less extreme than the hard light blending
mode.
view view N/A Sets the view(s) in which you want to draw your shape
(s) or stroke(s). This is only displayed if you have set up
more than one view in your Project Settings (typically,
if you’re working on a stereoscopic project).
spline key key 0 Quick access to spline keyframes:

• Numeric fields - show whether the spline is keyed at
this frame, which key you’re viewing, and the total
number of spline keys.
• - move to the previous or next spline key.
• - set or delete spline keys.
visible visible enabled Sets whether the selected shape, paint stroke, or
group is visible and rendered or not. You can still edit
an invisible shape/stroke and view its position in the
Viewer.
locked locked disabled Locks the selected shape, stroke, or group to prevent
it from being edited.
stereo offset stereo_offset 0 Moves the selected stroke, shape, or group on the x
x,y and y axes. This is an extra transform that is applied
after all other transforms. Typically, you would
position the stroke, shape, or group correctly in the
hero view, then split this control, and drag the stroke
REFERENCE GUIDE
166
RotoPaint |

Value
or shape to its correct location in any other views.

Note that you can also press Shift while dragging to
constrain the movement to x or y axis only.
Stereo offset can be useful, for example, if you have a

stroke, shape, or group that is correctly positioned in
one view and you want to move it to its correct
location in another view, but can’t use the translate
control on the Transform tab because that’s being
driven by Tracker data.

shape/stroke curves N/A Shows the hierarchy of shapes, paint strokes, and
list groups, allowing you to adjust them and how they’re
displayed in the Viewer:
• Name - double-click to edit the shape, stroke, or
group Name.
• - whether the shape, stroke, or group is visible
and rendered.
• - lock or unlock the shape, stroke, or group.
• - set the shape outline color to appear in the

Viewer.
• - set the color in which you want to render your
shape/stroke.
• - invert the shape, stroke, or group.
• - set the shape, stroke, or group blending mode.
• - whether motion blur is applied to the shape.

• View - displays the view(s) the shape/stroke appears
in. This is only displayed if you have set up more
than one view in your Projects Settings (typically, if
REFERENCE GUIDE
167
RotoPaint |

Value
• Lifetime - the range of frames during which the

shape/stroke is visible.
• Source - the shape or stroke source (as set on the
Shape, Stroke, or RotoPaint tab).
Use the shape/stroke list to:

groups.
• Right-click for a context menu containing Add new
layer and shape/stroke editing functions such as
Copy and Paste.
Transform Tab
translate x,y translate 0 Moves the selected shape, stroke, or group on the x
and y axes.
You can also adjust tranlate values by dragging the

rotate rotate 0 Spins the selected shape, stroke, or group around the
pivot point. Use center x,y to position the pivot point.

REFERENCE GUIDE
168
RotoPaint |

Value
scale scale 1 Resizes the selected shape, stroke, or group on the x

and y axes.

skew X skewX 0 Skews the selected shape, stroke, or group along the X
axis from the pivot point. Use center x,y to position
the pivot point.

skew Y skewY 0 Skews the selected shape, stroke, or group along the Y
axis from the pivot point. Use center x,y to position
the pivot point.

applied:
center x,y center 1024, 778 Adjusts the center of rotation and skew on the X and Y
axes. This control does not translate shapes or strokes.

extra matrix transform_matrix N/A Adds an extra matrix which is calculated after the
transforms defined by the other controls. This can be
useful, for example, if you are using a 3rd party
application, such as a tracker, and your tracking data
cannot be fully represented using the available
transform controls.
shape/stroke curves N/A See the RotoPaint tab for the shape list controls.
list
REFERENCE GUIDE
169
RotoPaint |

Value
Motion Blur Tab
shape global_ enabled When enabled, motion blur is applied to shapes

motionblur_on selected in the shape list or Viewer.
shape motion blur determines the exposure for each

moving shape and blends the resulting blurred shape.
This may be more efficient than the global motion
blur since each shape will only be blended once.
Note: Shape motion blur may result in

artifacts when shapes blur over the same
region.
global disabled When enabled, motion blur is applied to all shapes in

the current RotoPaint node.
Note: Global motion blur overrides the per-

shape motion blur settings.
global motion blur correctly accounts for interaction

between motion blurred shapes.This may be more
expensive than the shape motion blur since it may
blend each shape for every sample.
Note: Global motion blur requires that

shutter and sampling parameters are the
same for all shapes and has been optimized
for consecutive shapes with the same
properties using the over blend mode.
Shape Blur
motionblur motionblur 1 Sets the number of motion blur samples for the
selected shape(s). Increase this for better quality but
REFERENCE GUIDE
170
RotoPaint |

Value
slower rendering, decrease it for faster rendering.
on motionblur_on disabled When enabled, motion blur is applied to the selected

shape(s).
shutter motionblur_ 0.5 Sets the number of frames the shutter stays open
shutter when motion blurring. For example, a value of 0.5
render time.
shutter motionblur_ start Controls how the shutter behaves with respect to the
offset shutter_offset_ current frame value. There are four options:
type • centred - center the shutter around the current
frame 30 to 31.
frame 29 to 30.
• custom - open the shutter at the time you specify. In
the field next to the dropdown menu, enter a value
(in frames) you want to add to the current frame. To
open the shutter before the current frame, enter a
negative value. For example, a value of - 0.5 would
open the shutter half a frame before the current
frame.
[offset] motionblur_ 0 If the shutter offset control is set to custom, this

shutter_offset control is used to set the time that the shutter opens
by adding it to the current frame.
REFERENCE GUIDE
171
RotoPaint |

Value
Global Blur
motionblur global_ 1 Sets the number of motion blur samples for all shapes
motionblur in the current RotoPaint node. Increase this for better
quality but slower rendering, decrease it for faster
rendering.
shutter global_ 0.5 Sets the number of frames the shutter stays open
motionblur_ when motion blurring. For example, a value of 0.5
shutter corresponds to half a frame. Increasing the value
render time.
shutter global_ start Controls how the shutter behaves with respect to the
offset motionblur_ current frame value. There are four options:
shutter_offset_ • centred - center the shutter around the current
type frame. For example, if you set the shutter value to 1
frame 30 to 31.
frame 29 to 30.
• custom - open the shutter at the time you specify. In
the field next to the dropdown menu, enter a value
(in frames) you want to add to the current frame. To
open the shutter before the current frame, enter a
negative value. For example, a value of - 0.5 would
open the shutter half a frame before the current
frame.
[offset] global_ 0 If the shutter offset control is set to custom, this
REFERENCE GUIDE
172
RotoPaint |

Value
motionblur_ control is used to set the time that the shutter opens
shutter_offset by adding it to the current frame.

list
Shape Tab
source shape_source color Sets the color or source input for the selected shape:
including any shapes drawn on it. This input is mainly
used with the Clone tab controls. Using it may cause
slower renders.
including any shapes drawn on it.
and bg3 inputs.
feather feather 0 Softens the edges of a roto shape by fading it from

more transparent to less transparent.
With positive feather values, your feather effect is

outward. If your feather values are negative, the
feather effect is inward.
You can also add feathering to individual points by

Ctrl/Cmd+dragging a point in the Viewer. Press E to
increase the feathering of selected points, or Shift+E
to remove the feathering.
If you check feather link in the Roto tool settings

above the Viewer, you can move the shape point and
the feather point together. Otherwise, they move
independently. You can also Ctrl/Cmd+drag a shape
point to temporarily disable feather link and move
REFERENCE GUIDE
173
RotoPaint |

Value
the points independently.
on feather_on enabled When enabled, feathering is applied to the selected

shape(s).
Disabling this is the same as setting feather to 0.
feather feather_falloff 1 Sets the rate of opacity falloff at a feathered edge for
falloff a shape. This is measured in pixels.
[feather feather_type linear Controls the center point of the feather falloff:
type] • linear - the falloff changes linearly from the shape
edge to the feather edge.
closer to the shape edge than the feather edge.
closer to the feather edge than the shape edge.
• smooth - the center point of the feather falloff lies
halfway between the shape edge and the feather
edge.
These options may be helpful, for example, in

matching the soft edge to motion blurred image
content.
overall openspline_width 10 Sets the overall spline width of selected splines and all
width new open splines.
start type openspline_start_ rounded Sets the style of the first point for the selected spline
• rounded - the starting point of the spline has a
REFERENCE GUIDE
174
RotoPaint |

Value
rounded end.
• square - the starting point of the spline has a square
end.
end type openspline_last_ rounded Sets the style of the last point for the selected spline
• rounded - the finishing point of the spline has a
rounded end.
• square - the finishing point of the spline has a
square end.
invert inverted disabled Swaps the colors of the currently selected shape and
the underlying image.
list
Stroke Tab
source brush_source color Sets the color or source input for the selected stroke:
including any shapes and strokes drawn on it. This
input is mainly used with the Clone tab controls.
Using it may cause slower renders.
including any shapes or strokes drawn on it.
and bg3 inputs.
brush type brush_type paint Brush types selected from the control panel only
affect existing strokes. Use the Viewer tools to create
new strokes.
Specify the brush type for the selected stroke:

• paint - applies colored or blended paint strokes.
• smear - pulls the pixels under the cursor in the
direction of the paint stroke.
REFERENCE GUIDE
175
RotoPaint |

Value
• blur - blurs the area under the paint stroke. You can
adjust the blur amount using the effect control.
• sharpen - sharpens the area under the paint stroke
by increasing contrast. You can adjust the sharpening
using the effect control.
brush size brush_size 25 Sets the brush stroke diameter in pixels.
Alternatively, you can Shift+drag in the Viewer to set

the brush size.
brush brush_spacing 0.05 Sets the distance in pixels between paint brush dabs. A
spacing higher setting will increase the space between dabs,
creating a dotted line effect when painting. A lower
setting will decrease the distance and create a solid
brush stroke.
brush brush_hardness 0.2 Sets the brush hardness, which defines the rate of
hardness falloff from the brush center to its edge. Increasing
the hardness decreases blurring at the stroke edges,
and vice versa.
You can also tie a stroke’s hardness to pen pressure by

checking hardness next to pressure alters.
effect effect_parameter1 0 Controls the strength of blur and sharpen brush

strokes. Paint and smear brushes do not use this
control.
pressure dynamic_ enabled When enabled, a stroke’s opacity is tied to pen

alters transparency pressure.
opacity
pressure dynamic_size disabled When enabled, a stroke’s brush size is tied to pen
alters size pressure.
pressure dynamic_hardness disabled When enabled, a stroke’s hardness is tied to pen
REFERENCE GUIDE
176
RotoPaint |

Value
alters pressure.
hardness
build up buildup enabled When enabled, the stroke builds up when painted
over itself.
write on writeon_start 0 When animating strokes, sets the order in which the
start dabs on a stroke appear over a number of frames.
writeon_end 1
• write on start - set where along the stroke length
write on end
the paint begins. 0 is the start of the stroke, 1 is the
end.
• write on end - set where along the stroke length the
paint ends.
list
Clone Tab
translate x,y source_transform_ 0,0 Moves the source image along the x and y axes.
translate
If source is set to color, this control is disabled.
rotate source_transform_ 0 Spins the source image around the pivot point. Use
rotate center x,y to position the pivot point.

scale source_transform_ 1 Resizes the source image on the x and y axes.

scale
REFERENCE GUIDE
177
RotoPaint |

Value
skew X source_transform_ 0 Skews the source along the X axis from the pivot
skewX point. Use center x,y to position the pivot point.

skew Y source_transform_ 0 Skews the source along the Y axis from the pivot
skewY point. Use center x,y to position the pivot point.

applied:
center x,y source_transform_ N/A Adjusts the center of rotation and skew on the x and y
center axes. This control does not translate the source image.

round to source_translate_ disabled Rounds the translate x,y amount to the nearest whole
pixel round integer pixel. This can help you avoid softening when
using filtering.
filter source_filter Cubic Select the filtering algorithm to use when remapping
highly aliased, or jaggy, edges may appear if pixels are
REFERENCE GUIDE
178
RotoPaint |

Value
not filtered and retain their original values).

tab, the filter dropdown is disabled.

values.
plus medium sharpening (as shown by the negative -
y portions of the curve).
plus significant sharpening (as shown by the negative
Sinc4 the most.
black source_black_ disabled When rotating or translating the source image, a part
outside outside of the image area may get cropped. To fill the
cropped portion with black, check black outside. To
fill the cropped portion by expanding the edges of
the image, uncheck black outside.
REFERENCE GUIDE
179
RotoPaint |

Value
time offset source_time_ 0 Allows you to clone or reveal pixels from a different
offset frame. Time offset is either relative to the current
frame (-1 is the frame previous to the current one) or
absolute (1 is the first frame in the clip).
source_time_ relative
offset_type Note: If source is set to color on the Shape
tab, this control is disabled.
view source_view current Sets the view used as the clone source. To use the
view currently displayed in the Viewer, select current.

list
Lifetime Tab
lifetime type lifetime_type single frame Sets the range of frames during which the currently
selected shape, stroke or group is visible:
• all - the selection appears in all frames of the
composition.
• start to frame - the selection appears from the first
frame to the specified to frame.
• single frame - the selection appears on one frame
only. You can specify the frame in the from control.
• to end - the selection appears from the specified
from frame to the last frame.
• range - the selection appears from the specified
from frame up to the specified to frame.
REFERENCE GUIDE
180
RotoPaint |

Value
from lifetime_start 1 Sets the first frame for single, to end, and range
lifetime types.
to lifetime_end 1 Sets the last frame for start to frame and range
lifetime types.
list

(Scripting) Value
Tracking Tab
Settings
track channels channels rgb The color correction is only applied to these channels.

pre-track pretrack_filter none Before image patches are compared, the selected filter
filter is applied. You can select one of the following options:
none - This disables all pre-filtering, which allows you to

have full control of tuning the input image for tracking.
adjust contrast - This stretches the image contrast to

better suit the tracking algorithm. This option is
recommended.
adjust for adjust_for_ disabled Enabling this option performs extra pre-filtering to help
luminance luminance_ compensate for changes in brightness in the image over
changes changes time.
clamp super- clamp_ enabled When enabled, the pixel values in the tracked area are
white, sub- footage clamped between 0-1.
zero footage
If you want to track the full dynamic range of your
footage, you should disable this control.
hide progress hide_ disabled When enabled, the tracking progress bar is hidden.
REFERENCE GUIDE
181
RotoPaint |

(Scripting) Value
bar progress_bar
Export
export menu export_menu CornerPin2D Sets the node type to export the track data:
(relative)
CornerPin2D (relative) - warp the image according to
the relative transform between the current frame and the
reference frame.
CornerPin2D (absolute) - set the to positions to the

corners of the planar surface, and the from positions to
the dimensions of the currently selected node. This
option places the incoming image entirely inside the
planar surface.
CornerPin2D (stabilize) - applies the inverse transform

of the track to stabilize the image. This option locks
down the image so that the tracked plane effectively
doesn't move - can be very useful for applying drift
corrections.
Tracker - creates a four-corner Tracker node with each

of the trackers taking the positions of planar surface
corners. This allows you to use the Tracker's transform
functions to stabilize, reduce jitter, and so on. This
export option is always baked out in order for the
transform to work correctly.
create export_button N/A Click to create the selected node containing the track
data.
link output link_output enabled When enabled, create a link between RotoPaint and the
exported node, so that when the track updates, the
points are updated too.
Correction
CornerPin points
point1 xy pt1 N/A The CornerPin points are populated automatically when
REFERENCE GUIDE
182
ScannedGrain |

(Scripting) Value
point2 xy pt2 you track an object. When you draw a roto shape and
convert it into a track object, Nuke automatically places
point3 xy pt3 4 corner pins around the shape. These are the points that
point4 xy pt4 are tracked.
Offsets
offset1 xy of1 N/A You can correct the four automatically placed points by
offsetting any or all of the four points. To offset a point,
offset2 xy of2 simply click and drag it in the Viewer to the correct
offset3 xy of3 position.
offset4 xy of4
reference reference_ The first The frame used as a reference to compare all other
frame frame frame used to frames containing tracking data.
track from.
ScannedGrain
Applies actual scans of film grain to replicate the grain of the film stock. Frames are selected from a loop
of scanned film grain and applied to the input image. This allows you to add grain to CG creations to
match the original plate or ’age’ footage artificially.
See also Grain and F_ReGrain.
REFERENCE GUIDE
183
ScannedGrain |
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the grain.
mask An optional image to use as a mask. By default, the grain is limited to

or set to none.
ScannedGrain Tab
grain fullGrain [file dir Sets the file path of your scanned grain images.
$program_
name]/FilmGrain/ To create the film grain images:
1. Film a gray card and scan the result to file.
2. Blur the scanned image.
3. Subtract the blurred image from the original.
4. Add 0.5 to the rgb channels.
5. Save the scanned grain image to the desired
file type (.rgb for example).
frame range fullGrain.first_ 1 Sets the first frame of the grain file to use.
frame
fullGrain.last_ 50 Sets the last frame of the grain file to use.

frame Approximately 50 frames are required to produce
random grain.
min. width minwidth 721 Sets the minimum width (in pixels) that images
must have in order for grain to be applied.
Set a minimum width to ensure that the grain
REFERENCE GUIDE
184
ScannedGrain |
resolution matches the image otherwise you may

get unwanted results.
resize resize enabled When enabled, resizes the grain plate to match
your input.
When disabled, the grain plate is cropped to

match.
flip flip disabled Increases the number of grain images available by

allowing random horizontal and vertical flipping
of the grain plate. Flipping can produce a more
random grain spread, but increases processing
time.
offset offset 0.5 Sets the value of neutral gray to subtract from
images. The closer this value is to the input gray,
the more grain is visible.
Amount of grain
amount rgb amount 0.30000001 Sets the multiplier for the red channel applied to
the grain.
0.40000001 Sets the multiplier for the green channel applied

to the grain.
0.5 Sets the multiplier for the blue channel applied to

the grain.
saturation saturation 1 Sets the intensity of grain hue across all channels.
use use_precomputed enabled When enabled, the output LUT is clamped

precomputed between 0 and 1. This control is enabled by
table default for backward compatibility.
When disabled, Nuke does not clamp the LUT.
curve editor weight N/A The grain sequence’s color curves. You can use
these to edit the gain, gamma, contrast, and so on
for individual channels. You can also look up color
information for the current pixel in the Viewer.
REFERENCE GUIDE
185
ScannedGrain |
reset N/A N/A Returns the selected curve(s) to the default values.
low_clip low_clip 0 Sets the low threshold based on the input image.
In general, when grain is added to an image,

some input pixel values become higher, while
others become lower. This control allows you to
stop pixel values from going below a certain
value, such as the blackpoint in your image.
seed seed 1 A different grain pattern is produced for each

frame. Change this number to get a different
initial pattern, for instance if you are using
multiple ScannedGrain nodes.
If you require an identical grain plate for every

frame, enter -frame in this field.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the grain

invert invert_mask disabled Inverts the use of the mask channel, so that grain
is limited to the non-white areas of the mask.
fringe fringe disabled When enabled, only apply the effect to the edge
of the mask.

mask.
(un)premult N/A disabled Enables the associated channel to the right.

by Disabling this checkbox is the same as setting the
channel to none.
REFERENCE GUIDE
186
Sparkles |
unpremult none The image is divided by this channel before

being processed, and multiplied again afterwards.
If you are using premultiplied input images, you

may want to check (un)premult by and select
rgba.alpha here. This will simulate applying the
grain before the premultiplication was done. It is
the same as adding an Unpremult node before
this node and a Premult node after, but allows
you to work faster if you’re only using one
ScannedGrain node.
If you are using unpremultiplied input images,

you should leave this set to none.
mix mix 1 Dissolves between the original image at 0 and the

full grain effect at 1.
Sparkles
Creates customizable rays or sparkles from a central position indicator in the Viewer.
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the sparkles.
REFERENCE GUIDE
187
Sparkles |

(Scripting) Value
Sparkles Tab
Shape
Sparkle type mode Rays Sets the sparkle type to output from the position
indicator in the Viewer:
• Rays - produces rays fading away from the origin.
• Sparkles - produces intermittent rays fading at either
end of each ray.
• Fireworks - produces rays fading toward the origin.
triangle triangle_think disabled When enabled, a triangle shape it used to create the
shapes rays increasing the thickness.
position xy position 700, 600 Sets the center of origin for the rays on the x and y axes.
size size 300 Sets the size of the rays, in pixels, and adjusts the ray
bounding box accordingly.
Note: When Sparkle type is set to Sparkles,

this control determines the size of the longest
ray.
anamorphic anamorphic 1 Sets the anamorphic ratio of the bounding box.

Negative values decrease width and vice versa.
rotation rotation 0 Sets the amount of rotation applied to the rays.

Negative values produce clockwise rotation and vice
versa.
relative relative_ disabled When enabled, 0 degrees always points to the center of
rotation the image and rotation is relative to that value.
rays Nb ray_nb 50 Sets the number of rays emitted.
thickness ray_thick 0.2 Sets the thickness of the emitted rays in pixels.
random random_thick disabled When enabled, ray thickness is randomized within the
range [1,thickness].
REFERENCE GUIDE
188
Sparkles |

(Scripting) Value
sparkle falloff spark_falloff 0.05 When Sparkel type is set to Sparkles, determines
whether or not rays are connected to the position
indicator. Higher values decrease the distance from the
sparkle to the position indicator and vice versa.
ray falloff ray_falloff 2 Sets how quickly rays falloff as distance from the
position indicator increases.
Note: This control behaves differently

depending on the Sparkle type selected. For
example, Fireworks falloff towards the
position indicator.
seed seed 23 Sets the random seed that controls ray characteristics. If
you’re using multiple Sparkle nodes, you can use the
same seed to produce similar random results on both
nodes.
Color
brightness brightness 1 Sets ray brightness - higher values increase brightness.
color color 1 Sets the ray color.
chroma shift chroma_shift 0 Sets the amount of chroma shift applied to the rays.
Positive values split red clockwise and blue counter-
clockwise and vice versa.
chroma chroma_ 0 Sets the spread of chroma values from the position
spread spread indicator. Positive values extend the spread of the red
channel while reducing the blue channel and vice versa.
Mask
channel to none.
vismask none The channel to use as a mask. By default, the sparkles are
REFERENCE GUIDE
189
Sparkles |

(Scripting) Value
mask blur vismask_size 4 Sets the size of the mask when mask is set to anything
other than none.
Effects Tab
Sparkle crawl - These controls do not affect Rays.
Sparkle crawl sparkle_crawl disabled When enabled, use the Sparkle crawl controls to cause
Sparkles or Fireworks to move, or crawl.
motion motion 300 Sets the amount of motion to apply. Animating this
control causes Sparkles or Fireworks to crawl.
direction direction 0 Sets the direction of crawl in degrees. Animate this

control to produce more crawl.
ray fade fadeTolerance 30 Controls the amount of fade to apply to Sparkles only.
Broken rays
Broken rays broken_rays disabled When enabled, use the Broken rays controls affect the
rays emitted from the position indicator.
% ray broken_ 50 Sets the percentage of rays to affect with breaks.

affected affected
break start broken_start 0.7 Sets where the first break occurs as a percentage of the
total length. For example, 0.5 is equal to 50% of the
length.
random random_ disabled When enabled, the break start control is randomized.
broken_start
breaks broken_holes 0.2 Sets the number of breaks per ray after the break start.
random random_ disabled When enabled, the breaks control is randomized.

broken_holes
Sparks
sparks sparks disabled When enabled, use the Sparks controls to clip the
(direction emission angle and direction.
clip)
REFERENCE GUIDE
190
Text |

(Scripting) Value
direction sparks_ 0 Sets the ray’s emission direction in degrees.

direction
open angle sparks_angle 45 Sets the angle from which rays can be emitted. For
example, 360 causes rays to emit in all directions.
Text
This node lets you add text overlays on your images. You can simply type in the text you want to have
displayed or use Tcl expressions or Tcl variables to create a text overlay. Text overlays can also be
animated using animation layers in the Groups tab, so that their properties (such as position, size, and
color) change over time. These features make the Text node useful, for example, for creating slates or
scrolling credits.
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the text overlay.
mask An optional image to use as a mask. By default, the text is limited to

or set to none.
REFERENCE GUIDE
191
Text |

Value
Text Tab
output output rgba The text only appears in these channels.

premult premult none These channels are multiplied by the drawn text, so
that they are set to black outside the text shape.

bounding box.
format area.
• union bbox+format - restrict the output image to
the combination of the incoming bounding box
and format area.
to the intersection of the incoming bounding box
and format area.
replace replace disabled When enabled, the affected channels are cleared to
black before drawing on them.
When disabled, the text is drawn on top of the input

image.
invert invert disabled Inverts the inside and outside of the text shape.
opacity opacity 1 Adjusts the opacity of the text. The possible values
run from 0 (invisible) to 1 (fully opaque).
channel to none.
REFERENCE GUIDE
192
Text |

Value
maskChannelInput none The channel to use as a mask. By default, the text is

invert invert_mask disabled Inverts the use of the mask channel so that the text is
message message None Enter the text you want to display, a Tcl expression, a
Tcl variable, or a combination of these. Enter Tcl
expressions in square brackets, for example, [date].
To begin a new line, press Return.
To display special Unicode characters, such as

foreign language characters and copyright signs, you
can:
• use HTML named entities, such as © to
display ©
• use hex entities, such as © to display ©
• use decimal entities, such as © to display ©
• type Unicode characters, such as ©, on your
keyboard or cut and paste them from other
applications. UTF-8 character encoding is used to
store them in the control's value and in the saved
Nuke script.
The above only work if the font you are using has the
character you want to display in it.
REFERENCE GUIDE
193
Text |

Value
Note: We recommend using the above

entities rather than typing <, for example.
This is because future versions of the Text
node may interpret HTML mark-up. In
HTML, some characters, such as the greater
than and less than signs, are reserved. If you
used these signs within your text now,
future versions could mistake them for
HTML mark-up.
See Creating Text Overlays in the Nuke User Guide,

for examples of Tcl expressions, Tcl variables, HMTL
named entities, hex entities, and decimal entities you
can use in the Text node.
Tip: To get a list of all the Tcl expressions

you can use with date, you can also type X
on the Node Graph, set the script
command dialog that opens to Tcl, enter
date -h, and click OK.
box xyrt (or box N/A Adjusts the on-screen box that limits the text inside a
xywh) certain area of the frame:
• To define the left boundary of the box, adjust the x
field.
• To define the bottom boundary of the box, adjust
the y field.
• To define the right/width boundary of the box,
adjust the r field.
• To define the top/height boundary of the box,
adjust the t field.
You can also move and resize the box in the Viewer.
Your text is wrapped inside the box you defined.
REFERENCE GUIDE
194
Text |

Value
justify xjustify left Sets how to align the text horizontally:

• left - align the text along the left edge of the on-
screen text box. This leaves the right edge of the
text ragged.
• center - align the text from the center of the on-
screen text box. This leaves both edges of the text
ragged.
• right - align the text along the right edge of the
on-screen text box. This leaves the left edge of the
text ragged.
• justify - align the text both along the left and the
right edge of the on-screen text box. This leaves no
ragged edges. The justification is done by
expanding the spaces between letters. If there are
no spaces or the spaces get more than about three
times wider than they were, letters are expanded.
yjustify top Sets how to align the text vertically:

• top - align the text against the top edge of the on-
screen text box.
• center - align the text from the center of the on-
screen text box.
• bottom - align the text against the bottom edge of
the on-screen text box.
font font Utopia Sets the font to use for the text. This control uses the
FreeType library and supports a large number of
fonts, including TrueType (.ttf) fonts and PostScript
fonts (.pfa and .pfb).
Tip: You can liimit the Text node to only

use the fonts that ship with Nuke by
navigating to Project Settings > Font and
disabling include system fonts.
REFERENCE GUIDE
195
Text |

Value
index Regular Sets the font style to use for the text. The options
available in this control depend on the font selected.
global font global_font_scale 1 Sets the scale used to multiply the font size for all
scale characters in the message field.
font size font_size 100 Adjusts the size of the font. When leading is set to 0,
this parameter also controls the spacing between
each line of text.
When rendering the font, the size parameter controls

the font hinting used. Font hinting adjusts which
pixels are interpolated to more clearly render a font.
At small sizes and on low resolution output devices,
it has a big impact on the legibility of the font. For
best results, you should use this parameter (rather
than the scale parameter on the Transform tab) to
control the size of the font and keep scale set to 1.
font width font_width 100 Sets the width of the font, as a percentage of the
font size.
font height font_height 100 Sets the height of the font, as a percentage of the
font size.
kerning kerning 0 Sets the spacing between the currently selected

character and the previous character. Negative values
move the character closer to the previous character
and vice versa.
Note: Kerning has no effect on multiple

selections or the first character in a string.
tracking tracking 0 Sets the spacing between each character and the
previous character. Negative values move characters
closer to each other and vice versa.
REFERENCE GUIDE
196
Text |

Value
Note: Tracking has no effect on the last

character in a string.
baseline shift baseline_shift 0 Sets the height above the font baseline for the
currently selected characters. Negative values lower
the characters and vice versa.
leading leading 0 Sets the spacing between lines of text. Negative

values decrease the spacing and vice versa.
Groups Tab - the controls on this tab only affect groups in the animation layers table.
translate xy translate 0, 0 Translates the text on the x and y axes.
rotate rotate 0 Rotates the text by degrees. Negative values

scale scale 1 Scales the text. Scale width and height are ganged by
default.
skew X skewX 0 Skews the text on the x axis.
skew Y skewY 0 Skews the text on the y axis.
applied:
• XY - skew X is applied before skew Y.
• YX - skew Y is applied before skew X.
center xy center N/A Sets the center of rotation and scaling. You can also
move it using Ctrl/Cmd+drag.
animation group_animation N/A Displays groups created from the text in the
layers message field, allowing you to animate the required
characters using the transform controls.
Select the required text in the Viewer or message

field and then click the + button to create an
animation layer. Select the group and apply the
REFERENCE GUIDE
197
Text |

Value
required transforms using the controls on the Group

tab.
Color Tab
ramp ramp none Select whether to create a color gradient across the
text:
into another.
point 0 end. This means colors in the point 0 end
are spread wider than colors in the point 1 end.
end. This means colors in the point 1 end are
the ramp.
color color 1 Sets the color for the text.

top end).
The following controls are only available if ramp is set to anything other than none.
point 1 p1 100, 400 Sets the position of the point 1 indicator. This allows
you to adjust the spread and angle of the ramp.
point 0 p0 100, 100 Sets the position of the point 0 indicator. This allows
you to adjust the spread and angle of the ramp.
REFERENCE GUIDE
198
Text |

Value
Shadows Tab
enable drop enable_shadows disabled When enabled, drop shadows are applied to the text.
shadows
When disabled, all other controls on this tab are also
disabled.
inherit input shadow_inherit_ disabled When enabled, shadows inherit their color from the
color input_color text's input color.
When disabled, the color controls are enabled,

allowing you to set the shadow color independently.
color color 0,0,0,1 Sets the color of the drop shadow when inherit
input color is disabled.
opacity opacity 0.7 Sets the opacity of any drop shadow applied to the
text, where 0 is totally transparent and 1 is totally
opaque.
angle angle 225 Sets the cast angle for drop shadows relative to the
text, in degrees, where 0 is left to right.
distance distance 5 Sets the seperation distance between the text and
the drop shadow.
softness softness 0 Sets the softness, or blur, of any drop shadow

applied to the text. Higher values produce greater
blur.
shrink/expand shrink/expand 0 Sets the thickness of drop shadows relative to the

text. Negative values produce thinner shadows and
vice versa.
REFERENCE GUIDE
199
Time Nodes
Time nodes deal with distorting time (that is, slowing down, speeding up, or reversing clips), applying
motion blur, and performing editorial operations like slips, cuts, splices, and freeze frames.
Add 3:2 pulldown

Adds 3:2 pulldown to the input footage.
3:2 pulldown is a process of converting 24 frames per second (fps) film to 29.97 fps interlaced video. In
this process, the film is first slowed down 0.1% to match the speed difference between 29.97 fps and 30
fps. Then, film frames are distributed across video fields in a repeating 3:2 pattern in order to make 24 film
frames fill the space of 30 video frames per second:
• The first frame of film is copied to 3 fields of video,
• the second frame of film is copied to 2 fields of video,
• the third frame of film is again copied to 3 fields,
• the fourth frame of film is copied to 2 fields, and so on.
If you’re using footage that was originally film but was converted to video, you may have used the
Remove 3:2 pulldown node when importing the footage to Nuke. In this case, the Add 3:2 pulldown node
allows you to reintroduce 3:2 pulldown before rendering the footage back to video.
See also Remove 3:2 pulldown.
REFERENCE GUIDE
200
AppendClip |
Inputs and Controls

Type Name
Input 1 The image sequence to add 3:2 pulldown to.

(Scripting) Value
User Tab
phase phase 0 The 3:2 pulldown process results in:

• three whole frames that contain two fields from the
same film frame, and
• two split-field frames that contain fields from two
different film frames. The two split-field frames are
always adjacent to each other.
This control lets you shift the point at which the two
split-field frames occur within the first five frames of the
footage, relative to your first input frame.
AppendClip
AppendClip lets you splice clips. Splicing refers to joining clips head-to-tail, thus allowing action to flow
from one shot to the next. When you splice clips, you have options for:
• Fading to or from black.
• Dissolving between input clips.
• Slipping the combined clip in time.
To make an edit, you can first use the FrameRange node to cut portions out of input sequences and then
use the AppendClip node to link them together.
See also FrameRange.
REFERENCE GUIDE
201
AppendClip |
Inputs and Controls

Type Name
Input 1 The image sequence to which subsequent clips are appended. You
can connect any number of sequences to an AppendClip node.

(Scripting) Value
AppendClip Tab
Fade In fadeIn 0 Sets the number of frames to fade in from black on the
first clip in the sequence.
Fade Out fadeOut 0 Sets the number of frames to fade out to black on the
last clip in the sequence.
Cross Dissolve dissolve 0 Sets the number of frames to use to dissolve between
clips.
Bear in mind that differences in format between

dissolved clips are not catered for, so you may want to
use Cross Dissolve in conjunction with Reformat nodes.
First Frame firstFrame 1 Sets the amount of frames to slip the start of the first
clip in the sequence. For example, a value of 5 causes
the first clip to start playing at frame 5 instead of frame
1.
Last Frame lastFrame Depending on the First Frame value, displays the frame
number of the end of the sequence after the slip.
take metadata meta_from_ disabled When disabled, the metadata is taken from whichever
from first clip first input is currently taking precedence.
This control was added to preserve the behavior of

previous versions of Nuke, where the metadata was
always taken from the first input, and is on by default
when a legacy script is loaded.
REFERENCE GUIDE
202
FrameBlend |
FrameBlend
The FrameBlend node interpolates frames by generating an additive composite of the frames that precede
and follow it, rather than creating mere copies between the existing frames. This method creates
“ghosting” around all fast moving features. The results may look odd when viewed as part of a still frame,
but will contribute to smoother motion during actual playback.
You should insert this node before the temporal effect you want to influence (for example, before a
Retime node).
Inputs and Controls

Type Name
Input unnamed The image sequence to blend.
mask An optional image to use as a mask. By default, the frame blend is

or set to none.

Value
FrameBlend Tab
channels channels all The frame blend is only applied to these channels.

Number of numframes 5 Sets the number of frames to blend together to
REFERENCE GUIDE
203
FrameBlend |

Value
frames create the effect:

• When Custom is disabled, blend together Number
of frames-1 previous frames and the current frame,
creating a trail-like effect on moving objects.
• When Custom is enabled, then the Number of
frames are evenly distributed across the specified
range.
Frame Range startframe -1 When Custom is enabled, the specified frame range is
blended together producing the same result for every
endframe -1 output frame (unless you put an expression into the
frame numbers).
Custom userange disabled When enabled, use the Frame Range controls to
specify the blended frame range.
Input Range getinputrange N/A Click to automatically use the input sequence frames
as the blend FrameRange.
Foreground N/A disabled Enables the associated matte channel to the right.
matte Disabling this checkbox is the same as setting the
channel to none.
inputmask none Use this channel as a matte around a moving

foreground object so that only pixels with a value of
zero are blended. If you blend enough frames
together and the object moves enough, the object is
removed from the output to produce a clean
background plate.
output N/A disabled Enables the associated mask channel to the right.
Image count Disabling this checkbox is the same as setting the
to channel to none.
blendmask none When enabled, saves a floating point alpha image to a

channel you specify; the result indicates the number
of images that contributed to each pixel of the matte.
To normalize the alpha, divide the number 1 by the
REFERENCE GUIDE
204
FrameHold |

Value
number of frames averaged, and then multiply the

alpha channel by this result. You can also use the
inverse of this matte for additional degraining.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the blend is

invert invert_mask disabled Inverts the use of the mask channel so that the blend
the mask.

mask.
mix mix 1 Mixes between the original image at 0 and the full
frame blend at 1.
FrameHold
Lets you either:
• pick one frame and use that frame at every frame of the input clip, or
• use every certain number of frames of the input clip (for example, every fifth frame).
REFERENCE GUIDE
205
FrameRange |
Inputs and Controls

Type Name
Input unnamed An image sequence with a single frame or every certain number of
frames you want to use, disregarding all other frames.

(Scripting) Value
FrameHold Tab
first frame first_frame 0 The first frame to use. If increment is set to 0, first
frame is used at every frame of the input clip.
increment increment 0 The increment to use when rendering every certain

number of frames from the input clip. For example, if
this is set to 5 and first frame is set to 1, Nuke only uses
frames 1, 6, 11, 16, and so on from the input.
FrameRange
Lets you set a frame range for a clip. This controls which frames are sent to the flipbook and displayed in
the Viewer when the frame range source dropdown menu is set to Input.
After setting the frame range for a clip, you may want to adjust the script length for the new output range.
Press S on the Node Graph to open the Project Settings, and enter frame range values that match the
output range you specified.
To make an edit, you can first use this node to cut portions out of input sequences and then append the
results together using the AppendClip node.
REFERENCE GUIDE
206
Kronos |
Tip: Using FrameRange, you can also set the frame range for a clip directly in the Dope Sheet.
Inputs and Controls

Type Name
Input unnamed The image sequence to cut.

(Scripting) Value
FrameRange Tab
frame range first_frame 1 Enter the appropriate in and out point values.
For example, if your original clip is 50 frames but you

last_frame 100 want to use only the last 25 frames in your composite,
you would enter 25 as the first frame and leave the last
frame at 50.
reset N/A N/A Resets the frame range to its original in and out points.
Kronos
Kronos (NukeX and Nuke Studio only) is a retimer, designed to slow down or speed up footage. It works
by calculating the motion in the sequence in order to generate motion vectors, which in turn, enables you
to generate an output image at any point in time throughout the sequence by interpolating along the
direction of the motion. Kronos also contains a number of controls to allow you to trade off render time
versus accuracy of vectors.
See also OFlow, TimeWarp, and Retime.
REFERENCE GUIDE
207
Kronos |
Inputs and Controls

Type Name
Input FgVecs If the motion in your input sequence has been estimated before (for
example, using VectorGenerator), you can supply one or more vector
BgVecs sequences to Kronos to save processing time.
Note: The BgVecs input appears as an arrowhead on the

side of the node.
MoSrc An alternate Motion Source input, for example, if your Source

sequence is very noisy and interfering with the motion estimation.
Matte An optional matte of the foreground, which may improve the motion
estimation by reducing the dragging of pixels that can occur
between foreground and background objects.
Source The sequence to retime.

Value
Kronos Tab
Local GPU gpuName N/A Displays the GPU used for rendering when Use GPU if
available is enabled. Local GPU displays Not
available when:
• Use CPU is selected as the default blink device in
the Preferences.
processing on the selected GPU, such as when there
is not enough free memory available on the GPU.

REFERENCE GUIDE
208
Kronos |

Value
alternative from the default blink device dropdown.
Note: Selecting a different GPU requires

you to restart Nuke before the change takes
effect.
Use GPU if useGPUIfAvailable enabled When enabled, rendering occurs on the Local GPU
available specified, if available, rather than the CPU.
Note: Enabling this option with no local

GPU allows the script to run on the GPU
whenever the script is opened on a machine
You should also select this if you wish to
render from the command line with the --
gpu option.
See Nuke's Online Help Installation pages for more

information on the GPUs supported by your
operating system.
Input Range input.first N/A The first frame of input to use.
When you first create the node, this is automatically

set to the first frame of the Source clip. After that, it
is only updated if you click Reset.
input.last N/A The last frame of input to use.

set to the last frame of the Source clip. After that, it is
only updated if you click Reset.
Reset resetInputRange N/A Resets Input Range to the frame range of the current
Source input. This can be useful if the clip in the
Source input has changed.
REFERENCE GUIDE
209
Kronos |

Value
Channels retimedChannels all Sets the channels affected by the retime.
Method interpolation Motion Sets the interpolation algorithm to use:

• Frame - the nearest original frame is displayed.
• Blend - a mix between two frames is used for the
in-between frame. This is quick to render and is
useful when tweaking the timing on a curve before
setting the method to motion.
• Motion - vector interpolation is used to calculate
the in-between frame.
Timing timing2 Output Sets how to control the new timing of the clip:
Speed • Output/Input Speed - describes the retiming in
terms of overall output duration. For example,
double speed halves the duration of the clip and
half speed doubles the duration of the clip.
• Frame - describes the retiming in relative terms, for
example, ’at frame 100 in the output clip, display
frame 50 of the source clip‘. You’ll need to set at
least 2 key frames for this to retime the clip.
Output Speed timingOutputSpeed 0.5 This control is only active if Timing is set to Output
Speed.
Values below 1 slow down the clip and vice versa. For
example, to slow down the clip by a factor of two
(half speed), set this value to 0.5.
Note: Retimes can be difficult to control

using Output Speed, because you're
retiming output frames. Altering a keyframe
alters the relative position of all later
frames. However, Output Speed keyframes
do allow you to use negative speed values,
unlike Input Speed retimes.
REFERENCE GUIDE
210
Kronos |

Value
Input Speed timingInputSpeed 0.5 This control is only active if Timing is set to Input
Speed.
example, to slow down the clip by a factor of two
(half speed), set this value to 0.5.
Note: Retimes can be easier to control

using Input Speed, because keyframes are
set against the source frames, rather than
the output frames. However, Input Speed
keyframes do not allow you to use negative
speed values, unlike Output Speed retimes.
Frame timingFrame2 1 This control is active only if Timing is set to Frame.
Use this to specify the source frame at the current

frame in the timeline. For example, to slow down a 50
frame clip by half, key the Frame to 1 at frame 1 and
the Frame to 50 at frame 100.
Motion motionEstimation Dependen Sets the method of calculating motion estimation

t on script vectors:
• Local - uses local block matching to estimate
motion vectors. This method is faster to process,
but can lead to artifacts in the output. This option is
for legacy compatibility.
• Regularized - uses semi-global motion estimation
to produce more consistent vectors between
regions.
REFERENCE GUIDE
211
Kronos |

Value
Note: Scripts loaded from previous

versions of Nuke default to Local motion
estimation for backward compatibility.
Adding a new Kronos node to the Node
Graph defaults the Method to Regularized
motion estimation.
Vector Detail vectorDetail 0.3 Adjust this to vary the density of the vector field. A
value of 1 generates a vector at each pixel, whereas a
value of 0.5 generates a vector at every other pixel.
Higher values pick up finer movement, but also

increase processing time.
Strength strength 1.5 This control is only active if Method is set to

Regularized.
Sets the strength in matching pixels between frames.

Higher values allow you to accurately match similar
pixels in one image to another, concentrating on
detail matching even if the resulting motion field is
jagged. Lower values may miss local detail, but are
less likely to provide you with the odd spurious
vector, producing smoother results.
Note: The default value should work well

for most sequences.
Smoothness smoothness 0.5 This control is only active if Method is set to Local.
A high smoothness can miss lots of local detail, but is

vector, whereas a low smoothness concentrates on
detail matching, even if the resulting field is jagged.
REFERENCE GUIDE
212
Kronos |

Value

for most sequences.
Resampling resampleType Bilinear Sets the type of resampling applied when retiming:
• Bilinear - the default filter. Faster to process, but
can produce poor results at higher zoom levels. You
can use Bilinear to preview a retime before using
one of the other resampling types to produce your
output.
• Lanczos4 and Lanczos6 - these filters are good for
scaling down, and provide some image sharpening,
but take longer to process.
Shutter
Shutter shutterSamples 1 Sets the number of in-between images used to create

Samples an output image during the shutter time. Increase this
value for smoother motion blur, but note that it takes
much longer to render.
Shutter Time shutterTime 0 Sets the equivalent shutter time of the retimed
sequence. For example, a shutter time of 0.5 is
equivalent to a 180 degree mechanical shutter, so at
24 frames per second the exposure time will be
1/48th of a second.
Imagine a gray rectangle moving left to right

horizontally across the screen. The figures below
show how Shutter Time affects the retimed
rectangle.
REFERENCE GUIDE
213
Kronos |

Value
Automatic autoShutterTime disabled When enabled, Kronos automatically calculates the

Shutter Time shutter time throughout the sequence.
Note that this only produces motion blur when the

retimed speed is greater than the original speed.
Output output Result Sets the final output display for the retimed image.
Selecting anything other than Result is only useful
when a Matte Channel is used.
• Result - displays the retimed Source image.
• Matte - displays the retimed Matte input.
• Foreground - displays the retimed foreground. The
background regions outside the matte input may
show garbage.
• Background - displays the retimed background.
The foreground regions inside the matte input may
show garbage.
Matte matteChannel None Where to get the (optional) foreground mask to use
Channel for motion estimation:
• None - do not use a matte.
• Source Alpha - use the alpha of the Source input.
• Source Inverted Alpha - use the inverted alpha of
the Source input.
• Matte Luminance - use the luminance of the Matte
input.
• Matte Inverted Luminance - use the inverted
luminance of the Matte input.
REFERENCE GUIDE
214
Kronos |

Value
• Matte Alpha - use the alpha of the Matte input.

• Matte Inverted Alpha - use the inverted alpha of
the Matte input.
Advanced
Flicker flickerCompensatio disabled When enabled, Kronos takes into account variations in
Compensatio n luminance and overall flickering, which can cause
n problems with your output.
Examples of variable luminance include highlights on

metal surfaces, like vehicle bodies, or bodies of water
within a layer that reflect light in unpredictable ways.
Note: Using Flicker Compensation

increases rendering time.
Legacy Mode legacymodeNuke9 Dependen Scripts loaded from previous versions of Nuke
t on Nuke default to Legacy Mode for backward compatibility.
version Adding a new Kronos node to the Node Graph hides
this control.
When enabled, Kronos reverts to the legacy method

for calculating Output Speed using the Local motion
estimation method.
Advanced > Tolerances
REFERENCE GUIDE
215
NoTimeBlur |

Value
Weight Red weightRed 0.3 For efficiency, much of the motion estimation is done
on luminance only - that is, using monochrome
images. The tolerances allow you to tune the weight
Weight Green weightGreen 0.6 of each color channel when calculating the image
luminance. These parameters rarely need tuning.
However, you may, for example, wish to increase the
Weight Blue weightBlue 0.1 red weighting Weight Red to allow the algorithm to
concentrate on getting the motion of a primarily red
object correct, at the cost of the rest of the items in a
shot.
Vector vectorSpacing 20 Sets the spacing between motion vectors displayed

Spacing on the Viewer when Overlay Vectors is enabled. The
default value of 20 means every 20th vector is drawn.
Overlay showVectors disabled When enabled, the motion vectors are displayed in
Vectors the Viewer. Forward motion vectors are drawn in red,
and backward motion vectors in blue.
Note: Motion vectors displayed in the

Viewer are added to your output if you
don’t turn off the overlay before rendering.
NoTimeBlur
NoTimeBlur prevents nodes downstream from asking for fractional or multiple frames. You can use this
node to avoid unwanted computation. For example, you may find that using TimeBlur on all the upstream
nodes in your composition can be unnecessary and very time consuming. In these cases, you can use
NoTimeBlur node to limit the number of nodes to which you’re applying TimeBlur. Just insert the
NoTimeBlur node in your node tree above the TimeBlur and any nodes you want the TimeBlur node to
process.
You can also use the NoTimeBlur node to force animated objects to align with input images.
REFERENCE GUIDE
216
OFlow |
Inputs and Controls

Type Name
Input unnamed NoTimeBlur can be inserted between any two existing nodes to limit
requests for fractional or multiple frames downstream.

(Scripting) Value
NoTimeBlur Tab
rounding rounding rint Sets the rounding type to use when blocking fractional
or multiple frames:
• rint - rounds the frame(s) to the nearest integer value.
• floor - rounds the frame(s) down to the nearest integer
value.
• ceil - rounds the frame(s) up to the nearest integer
value.
• none - no rounding is applied to the requested frame
(s).
use single single enabled When enabled, only the first frame requested is used.
frame
OFlow
Lets you generate high-quality retiming operations, such as slowing down or speeding up the input
footage. OFlow analyzes the movement of all pixels in the frames, generates motion vectors based on the
analysis, and then renders new “in-between” images by interpolating along the direction of the motion.
You can also use OFlow to add motion blur or enhance the existing motion blur in the image.
See also Retime and Kronos.
REFERENCE GUIDE
217
OFlow |
Inputs and Controls

Type Name
Input Source The sequence to retime.

Value
OFlow Tab
available is enabled. Local GPU displays Not available
when:
the Preferences.
• it was not possible to create a context for processing
on the selected GPU, such as when there is not
enough free memory available on the GPU.


effect.
REFERENCE GUIDE
218
OFlow |

Value

gpu option.

information on the GPUs supported by your operating
system.
Input Range input.first N/A The first frame of input to use.

set to the first frame of the Source clip. After that, it is
input.last N/A The last frame of input to use.

set to the last frame of the Source clip. After that, it is
Reset resetInputRange N/A Resets Input Range to the frame range of the current
Source input. This can be useful if the clip in the
Source input has changed.
Channels retimedChannels all Sets the channels affected by the retime.
Method method Motion Sets the interpolation algorithm:

• Frame - the nearest original frame is displayed.
• Blend - a mix between two frames is used for the in-
between frame. This is quick to render and is useful
when tweaking the timing on a curve before setting
the method to motion.
• Motion - vector interpolation is used to calculate
the in-between frame.
REFERENCE GUIDE
219
OFlow |

Value
Timing timing2 Output Sets how to control the new timing of the clip:
Speed • Output/Input Speed - describes the retiming in
terms of overall output duration. For example,
double speed halves the duration of the clip and half
speed doubles the duration of the clip.
• Frame - describes the retiming in relative terms, for
example, ’at frame 100 in the output clip, display
frame 50 of the source clip‘. You’ll need to set at
least 2 key frames for this to retime the clip.
Output Speed timingOutputSpeed 0.5 This control is only active if Timing is set to Output
Speed.
example, to slow down the clip by a factor of two (half
speed), set this value to 0.5.
Note: Retimes can be difficult to control

using Output Speed, because you're
retiming output frames. Altering a keyframe
alters the relative position of all later frames.
However, Output Speed keyframes do allow
you to use negative speed values, unlike
Input Speed retimes.
Input Speed timingInputSpeed 0.5 This control is only active if Timing is set to Input
Speed.
example, to slow down the clip by a factor of two (half
speed), set this value to 0.5.
REFERENCE GUIDE
220
OFlow |

Value
Note: Retimes can be easier to control

using Input Speed, because keyframes are
set against the source frames, rather than the
output frames. However, Input Speed
keyframes do not allow you to use negative
speed values, unlike Output Speed retimes.
Frame timingFrame2 1 This control is active only if Timing is set to Frame.
Use this to specify the source frame at the current

frame in the timeline. For example, to slow down a 50
frame clip by half, key the Frame to 1 at frame 1 and
the Frame to 50 at frame 100.
Vector Detail vectorDetail 0.2 Adjust this to vary the density of the vector field. A
value of 1 generates a vector at each pixel, whereas a
value of 0.5 generates a vector at every other pixel.
Higher values pick up finer movement, but also

increase processing time.
Smoothness smoothness 0.5 A high smoothness can miss lots of local detail, but is
vector, whereas a low smoothness concentrates on
detail matching, even if the resulting field is jagged.

for most sequences.
• Bilinear - the default filter. Faster to process, but can
produce poor results at higher zoom levels. You can
use Bilinear to preview a retime before using one of
the other resampling types to produce your output.
REFERENCE GUIDE
221
OFlow |

Value
Shutter

Shutter Time shutterTime 0 Sets the equivalent shutter time of the retimed
1/48th of a second.

horizontally across the screen. The figures below show
how Shutter Time affects the retimed rectangle.
Automatic autoShutterTime disabled When enabled, OFlow automatically calculates the

Shutter Time shutter time throughout the sequence.
Note that this only produces motion blur when the
REFERENCE GUIDE
222
OFlow |

Value
retimed speed is greater than the original speed.
Advanced
Flicker flickerCompensation disabled When enabled, OFlow takes into account variations in
Compensation luminance and overall flickering, which can cause
problems with your output.
Examples of variable luminance include highlights on

metal surfaces, like vehicle bodies, or bodies of water
within a layer that reflect light in unpredictable ways.

Weight Red weightRed 0.3 For efficiency, much of the motion estimation is done
on luminance only - that is, using monochrome
images. The tolerances allow you to tune the weight
Weight Green weightGreen 0.6 of each color channel when calculating the image
luminance. These parameters rarely need tuning.
However, you may, for example, wish to increase the
Weight Blue weightBlue 0.1 red weighting Weight Red to allow the algorithm to
concentrate on getting the motion of a primarily red
object correct, at the cost of the rest of the items in a
shot.
Vector vectorSpacing 20 Sets the spacing between motion vectors displayed

Spacing on the Viewer when Overlay Vectors is enabled. The
default value of 20 means every 20th vector is drawn.
Overlay showVectors disabled When enabled, the motion vectors are displayed in
Vectors the Viewer. Forward motion vectors are drawn in red,
and backward motion vectors in blue.
REFERENCE GUIDE
223
Remove 3:2 pulldown |

Value
Note: Motion vectors displayed in the

Viewer are added to your output if you don’t
turn off the overlay before rendering.
Remove 3:2 pulldown

Removes 3:2 pulldown from the input footage.
If you’re using footage that was originally film but was converted to video, you may want to use the
Remove 3:2 pulldown node when importing the footage to Nuke. This allows you to convert the footage
to its original state (24 fps non-interlaced film) before adding effects to it. If you later want to render the
footage back to video again, you can use the Add 3:2 pulldown node to reintroduce 3:2 pulldown.
Inputs and Controls

Type Name
Input 1 The image sequence to remove 3:2 pulldown from. For example,
footage that was originally film but was converted to video.

(Scripting) Value
User Tab
phase phase 0 The 3:2 pulldown process results in:

• three whole frames that contain two fields from the
same film frame, and
• two split-field frames that contain fields from two
different film frames. The two split-field frames are
always adjacent to each other.
REFERENCE GUIDE
224
Retime |

(Scripting) Value
Phase is the point at which the two split-field frames

occur within the first five frames of the footage, relative
to your first input frame. The value here should match
the phase used when 3:2 pulldown was added.
invert field field_dom disabled Video frames are usually divided into two fields. Each
dominance field contains only half of the image information, drawn
as horizontal scan lines:
• The first field of a frame contains every other scan line.
• The second field of a frame contains the remaining
scan lines.
When both fields are displayed in rapid sequence, they

appear to produce a normal, complete frame.
Invert field dominance determines which field is

displayed first:
• When enabled, the odd lines come first.
• When disabled, the even lines come first.
Retime
Lets you slow down, speed up, or even reverse select frames in a clip without necessarily altering its
overall length.
See also OFlow, TimeWarp, and Kronos.
REFERENCE GUIDE
225
Retime |
Inputs and Controls

Type Name
Input unnamed The sequence to retime.
Control Knob Default Function

(UI) (Scripting) Value
Retime Tab
input range input.first 1 When enabled, sets the first frame of the input sequence
to use for the retime.
input.first_ disabled Enable this control to lock the retime to a specific first
lock frame.
Note: If both output range fields are locked,

the retime is calculated to make the frame ranges
match. Otherwise the output frames move to
accommodate the speed.
input.last Dependent When enabled, sets the last frame of the input sequence to
on input clip use for the retime.
input.last_ disabled Enable this control to lock the retime to a specific last
lock frame.
Note: If both output range fields are locked,

the retime is calculated to make the frame ranges
match. Otherwise the output frames move to
accommodate the speed.
reverse reverse disabled When enabled, the input frames run backwards.
output output.first 1 When enabled, sets the first frame of the clip length after
REFERENCE GUIDE
226
Retime |

range the retime.
output.first_ disabled Enable this control to lock the output to a specific first
lock frame.
output.last Dependent When enabled, sets the last frame of the clip length after
on input clip the retime.
output.last_ disabled Enable this control to lock the output to a specific last
lock frame.
speed speed 1 Sets the retime speed enabling Nuke to calculate the
output range for you, rather than using the output range
controls. Values higher than 1 increase playback speed;
values less than 1 decrease playback speed.
before before hold Sets the behavior of frames before the output.first frame
(examples refer to a 20 frame sequence with an
output.first value of 5):
• continue - the first frame in the sequence is held until
the output.first frame is reached.
• loop - substitutes an equal number of frames, effectively
creating a clip loop.
Example: 17, 18, 19, 20, 1, 2, 3, 4, etc.

• bounce - substitutes a reversed equal number of frames,
creating a clip bounce.
Example: 5, 4, 3, 2, 1, 2, 3, 4, etc.
• hold - the first frame in the sequence is held until the
output.first frame is reached.
Example: 1, 1, 1, 1, 1, 2, 3, 4, etc.
• black - frames are black until the output.first frame is
reached.
after after hold Sets the behavior of frames after the output.last frame
(examples refer to a 20 frame sequence with an
REFERENCE GUIDE
227
Retime |

output.last value of 5):

• continue - the output.last frame in the sequence is held
until the end of the sequence is reached.
• loop - substitutes an equal number of frames, effectively
creating a clip loop.
Example: 16, 17, 18, 19, 20, 1, 2, 3, 4, etc.

• bounce - substitutes a reversed equal number of frames,
creating a clip bounce.
Example: 16, 17, 18, 19, 20, 19, 18, 17, etc.
• hold - the output.last frame in the sequence is held until
the end of the sequence is reached.
Example: 16, 17, 18, 19, 20, 20, 20, 20, etc.
• black - frames are black from output.last until the end of
the sequence is reached.
filter filter box Sets the type of filtering to apply to the retime:
• none - passes fractional frame numbers to the input,
which is useful if the input can calculate its own frame
interpolation.
• nearest - rounds the center of the range to nearest
integer frame.
• box - uses a weighted average of several frames together
to cover the output range.
shutter shutter 1 Controls frame-blending by manipulating the shutter value.

Lower shutter values generate less frame-blending.
TimeWarp Tab
warp warp N/A To warp the input clip, edit this curve are as follows:
• To slow down motion, decrease the slope of the curve.
• To speed up motion, increase the slope of the curve.
• To reverse motion, create a downward sloping portion
REFERENCE GUIDE
228
SmartVector |

on the curve (a dip, in other words).
You can:
• Ctrl/Cmd+Alt+click to insert keyframe knots on the
curve.
• Ctrl/Cmd+drag to reposition keyframe knots.
• Ctrl/Cmd+drag to rotate a keyframe knot control
handles.
Note: The curve must pass through 0,0 and 1,1

so that the in and out points work, respectively.
reset N/A N/A Click to reset the curve to the default values.
SmartVector
The SmartVector node writes motion vectors to the .exr format, which are then used to drive the
VectorDistort or VectorCornerPin nodes as part of the Smart Vector toolset.
See VectorDistort and VectorCornerPin.
Inputs and Controls

Type Name
Inputs Matte An optional matte of the foreground, which may improve the motion
Source The sequence from which to generate motion vectors.
REFERENCE GUIDE
229
SmartVector |

Value
SmartVector Tab
Local GPU gpuName N/A Displays the GPU used for rendering when Use
GPU if available is enabled. Local GPU displays
Not available when:
• Use CPU is selected as the default blink device
in the Preferences.
processing on the selected GPU, such as when
there is not enough free memory available on
the GPU.

dropdown.

you to restart Nuke before the change
takes effect.

whenever the script is opened on a
machine that does have a GPU available.
render from the command line with the -
-gpu option.
REFERENCE GUIDE
230
SmartVector |

Value

operating system.
Vector Detail vectorDetailReg 0.3 Sets the vector quality. The default value of 0.3 is
sufficient for sequences with low detail and
movement, but you may want to increase the
detail to improve the vector quality in some cases.
Note: High detail vectors take longer to

render, but can improve the results you
get from the VectorDistort node.
Strength strengthReg 1.5 Sets the strength in matching pixels between

frames. Higher values allow you to accurately
match similar pixels in one frame to another,
concentrating on detail matching even if the
resulting motion field is jagged. Lower values may
miss local detail, but are less likely to provide you
with the odd spurious vector, producing smoother
results.
Note: The default value works well for

most sequences.
Matte Channel matteChannel None Where to get the optional foreground mask to use
when estimating motion:
• Source Alpha - use the alpha of the Source
input.
• Source Inverted Alpha - use the inverted alpha
of the Source input.
• Matte Luminance - use the luminance of the
Matte input.
REFERENCE GUIDE
231
SmartVector |

Value

• Matte Inverted Alpha - use the inverted alpha
of the Matte input.
Output output Background When Matte Channel is set to anything other than
None, the Output control determines whether
Foreground or Background vectors are passed
down stream or written to file.
Inpaint Matte inpainting disabled When enabled, any areas where vectors are absent
Region are infilled using the nearest available vectors.
Note: This control is only available when

Matte Channel is set to one of the matte
options, such as Matte Alpha.
Matte Dilation matteDilation 50 Controls the amount of dilation applied to the

matte before inpainting the matte region.

Inpaint Matte Region is enabled.
Export
Export Write export_write N/A Click to automatically create a Write node with
suitable .exr settings to contain the vector data.
Advanced
Flicker flickerCompensation disabled When enabled, SmartVector takes into account

Compensation variations in luminance and overall flickering,
which can cause problems with your output.
Examples of variable luminance include highlights
on metal surfaces, like vehicle bodies, or bodies of
water within a layer that reflect light in
unpredictable ways.
REFERENCE GUIDE
232
TemporalMedian |

Value

Tolerances
Weight Red weightRed 0.3 For efficiency, much of the motion estimation is
done on luminance only - that is, using
monochrome images. The tolerances allow you to
tune the weight of each color channel when
calculating the image luminance. These parameters
Weight Green weightGreen 0.6 rarely need tuning. However, you may, for
example, wish to increase the Weight Red to
allow the algorithm to concentrate on getting the
motion of a primarily red object correct, at the
Weight Blue weightBlue 0.1 cost of the rest of the items in a shot.
TemporalMedian
This node applies a temporal median filter to the input clip. It outputs each pixel by calculating the
median of the current frame, the frame before, and the frame after. This can be useful for removing grain,
for example.
Inputs and Controls

Type Name
Input unnamed The image sequence to apply a temporal median filter to (typically,
the image sequence to degrain).
REFERENCE GUIDE
233
TemporalMedian |

Type Name
mask An optional image to use as a mask. By default, the temporal median

filter effect is limited to the non-black areas of the mask.
or set to none.

Value
TemporalMedian Tab
channels channels all The temporal median filter effect is only applied to
these channels.

core core 0.05 If the differences between pixel values on the current,
previous, and next frame are greater than this value,
those pixels are left unchanged. This is because higher
differences between frames probably indicate
something other than film grain.
mask N/A disabled Enable the associated mask channel to the right.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the temporal

median filter effect is limited to the non-black areas
of this channel.
temporal median filter effect is limited to the non-
REFERENCE GUIDE
234
TimeBlur |

Value
white areas of the mask.
the mask.

mask.
effect at 1.
TimeBlur
When a fast moving subject is recorded on film or video, its edges appear to smear as a result of the
object's movement while the shutter is open. The longer the shutter remains open at each frame interval,
the more obvious this effect. TimeBlur simulates this phenomenon by sampling its input at divisions times
over shutter frames starting at the current frame plus offset.
Time blur is commonly applied to garbage masks that are tracked to a fast moving feature. The time blur
averages the incoming mask image over the shutter period, to better match the motion blur in the original
image and creating a more convincing integration.
See also NoTimeBlur.
Inputs and Controls

Type Name
Input unnamed The image sequence to which you want to apply blur.
REFERENCE GUIDE
235
TimeBlur |

Value
TimeBlur Tab
divisions divisions 10 The number of times you want to sample the input
over the shutter time. For images with fast-moving
content, higher values are necessary to eliminate
"steppiness" or banding in the output.
shutter shutter 0.5 Sets the number of frames the shutter stays open
when motion blurring. For example, a value of 0.5
produces more blur, and decreasing the value less.
shutter offset shutteroffset start Controls how the shutter behaves with respect to the
current frame value:
• centred - center the shutter around the current
frame 30 to 31.
frame 29 to 30.
frame. To open the shutter before the current
frame, enter a negative value. For example, a value
of - 0.5 would open the shutter half a frame before
the current frame.
shuttercustomoffset 0 If the shutter offset control is set to custom, this

field is used to set the time that the shutter opens
by adding it to the current frame. Values are in
frames, so -0.5 would open the shutter half a frame
before the current frame.
REFERENCE GUIDE
236
TimeClip |
TimeClip
Just like TimeOffset, the TimeClip node lets you move the clip forwards or backwards in time and reverse
the order of frames in the clip. In addition to this basic functionality, you can slip a clip, set the frame
range for the clip, set what happens to frames outside of this frame range, fade the clip to or from black,
and set expressions to adjust the node’s behavior.
Tip: Using TimeClip, you can also offset, trim, and slip clips directly in the Dope Sheet. See
Nuke's online help for more information.
See also TimeOffset.
Inputs and Controls

Type Name
Input unnamed The image sequence to slip.

(Scripting) Value
TimeClip Tab
Fade In fadeIn 5 The number of frames to fade from black.
Fade Out fadeOut 5 The number of frames to fade to black.
frame range first Dependent The first frame of the range.

on Project
Settings You can also adjust this by going to the Dope Sheet tab
and dragging the head of the clip to a new location.
before hold Sets the behavior of frames before the first frame
(examples refer to a 20-frame sequence with a first
value of 5):
REFERENCE GUIDE
237
TimeClip |

(Scripting) Value
• hold - the first frame in the sequence is held until the

first frame is reached. Example: 1, 1, 1, 1, 1, 2, 3, 4, etc.
• loop - substitutes an equal number of frames,
effectively creating a clip loop. Example: 17, 18, 19, 20,
1, 2, 3, 4, etc.
• bounce - substitutes a reversed equal number of
frames, creating a clip bounce. Example: 5, 4, 3, 2, 1, 2,
3, 4, etc.
• black - frames are black until the first frame is reached.
last Dependent The last frame of range.

on Project
Settings You can also adjust this by going to the Dope Sheet tab
and dragging the tail of the clip to a new location.
after hold Sets the behavior of frames after the last frame
(examples refer to a 20 frame sequence with a last value
of 5):
• hold - the last frame in the sequence is held until the
end of the sequence is reached. Example: 16, 17, 18, 19,
20, 20, 20, 20, etc.
• loop - substitutes an equal number of frames,
effectively creating a clip loop. Example: 16, 17, 18, 19,
20, 1, 2, 3, 4, etc.
• bounce - substitutes a reversed equal number of
frames, creating a clip bounce. Example: 16, 17, 18, 19,
20, 19, 18, 17, etc.
• black - frames are black from last until the end of the
sequence is reached.

• expression - Lets you enter an expression in the field
on the right. The expression changes the relation
between the current frame and the frame read in. For
example, if your clip begins from image.0500.rgb and
you want to place this first frame at frame 1 rather than
REFERENCE GUIDE
238
TimeClip |

(Scripting) Value
frame 500, you can use the expression frame+499. This

way, 499 frames are added to the current frame to get
the number of the frame that’s read in. At frame 1,
image.0500.rgb is read in; at frame 2, image.0501.rgb is
read in; and so on. Another example of an expression is
frame*2. This expression multiplies the current frame
by two to get the number of the frame that’s read in.
This way, only every other frame in the clip is used. At
image.0004.rgb is read in; at frame 3, image.0006.rgb is
read in; and so on.
• start at - Lets you enter a start frame number in the
field on the right. This specifies the frame where the
first frame in the sequence is read in. In other words, all
frames are offset so that the clip starts at the specified
frame. For example, if your sequence begins from
image.0500.rgb and you enter 1 in the field,
image0500.rgb is read in at frame 1. Similarly, if you
enter 100 in the field, image0500.rgb is read in at
frame 100.
• offset - Lets you enter a constant offset in the field on
the right. This constant value is added to the current
frame to get the number of the frame that’s read in.
For example, if your clip begins from image.0500.rgb
and you want to place this first frame at frame 1 rather
than frame 500, you can use 499 as the constant offset.
This way, 499 is added to the current frame to get the
frame that’s read in. At frame 1, image.0500.rgb is read
in; at frame 2, image.0501 is read in, and so on. You can
also use negative values as the constant offset. For
example, if you use the value -10, Nuke subtracts ten
from the current frame to get the frame that’s read in.
At frame 20, image.0010.rgb is read in; at frame 21,
original range origfirst Dependent The original first frame of the sequence.
REFERENCE GUIDE
239
TimeEcho |

(Scripting) Value
on Project
Settings
origlast Dependent The original last frame of the sequence.

on Project
Settings
reverse reverse disabled Reverse the clip within the specified frame range.
TimeEcho
Merges multiple frames from the input into a single frame to create an echo-like effect over time. This
allows you to create streaking and smearing effects, for example.
Inputs and Controls

Type Name
Input unnamed The image sequence with the frames to merge.

(Scripting) Value
TimeEcho Tab
TimeEcho mode max Sets how the frames are blended together:
Method • Plus - uses the sum of all frames. Note that this may
result in pixel values higher than 1.0.
• Max - uses the highest value from the frames.
• Average - averages the frames (uses the sum divided
by the number of frames). The result is darker than the
REFERENCE GUIDE
240
TimeOffset |

(Scripting) Value
original images.
Frames to framesbehind 1 Sets the number of frames to merge together, counting

look at backwards from the current frame (and including the
current frame).
For example, if the current frame is 10, and you set this
value to 4, TimeEcho merges frames 7, 8, 9, and 10.
Frames to framesfade 1 Allows you to fade out frames, so that the further back a
fade out frame is in time, the less effect it has on the final
images. Note that this is the total number of frames
over which the fade happens, with the last of these
frames remaining at maximum.
For example, if you set this to 5, the fade will gradually

happen over the first five frames used, with the fifth
frame remaining at maximum:
• the first image is reduced to 20% of its original
opacity
• the second image is reduced to 40% of its original
opacity
• the third image is reduced to 60% of its original
opacity
• the fourth image is reduced to 80% of its original
opacity
• the fifth image remains at 100% of its original opacity.
TimeOffset
REFERENCE GUIDE
241
TimeOffset |
Lets you offset a clip. Offsetting a clip refers to moving it backward or forward in time. There are any
number of reasons why you might want to do this (for example, to synchronize events in a background
and foreground clip).
After offsetting the clip, you may want to adjust the script length for the new output range. Press S on the
Node Graph to open the Project Settings, and enter frame range values that match the output range you
specified. If you don’t do this, Nuke fills the empty frames at the head or tail of the clip by holding the
first or last frame.
Tip: Using TimeOffset, you can also offset clips directly in the Dope Sheet. See Nuke's online
help for more information.
Inputs and Controls

Type Name
Input unnamed The image sequence to offset.

(Scripting) Value
TimeOffset Tab
time offset time_offset 0 The number of frames by which you want to offset the
(frames) clip.
Enter a negative value to subtract frames from the head

of the clip. Enter a positive value to add frames to the
head of the clip.
reverse input reverse_input disabled When enabled, TimeOffset inverts the clip (making the
last frame the first, and so on).
REFERENCE GUIDE
242
TimeWarp |
TimeWarp
Lets you slow down, speed up, or even reverse selected frames in a clip without necessarily altering its
overall length.
See also OFlow, Retime, and Kronos.
Inputs and Controls

Type Name
Input unnamed The sequence to warp.
Note: When you connect the TimeWarp node, frame 1 is

automatically keyframed as the first frame.

(Scripting) Value
TimeWarp Tab
input frame lookup 1 Sets which frame of the input sequence appears at the
current frame.
For example, you could set frame 10 of the input to

output at frame 5, effectively doubling the speed of the
sequence.
curve editor N/A N/A Click to open up the Curve Editor.
input length length 1 Sets the length of time to average together to produce
the output frame.
Click the set to df/dt button to automatically set the

inputlength to the correct formula based on the
derivative of the inputframe.
REFERENCE GUIDE
243
VectorGenerator |

(Scripting) Value
set to df/dt N/A N/A Click to automatically set the inputlength to the correct
formula based on the derivative of the inputframe.
filter filter box Sets the type of filtering to apply to the time warp:
• none - passes fractional frame numbers to the input,
which is useful if the input can calculate its own frame
interpolation.
• nearest - rounds the center of the range to nearest
integer frame.
• box - uses a weighted average of several frames
together to cover the output range.
VectorGenerator
VectorGenerator (NukeX and Nuke Studio only) produces images containing motion vector fields. In
general, once you have generated a sequence of motion vector fields that describe the motion in a
particular clip well, they will be suitable for use in any nodes which can take vector inputs. These include
Kronos and MotionBlur available in NukeX.
The output from VectorGenerator consists of two sets of motion vectors for each frame. These are stored
in the vector channels.
Inputs and Controls

Type Name
Input Matte An optional matte of the foreground, which may improve the motion
An optional matte of the foreground. This can be used to help the
REFERENCE GUIDE
244
VectorGenerator |

Type Name
motion estimation algorithm inside VectorGenerator understand

what is foreground and background in the image, so that the
dragging of pixels between overlapping objects can be reduced.
White areas of the matte are considered to be foreground, and black
areas background. Grey areas are used to attenuate between
foreground and background.
Source The sequence from which to generate motion vectors.

Value
VectorGenerator Tab
Not available when:
in the Preferences.
the GPU.

dropdown.

takes effect.
REFERENCE GUIDE
245
VectorGenerator |

Value

render from the command line with the -
-gpu option.

operating system.
Method motionEstimation Dependent Sets the method of calculating motion estimation

on script vectors:
• Local - uses local block matching to estimate
motion vectors. This method is faster to process,
but can lead to artifacts in the output.
• Regularized - uses semi-global motion
estimation to produce more consistent vectors
between regions.
Note: Scripts loaded from previous

versions of Nuke default to Local motion
estimation for backward compatibility.
Adding a new VectorGenerator node to
the Node Graph defaults the Method to
Regularized motion estimation.
Vector Detail vectorDetail 0.3 This determines the resolution of the vector field.
The larger vector detail is, the greater the
processing time, but the more detailed the vectors
should be. A value of 1.0 generates a vector at
each pixel. A value of 0.5 generates a vector at
every other pixel. For some sequences, a high
REFERENCE GUIDE
246
VectorGenerator |

Value
vector detail near 1.0 generates too much

unwanted local motion detail, and often a low
value is more appropriate.
Strength strength 1.5 This control is only active if Method is set to

Regularized.
Sets the strength in matching pixels between

frames. Higher values allow you to accurately
match similar pixels in one image to another,
concentrating on detail matching even if the
resulting motion field is jagged. Lower values may
miss local detail, but are less likely to provide you
with the odd spurious vector, producing smoother
results.
Note: The default value should work

well for most sequences.
Smoothness smoothness 0.5 This control is only active if Method is set to

Local.
A high smoothness can miss lots of local detail,

but is less likely to provide you with the odd
spurious vector, whereas a low smoothness
concentrates on detail matching, even if the
resulting field is jagged.
Note: The default value should work

well for most sequences.
Matte Channel matteChannel None Where to get the (optional) foreground mask to
use for motion estimation:
REFERENCE GUIDE
247
VectorGenerator |

Value
• Source Alpha - use the alpha of the Source

input.
• Mask Luminance - use the luminance of the
Matte input.
• Mask Inverted Luminance - use the inverted
• Mask Alpha - use the alpha of the Matte input.
• Mask Inverted Alpha - use the inverted alpha of
the Matte input.
Output output Foreground When a matte input is supplied, this determines

whether the motion vectors corresponding to the
background or the foreground regions are output.
• Foreground - the vectors for the foreground
motion are output.
• Background - the vectors for the background
motion are output.
Advanced
Flicker flickerCompensation disabled When enabled, VectorGenerator takes into account

Compensation variations in luminance and overall flickering,
which can cause problems with your output.
Examples of variable luminance include highlights

on metal surfaces, like vehicle bodies, or bodies of
water within a layer that reflect light in
unpredictable ways.

REFERENCE GUIDE
248
VectorToMotion |

Value
Weight Red weightRed 0.3 For efficiency, much of the motion estimation is
done on luminance only - that is, using
monochrome images. The tolerances allow you to
tune the weight of each color channel when
Weight Green weightGreen 0.6
calculating the image luminance. These parameters
rarely need tuning. However, you may, for
example, wish to increase the red weighting
Weight Blue weightBlue 0.1 Weight Red to allow the algorithm to concentrate
on getting the motion of a primarily red object
correct, at the cost of the rest of the items in a
shot.
VectorToMotion
VectorToMotion converts SmartVectors to motion, forward, and backward channels that can be used
with VectorBlur to create motion blur, without using a VectorGenerator.
It doesn't have any controls of its own, it's a conversion tool to convert vectors to motion channels. See
Nuke's online help for more information.
REFERENCE GUIDE
249
Channel Nodes
Channel nodes deal with the use of channels and layers within your composite. Typical channels are red,
green, blue, and alpha - however, there are many other pieces of useful data that can be stored in unique
channels. A layer is simply a collection of channels, for example rgba.
AddChannels
Adds channels to the input image. Channels that do not already exist are created and filled with the color
defined by the color knob.
Inputs and Controls

Type Name
Input unnamed The image to which the new channels are added.

(Scripting) Value
AddChannels Tab
channels channels none The channels to add.

channels.
and channels2 none Additional channels to add.

channels.
REFERENCE GUIDE
250
ChannelMerge |

(Scripting) Value

channels.

channels.
color color 0 The default color to fill in any new channels - existing
channels are not changed.
Note: The current version only accepts rgba

and gray shades for other channels.
ChannelMerge
The ChannelMerge node lets you merge together one channel from each input and save the result in the
selected output channel. All other channels are copied unchanged from the B input.
If no A input is connected, both channels to merge are taken from the B input.
By default, ChannelMerge combines the inputs’ alphas.
Inputs and Controls

Type Name
Input A The image that contains a channel to merge with a channel from
input B. This input is optional.
REFERENCE GUIDE
251
ChannelMerge |

Type Name
B The image that contains a channel to merge with a channel from

input A. If no A input is connected, both channels to merge are
taken from this input.
mask An optional image to use as a mask. By default, the merge is limited

to the non-black areas of the mask.

Value
ChannelMerge Tab
A channel N/A enabled Enables the associated A channel to the right.

channel to none.
A rgba.alpha The channel to merge from input A. If no A input is

connected, this channel is taken from input B.
operation operation union Sets how the pixel values from input A are calculated
with the pixel values from input B to create the new
pixel values that are output as the merged image:
• absminus abs(A-B) - how much the pixels differ.
• b if not a A?A:B - shows A wherever A exists;
otherwise shows B.
• divide A/B, 0 if A<0 and B<0 - divides the values but
stops two negative values from becoming a positive
number.
• in Ab - only shows the areas of image A that overlap
with the alpha of B.
REFERENCE GUIDE
252
ChannelMerge |

Value

• minus (A-B) - subtracts B from A.
• multiply (AB, A if A<0 and B<0) - multiplies A by B.
The result is always darker. Blending with black gives
• out A(1-b) - only shows the areas of image A that do
not overlap with the alpha of B.
• plus A+B - the sum of the two colors. Increases
brightness to lighten A and reflect B.
• stencil B(1-a) - this is the reverse of the out
operation. Only shows the areas of image B that do
not overlap with the alpha of A.
• union A+B-AB - shows both image A and B.
• xor A+B-2AB - shows both image A and B where the
images do not overlap.
B channel N/A enabled Enables the associated B channel to the right.

channel to none.
B rgba.alpha The channel to merge from input B.
output N/A enabled Enables the associated output channel to the right.
channel to none.
output rgba.alpha Render the merged image into this output channel.
Set BBox to bbox union How to output the bounding box:

• union - combine the two bounding boxes.
REFERENCE GUIDE
253
Copy |

Value
• B side - use the bounding box from input B. Any

data from the A input that’s outside this box is
clipped.
• A side - use the bounding box from input A. Any
data from the B input that’s outside this box is
clipped.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the merge is

invert invert_mask disabled Inverts the use of the mask channel so that the merge
the mask.

mask.
mix mix 1 Dissolves between the original B input at 0 and the

full merge effect at 1.
Copy
Replaces channels in the B input with channels from the A input. You can use a mask channel to indicate
which input is used for each pixel or whether both inputs should be mixed for some pixels.
Using the mix slider, you can use this node as a dissolve. You can also create this node by pressing K on
the Node Graph.
See also Shuffle.
REFERENCE GUIDE
254
Copy |
Inputs and Controls

Type Name
Input A The image that contains a channel to copy to input B.
B The image that contains a channel to replace with a channel from the
A input.
mask An optional image to use as a mask. By default, the A input is used

where the mask is white, the B input where the mask is black, and a
blend between the inputs where the mask is gray.

Value
Copy Tab
Copy channel N/A enabled Enables the associated channel to the right. Disabling
this checkbox is the same as setting the channel to
none.
from0 rgba.alpha A channel to copy from input A.
to N/A enabled Enables the associated channel to the right. Disabling

none.
to0 rgba.alpha The channel that the above Copy channel replaces in
input B.
none.
from1 rgba.alpha Another channel to copy from input A.
REFERENCE GUIDE
255
Copy |

Value

none.
input B.
none.

none.
input B.
none.

none.
input B.
Layer Copy channels none Copies all the selected channels from A to B, if they
exist in A. If they don’t exist in A, the channels from B
are used unchanged.
This allows you to easily copy multiple channels from

A to B. For example, setting this to all copies all the
channels that exist in A.
REFERENCE GUIDE
256
Copy |

Value


• B side - use the bounding box from input B. Any
data from the A input that’s outside this box is
clipped.
• A side - use the bounding box from input A. Any
data from the B input that’s outside this box is
clipped.
metadata metainput B Sets the source of the metadata passed down the
from node tree:
• B - only input B metadata is preserved.
• A - only input A metadata is preserved.
• All - both input A and B metadata is preserved.
range from rangeinput B Sets the output frame range passed down the node
tree:
• B - only frames from input B are output.
• A - only frames from input A are output.
• All - both input A and B frames are output.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the A input is

used where the mask is white, the B input where the
mask is black, and a blend between the inputs where
the mask is gray.
REFERENCE GUIDE
257
Remove |

Value
invert invert_mask disabled Inverts the use of the mask channel so that the B input
is used where the mask is white and the A input where
the mask is black.
the mask.

mask.
none.


here. This will simulate doing the copy before the
using one Copy node.

mix mix 1 Dissolves between the original B input at 0 and the

full copy effect at 1.
Remove
REFERENCE GUIDE
258
Remove |
Removes channels from the input clip. When you are done using a layer or a channel within a set, you may
wish, for the sake of clarity, to remove it so that it is no longer passed to downstream nodes. Note that
leaving channels in the stream will not itself cause them to be computed; only channels required are
computed.
Removing layers and or channels does not free up space for the creation of new channels and layers. Once
you create a channel, it permanently consumes one of the script’s 1023 available channel name slots. You
are free, however, to rename channels and/or assign them new outputs.
Inputs and Controls

Type Name
Input unnamed The image sequence with channels to remove.

(Scripting) Value
Remove Tab
operation operation remove Sets how channels are removed:

• remove - removes the channels you select below.
• keep - keeps the channels you select below and
removes all other channels.
channels channels all The channels to remove (or keep if you’ve set operation
to keep).

channels.
and channels2 none Additional channels to remove (or keep if you’ve set
operation to keep).

channels.
REFERENCE GUIDE
259
Shuffle |

(Scripting) Value
operation to keep).

channels.
operation to keep).

channels.
Shuffle
Note: Nuke 12.1 introduced a new and improved Shuffle2 node, which makes Shuffle and
ShuffleCopy obsolete. Shuffle and ShuffleCopy are still available for backward compatibility if
you press X in the Node Graph and enter Shuffle or ShuffleCopy.
Shuffle lets you:

• rearrange up to 8 channels from a single image (B input). For example, you can use it to swap rgba.red
for rgba.green, and vice versa,
• rearrange channels between two separate nodes (A and B input), like a foreground and background
branch,
• replace a channel with black (removing the alpha channel, for example) or with white (making the alpha
solid, for example),
• create new channels.
To simply copy a channel from one data stream into another, see Copy.
REFERENCE GUIDE
260
Shuffle |
Inputs and Controls

Type Name
Input B The primary image sequence with the channels to rearrange.
A The secondary image sequence with the channels to rearrange.

(Scripting) Value
Shuffle Tab
Input Layer
In fromInput1 B The incoming layer(s) that contains the channel(s) to

rearrange.
in1 rgba
You can also select new here to create new channels.
In fromInput2 B Additional incoming channels to rearrange. If you only

want to rearrange channels from one layer, you don’t
need these controls.
in2 none
Output Layer
N/A out1 rgba The rearranged channels are output to this layer.
N/A out2 none Additional rearranged channels are output to this layer. If
you only want to rearrange channels from one layer, you
don’t need these controls.
REFERENCE GUIDE
261
ShuffleCopy |
ShuffleCopy
Note: Nuke 12.1 introduced a new and improved Shuffle2 node, which makes Shuffle and
ShuffleCopy obsolete. Shuffle and ShuffleCopy are still available for backward compatibility if
you press X in the Node Graph and enter Shuffle or ShuffleCopy.
ShuffleCopy lets you:

• rearrange up to 8 channels from two images (two inputs). For example, you can use it to combine two
separate passes (such as the beauty pass and the reflection pass) into the same data stream,
• replace the alpha with black (removing the alpha channel) or with white (making the alpha solid),
• create new channels.
Channels not affected by this node are taken from input 2, which is considered the main data stream.
To rearrange channels in a single upstream node, Shuffle.
To simply copy a channel from one data stream into another, see Copy.
Inputs and Controls

Type Name
Input 1 The first image sequence with the channels to rearrange.
2 The second image sequence with the channels to rearrange. Any

channels not affected by this node are also taken from this input.

(Scripting) Value
ShuffleCopy Tab
1 in in rgba Channels supplied by input 1. If you check a column
REFERENCE GUIDE
262
ShuffleCopy |

(Scripting) Value
below, that channel is taken from this input, and gets

output to the channel indicated by the row.
2 in in2 rgba Channels supplied by input 2. If you check a column

below, that channel is taken from this input, and gets
output to the channel indicated by the row.
[top right out rgba The rearranged channels are output to this layer.
dropdown
menu]
[bottom right out2 none The second set of outputs from the node. This allows you
dropdown to output another four rearranged channels from the
menu] node, for a total of eight channels of output to match the
possible eight channels of input.
REFERENCE GUIDE
263
Color Nodes
Color nodes deal with color corrections, color space, and color management.
Add
The Add node lets you offset a channel’s values. It adds a fixed value to the channel, which, in effect,
lightens the whole channel. You can also add a negative value to a channel, in which case the channel gets
darker.
Inputs and Controls

Type Name
Input unnamed The image sequence to offset.
mask An optional image to use as a mask. By default, the offset is limited


Value
Add Tab
channels channels all The offset is only applied to these channels.
REFERENCE GUIDE
264
|

Value

value value 0 The value to add to the channels selected above. A

positive value lightens the selected channels, and a
negative value darkens them.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the offset is

invert invert_mask disabled Inverts the use of the mask channel so that the offset
the mask.

mask.
none.


here. This will simulate doing the addition before the
REFERENCE GUIDE
265
Clamp |

Value
using one color correct node.

offset effect at 1.
Clamp
This node constrains, or clamps, values in the selected channels to a specified range. By default, it clamps
all channels to values between 0 and 1. This can be useful for ensuring that the input’s blackest blacks and
whitest whites will be visible on an intended display device, or for restricting data fed to a subsequent
node that does not support numbers outside this range.
See also ClipTest.
Inputs and Controls

Type Name
Input unnamed The image sequence whose values you want to clamp.
mask An optional image to use as a mask. By default, the clamp is limited

REFERENCE GUIDE
266
Clamp |

Value
Clamp Tab

minimum minimum 0 The lower clamp value. By default, all values that are
lower than this number are set to this value. For
example, if you set this to 0.2, any value below 0.2 is
set to 0.2. This has the effect of causing black values to
go gray.
However, if you check enabled next to MinClampTo,

all values that are lower than this number are set to the
MinClampTo value. For example, if you set this to 0.2
and MinClampTo to 0.4, any value below 0.2 is set to
0.4.
minimum_enable enabled Whether to clamp selected channels to a minimum

value. Disabling both minimum and maximum
effectively disables the node.
maximum maximum 1 The upper clamp value. By default, all values that are
higher than this number are set to this value. For
example, if you set this to 0.8, any value above 0.8 is
set to 0.8. This has the effect of causing white values
to go gray.
However, if you check enabled next to MinClampTo,

all values that are higher than this number are set to
the MaxClampTo value. For example, if you set this to
0.8 and MinClampTo to 0.6, any value above 0.8 is set
to 0.6.
maximum_enable enabled Whether to clamp selected channels to a maximum

value. Disabling both minimum and maximum
effectively disables the node.
REFERENCE GUIDE
267
Clamp |

Value
MinClampTo MinClampTo 0 The value to which low illegal values are clamped
when MinClampTo is enabled. This sets the clamped
areas to a custom color, allowing you to visualize the
clamped range or create graphic effects.
MinClampTo_ disabled When enabled, all values below minimum are

enable clamped to the MinClampTo value.
When disabled, all values below minimum are

clamped to the minimum value.
MaxClampTo MaxClampTo 1 The value to which high illegal values are clamped
when MaxClampTo is enabled. This sets the clamped
areas to a custom color, allowing you to visualize the
clamped range or create graphic effects.
MaxClampTo_ disabled When enabled, all values above maximum are

enable clamped to the MaxClampTo value.
When disabled, all values above maximum are

clamped to the maximum value.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the clamp is

invert invert_mask disabled Inverts the use of the mask channel so that the clamp
the mask.

mask.
REFERENCE GUIDE
268
ClipTest |

Value
none.


here. This will simulate doing the clamp before the

clamp effect at 1.
ClipTest
ClipTest applies zebra stripes to all pixels outside a specified range. This can help you ensure that the
input’s blackest blacks and whitest whites are visible on an intended display device. By default, ClipTest
highlights any pixels that are outside the range from 0 to 1.
Note: This node affects the final output and not just the Viewer.
See also Clamp.
REFERENCE GUIDE
269
ClipTest |
Inputs and Controls

Type Name
Input unnamed The image sequence whose out-of-range pixels you want to
highlight.
mask An optional image to use as a mask. By default, the zebra stripes are

Value
ClipTest Tab
channels channels rgba The zebra stripes are only applied to these channels.

lower lower 0 Highlight any pixels whose values are lower than this
number.
upper upper 1 Highlight any pixels whose values are higher than this
number.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the zebra

stripes are limited to the non-black areas of this
channel.
REFERENCE GUIDE
270
CMSTestPattern |

Value
invert invert_mask disabled Inverts the use of the mask channel so that the zebra
stripes are limited to the non-white areas of the mask.
the mask.

mask.
none.


here. This will simulate doing the clip test before the

clip test effect at 1.
CMSTestPattern
This node generates calibration patterns for color management. You should run the output of this node
through a color management node and then render the result with a GenerateLUT node. This way, you can
produce 1D and 3D lookup table (LUT) files for various programs:
REFERENCE GUIDE
271
CMSTestPattern |
• 1D LUTs assume that the primary color channels (red, green, and blue) are independent, and the output
for a particular primary color (for example, red) only depends on the input value of that channel (rather
than the values of all three channels). For each input value, there is one output value. 1D LUTs are
represented by a curve. They are useful for quickly creating an image for viewing purposes, using less
processing power, but they cannot be used to adjust saturation independently of brightness.
• In 3D LUTs, the output values of each primary color channel are calculated based on the input values
from all three channels (red, green, and blue). For each input value, there are three output values. This is
represented by a 3D grid or cube. 3D LUTs can adjust saturation and brightness together, and are very
useful for emulating a particular known look, such as Cineon.
See GenerateLUT and Vectorfield for more information.
Inputs and Controls

Type Name
Output unnamed Add an arrangement of Nuke nodes that do the color correction you
want your 3D LUT to do. Then output the result using a GenerateLUT
node.
Alternatively, you can insert a Write node after the CMSTestPattern

node, write the image out, grade it in another application, and read
the result back in using a Read node. Then output the result using a
GenerateLUT node.

(Scripting) Value
CMSTestPattern Tab
RGB 3D LUT cube_size 32 Enter the cube size to use for the test pattern. The cube
cube size size defines the size of the LUT cube on the xyz axes with
higher values increasing the number of colors. In reality,
you would be unlikely to use a cube size greater than
128 and the test pattern is automatically clipped above
cube size 444.
REFERENCE GUIDE
272
ColorCorrect |
ColorCorrect
The ColorCorrect node is used to make quick adjustments to contrast, gamma, gain, and offset. You can
apply these to a clip’s master (entire tonal range), shadows, midtones, or highlights.
You can control the range of the image that is considered to be in the shadows, midtones, and highlights
using the lookup curves on the Ranges tab. However, do not adjust the midtone curve - midtones are
always equal to 1 minus the other two curves.
You can also create this node by pressing C on the Node Graph.
See also Grade and Multiply.
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the color correction.
mask An optional image to use as a mask. By default, the color correction

is limited to the non-black areas of the mask.

Value
ColorCorrect Tab
channels channels rgb The color correction is only applied to these

channels.
REFERENCE GUIDE
273
ColorCorrect |

Value
master
saturation saturation 1 Adjusts the saturation in the entire image.
contrast contrast 1 Adjusts the contrast in the entire image.
gamma gamma 1 Adjusts the gamma in the entire image.
gain gain 1 Adjusts the gain in the entire image.
offset offset 0 Adjusts the offset in the entire image.
shadows
saturation shadows.saturation 1 Adjusts the saturation in the darkest parts of the

image.
contrast shadows.contrast 1 Adjusts the contrast in the darkest parts of the image.
gamma shadows.gamma 1 Adjusts the gamma in the darkest parts of the image.
gain shadows.gain 1 Adjusts the gain in the darkest parts of the image.
offset shadows.offset 0 Adjusts the offset in the darkest parts of the image.
midtones
saturation midtones.saturation 1 Adjusts the saturation in the middle range of the

image.
contrast midtones.contrast 1 Adjusts the contrast in the middle range of the image.
gamma midtones.gamma 1 Adjusts the gamma in the middle range of the image.
gain midtones.gain 1 Adjusts the gain in the middle range of the image.
offset midtones.offset 0 Adjusts the offset in the middle range of the image.
highlights
saturation highlights.saturation 1 Adjusts the saturation in the highlights of the image.
contrast highlights.contrast 1 Adjusts the contrast in the highlights of the image.
REFERENCE GUIDE
274
ColorCorrect |

Value
gamma highlights.gamma 1 Adjusts the gamma in the highlights of the image.
gain highlights.gain 1 Adjusts the gain in the highlights of the image.
offset highlights.offset 0 Adjusts the offset in the highlights of the image.
Ranges Tab
test test disabled Lets you overlay the output with black, gray, or white
to show what is considered to be in the shadows,
midtones, or highlights. Green and magenta indicate
a mixture of ranges.
curve editor lookup N/A You can use the shadow and highlight lookup curves
to edit the range of the image that is considered to
be in the shadows or highlights. You can also look up
color information for the current pixel in the Viewer.
Warning: DO NOT adjust the midtone

curve. Midtones are always equal to 1 minus
the other two curves.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the color

correction is limited to the non-black areas of this
channel.
invert invert_mask disabled Inverts the use of the mask channel so that the color
correction is limited to the non-white areas of the
mask.
REFERENCE GUIDE
275
ColorCorrect |

Value
the mask.

mask.
none.


here. This will simulate doing the color correction
before the premultiplication was done. It is the same
as adding an Unpremult node before this node and a
Premult node after, but allows you to work faster if
you’re only using one color correct node.

mix N/A enabled Enables the mix luminance slider. Disabling this
luminance checkbox is the same as setting mix luminance to 0.
mix_luminance 0 Controls how much of the original luminance is

preserved after the color correction. A value of 0
means the altered luminance is used in the output
image. A value of 1 produces a luminance value close
to that of the original input image.
color correction at 1.
REFERENCE GUIDE
276
ColorLookup |
ColorLookup
Allows you to make contrast, gamma, gain, and offset adjustments (and, in fact, many others) using lookup
tables (LUTs). LUTs refer to line graphs of a given color channel’s brightness. The horizontal axis represents
the channel’s original, or input, values, and the vertical axis represents the channel’s new, or output values.
Inputs and Controls

Type Name
Input unnamed The image sequence to color correct.
mask An optional image to use as a mask. By default, the color correction is

or set to none.

Value
ColorLookup Tab
channels channels rgba The color correction is only applied to these channels.

use use_precomputed enabled When enabled, ColorLookup uses a precomputed

precomputed lookup table (LUT) between 0 and the value specified
table in the range field to increase processing speed.
range range 1 Sets the precomputed LUT search range between 0

and this value to increase processing speed.
REFERENCE GUIDE
277
ColorLookup |

Value
curve editor lut N/A You can use the lookup curves to edit the range of
the image in individual channels, or ganged using the
master curve (which represents all channels). To edit
one curve with reference to another, select multiple
curves.
You can look up color information for the current

pixel in the Viewer. Drag the cursor over the pixels
you want to sample for the correction. In the
ColorLookup curve editor, press Ctrl/Cmd+Alt while
clicking on the curve to set points at the places where
the red, green, and blue lines intersect with the color
curve. Edit the position of the points and adjust the
tangent handles to adjust the curve shape for the
color correction.
source source 0 Sets a source color for adding points to the curve
editor.
target target 0 Sets a destination color for adding points to the curve
editor.
Set RGB setRGB N/A Adds points on the r, g, and b curves, mapping source
to target.
You can use this control to match shadow, midtone,

and highlights on two plates, for example. Set source
to shadow rgb in one, target to shadow rgb in the
other, then press Set RGB. Same for midtone and
highlight areas.
Set RGBA setRGBA N/A Adds points on the r, g, b, and alpha curves, mapping
source to target.
Set A setA N/A Adds points on the alpha curve, mapping source to
target.
REFERENCE GUIDE
278
ColorLookup |

Value

channel to none.

channel.
mask.
the mask.

mask.
none.



REFERENCE GUIDE
279
ColorMatrix |

Value
ColorMatrix
ColorMatrix multiplies the rgb colors by an arbitrary 3x3 matrix. This is useful for transforming color
spaces.
Inputs and Controls

Type Name


Value
ColorMatrix Tab
channels channels rgb The color correction is only applied to these channels.

matrix matrix 0 The color matrix to apply to the input image. The
REFERENCE GUIDE
280
ColorMatrix |

Value
output values of each primary color channel are

calculated based on the input values from all three
channels (red, green, and blue).
For example, if you enter 0.5, 1, and 2 on the first row

of the matrix, the input red value is multiplied by 0.5,
the input green value by 1, and the input blue value by
2. These are then added together to produce the
output red value. The output green and blue values are
not affected by the first row of the matrix.
You can get the values for the matrix from the
Colorspace node’s color matrix output control, for
example. Ctrl/Cmd + drag the Animation menu to
copy the whole matrix at once.
invert invert disabled When enabled, the inverse of the matrix is used.
mask N/A 0 Enables the associated mask channel to the right.

channel to none.

channel.
REFERENCE GUIDE
281
Colorspace |

Value
invert invert_mask disabled Inverts the use of the mask channel, so that the color
mask.
the mask.

mask.
none.



effect at 1.
Colorspace
REFERENCE GUIDE
282
Colorspace |
The Colorspace node converts images from one colorspace to another, for example from Nuke’s native
colorspace to other color spaces more appropriate to a given process or intended display device. This
node supports RGB, HSV, YUV, CIE, and CMS formats (and various subformats). It can adjust for different
primaries, white point, and different encodings.
If you wish to reverse your colorspace conversion later in the script:

1. Copy the Colorspace node and insert it at the appropriate point in your script.
2. Set the out controls to sRGB, D55, and RGB.
3. Set the in controls to match the values you entered in the first Colorspace node.
If you wish to write out the input clip in the new colorspace, set the colorspace dropdown menu to linear
in the Write node controls. This halts the automatic conversion and lets the one you created have priority.
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the colorspace conversion.
mask An optional image to use as a mask. By default, the colorspace

conversion is limited to the non-black areas of the mask.
or set to none.

Value
Colorspace Tab
channels channels rgb The colorspace correction is applied only to these

channels.

individual channels
in colorspace_in linear Sets the input colorspace. This node converts the
input image from this colorspace to the out
REFERENCE GUIDE
283
Colorspace |

Value
colorspace.
illuninant_in D65 Sets the input illuminant to simulate colors under

different lighting. For example, the default D65
simulates illumination roughly equivalent to mid-day
sun in Western or Northern Europe.
primary_in sRGB Sets the input primaries (the exact "shade" of red,
green, and blue).
out colorspace_out linear Sets the required output colorspace. This node
converts the input image from the in colorspace to
this colorspace.
illuninant_out D65 Sets the output illuminant to simulate colors under

different lighting. For example, A simulates
illumination from a household tungsten filament light
bulb.
primary_out sRGB Sets the input primaries (the exact "shade" of red,
green, and blue).
Bradford bradford_matrix disabled If the illuminant_in and illuminant_out values are not
matrix the same, you can enable this to apply an illuminant
adaptation transform. This alters colors so they appear
the same.
When disabled, the input and output colors have the

same CIE colorspace XYZ values.
color matrix colormatrix N/A If a matrix is used in the color transform chosen
output above, the output appears here.
Changing the matrix values has no effect, but you can

copy them into a ColorMatrix node to apply them to
an image. Ctrl/Cmd + drag the Animation menu to
copy the whole matrix at once. See ColorMatrix for
more information.
REFERENCE GUIDE
284
Colorspace |

Value
Note: A colorspace conversion may be

doing far more than just a single 3x3 matrix
operation!
channel to none.
maskChannelInput none The channel to use as a mask. By default, the

colorspace conversion is limited to the non-black
areas of this channel.
colorspace conversion is limited to the non-white
areas of the mask.
the mask.

mask.
none.


here. This will simulate doing the color conversion
REFERENCE GUIDE
285
ColorTransfer |

Value


colorspace conversion at 1.
ColorTransfer
ColorTransfer lets you match the colors of one clip with those of another. It changes the colors in input B
to match input A.
See also Grade and MatchGrade.
Inputs and Controls

Type Name
Input A The image sequence whose colors you want to match.
B The image sequence whose colors you want to change.

(Scripting) Value
ColorTransfer Tab
ROI x, y, r, t ROI N/A The region of interest. This is the area in input A with the
REFERENCE GUIDE
286
Crosstalk |

(Scripting) Value
(or x, y, w, h) colors to match. By default, it is set to cover the entire

input.
You can adjust the following:

• x - the distance (in pixels) between the left edge of
image A and the left side of the ROI.
• y - the distance (in pixels) between the bottom edge
of image A and the bottom edge of the ROI.
• r - the distance (in pixels) between the left edge of
image A and the right side of the ROI.
• t - the distance (in pixels) between the bottom edge of
image A and the top edge of the ROI.
• w - the width of the ROI. This is only available if you
click the wh button.
• h - the height of the ROI. This is only available if you
click the wh button.
You can also adjust the ROI in the Viewer by dragging its
edges.
When adjusting the ROI, you may want to connect the

Viewer to input A rather than the ColorTransfer node, so
you can better see the area you are selecting.
Crosstalk
Adds or removes crosstalk, or bleeding, from one color channel to another. Each output color channel is
the sum of three lookup tables, each indexed by one of the input color channels. The horizontal axis
represents the channel’s original, or input, values, and the vertical axis represents the channel’s new, or
output values.
REFERENCE GUIDE
287
Crosstalk |
Inputs and Controls

Type Name
Input unnamed The image whose color you want to adjust.
mask An optional image to use as a mask. By default, the effects are limited
or set to none.

Value
Crosstalk Tab
[crossover crossover N/A From the curve list on the left, select the curve you
curves] want to modify.
For example, you’d select red to only modify the red

values.
You can also select R->G, for example, to use the red
value to evaluate the curve and add the result to the
green value. Adjust the curve as necessary. To insert
points on the curve, Ctrl/Cmd+Alt+click on the curve.
Reset N/A N/A Resets the selected curves to their initial positions.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the effects

are limited to the non-black areas of this channel.
REFERENCE GUIDE
288
Exposure |

Value

invert invert_mask disabled Inverts the use of the mask channel so that the effects
are limited to the non-white areas of the mask.
the mask.

mask.
none.


here. This will simulate applying the expressions

effect at 1.
Exposure
Allows you to adjust the exposure of the input sequence using f-stops, printer lights, film densities, or
Cineon values.
REFERENCE GUIDE
289
Exposure |
Inputs and Controls

Type Name
Input unnamed The image whose exposure you want to adjust.
mask An optional image to use as a mask. By default, the effects are limited
or set to none.

Value
Exposure Tab
channels channels rgb The exposure change is only applied to these

channels.

colorspace colorspace Linear Sets the output colorspace:

• Linear - performs a linear multiplication to calculate
the exposure.
• Cineon - data is assumed to be raw data from a
Cineon file requiring addition or subtraction. The
result, when passed through a Log2Lin node,
approximately matches the Linear result.
Note: You can match Linear exactly if you

set gamma to 0.6 and enable ignore in the
Log2Lin node Properties panel.
REFERENCE GUIDE
290
Exposure |

Value
blackpoint blackpoint 0 Sets the black point (typically the darkest pixel). Any
pixels of this value are set to 0. In other words, this
color is turned into pure black.
Typically, you would set this by sampling the darkest

pixels in the Viewer. To find these in your image, you
may want to temporarily increase the gain in the
Viewer node controls until only the darkest pixels are
visible. Adjusting the Viewer gain does not affect the
colors you sample.
Adjust in mode Densities Sets the range imposed on the red, green, and blue
sliders to alter the exposure of the input image:
• Stops
• Lights
• Densities (log10(density) of 0.6 gamma negative
stock)
• Cineon (offset of Cineon 10-bit log data)
Adjusting the color sliders has the same effect in all

modes, the range is a matter of user preference.
gang gang enabled When enabled, the color sliders are ganged together
- adjustments to one affect all three equally.
red red 0 Adjusts the exposure in the red channel using the
range specified by the Adjust in control.
green green 0 Adjusts the exposure in the green channel using the
blue blue 0 Adjusts the exposure in the blue channel using the
channel to none.
maskChannelInput none The channel to use as a mask. By default, the effects
REFERENCE GUIDE
291
Exposure |

Value
are limited to the non-black areas of this channel.
invert invert_mask disabled Inverts the use of the mask channel so that the effects
are limited to the non-white areas of the mask.
the mask.

mask.
none.



effect at 1.
REFERENCE GUIDE
292
Expression |
Expression
Allows you to apply complex mathematical formulas to a channel’s values using C-like syntax expressions.
If necessary, you can apply different expressions to different sets of channels.
• x or y is the pixel coordinate. If the input is a proxy image, then these are scaled and translated to the
coordinate that would be in the full size image.
• cx and cy can provide a more useful coordinate system. In this system 0,0 is the center of the picture.
-1,0 is the center of the left edge, 1,0, is the center of the right edge. 1,1 is a point 45 degrees up from
the origin on the right edge (outside the top of the picture if the aspect ratio is greater than 1).
• You can refer to any input channel by name, for example r for the red channel.
• You can refer to any control on any node by its name. For example, Blur1.size returns the size of the blur.
You can also evaluate animated controls, for example Blur1.size(t).
• A blank expression is the same as zero.
• All math is done with 32-bit floating point numbers.
See also MergeExpression.
Inputs and Controls

Type Name
Input unnamed The image sequence to which you want to apply expressions.
mask An optional image to use as a mask. By default, the expressions are

REFERENCE GUIDE
293
Expression |

Value
Expression Tab
[variable temp_name0 N/A If you need to use a long expression in several fields,
name field] you can use this row for assigning the expression
temporarily to a variable. Enter the variable name
here, and the expression on the right of the = sign.
You can then use the variable to represent the
expression in the = fields next to the channels.
= temp_expr0 N/A If you need to use a long expression in several fields,

you can use this row for assigning the expression
temporarily to a variable. Enter the variable name on
the left side of the = sign, and the expression in this
field. You can then use the variable to represent the


REFERENCE GUIDE
294
Expression |

Value


channels channel0 red The channel(s) to which you want to apply the
expression in the below = field.

= expr0 N/A The expression to apply to the above channels.
For example, to assign noise to the above channel(s)

and then boost the gain of that result by 20, you
would type (random*r)*20 here.
To reference pixels in other channels, use

layer.channel (for example, matte.garbage). If you
don’t specify the layer name, the Expression node
assumes the channel is in the current layer. As a
shortcut, you can use r, g, b, and a to reference the
red, green, blue, and alpha channels in the rgba layer.
channels channel1 green The channel(s) to which you want to apply the
REFERENCE GUIDE
295
Expression |

Value



channels channel2 blue The channel(s) to which you want to apply the



channels channel3 alpha The channel(s) to which you want to apply the
REFERENCE GUIDE
296
Expression |

Value



channel to none.

expressions are limited to the non-black areas of this
channel.
expressions are limited to the non-white areas of the
mask.
the mask.

mask.
REFERENCE GUIDE
297
Expression |

Value
none.



expressions effect at 1.
The following functions are supported:
Function Purpose Operator Usage Related Functions
abs (x) Returns the absolute value of a x See also: fabs.

floating-point number x.
acos (x) Calculates the arc cosine of x; that If x is less than -1 or See also: cos, cosh, asin, atan.
is the value whose cosine is x. greater 1, acos returns
nan (not a number)
asin (x) Calculates the arc sine of x; that is If x is less than -1 or See also: sin, sinh, acos, atan.
the value whose sine is x. greater 1, asin returns
nan (not a number)
atan (x) Calculates the arc tangent of x; x See also: tan, tanh, acos, asin,
that is the value whose tangent is atan2.
x. The return value will be
between -PI/2 and PI/2.
REFERENCE GUIDE
298
Expression |
atan2 (x, y) Calculates the arc tangent of the x, y See also: sin, cos, tan, asin,
two variables x and y. This acos, atan, hypot.
function is useful to calculate the
angle between two vectors.
ceil (x) Round x up to the nearest integer. x See also: floor, trunc, rint.
clamp (x, Return x clamped to the min and x, min, max See also: min, max.
min, max) max values specified.
clamp (x) Return x clamped to [0.0 ... 1.0]. x See also: min, max.
cos (x) Returns the cosine of x. x in radians See also: acos, sin, tan, cosh.
cosh (x) Returns the hyperbolic cosine of x See also: cos, acos, sinh, tanh.
x, which is defined
mathematically as:
(exp(x) + exp(-x)) / 2.
degrees (x) Convert the angle x from radians x See also: radians.
into degrees.
exp (x) Returns the value of e (the base of x See also: log, log10.
natural logarithms) raised to the
power of x.
exponent (x) Exponent of x. x See also: mantissa, ldexp.
fBm (x, y, z, Fractional Brownian Motion. This x, y, z, octaves, See also: noise, random,
octaves, is the sum of octave calls to noise lacunarity, gain turbulence.
lacunarity, (). The input point for each is
gain) multiplied by pow(lacunarity,i)
and the result is multiplied by
pow(gain,i). For normal use,
lacunarity should be greater than
1 and gain should be less than 1.
fabs (x) Returns the absolute value of the x See also: abs.
floating-point number x.
false () Always returns 0. none See also: true.
floor (x) Round x down to the nearest x See also: ceil, trunc, rint.
REFERENCE GUIDE
299
Expression |
integer.
fmod (x, y) Computes the remainder of x, y See also: ceil, floor.

dividing x by y. The return value is
x - n y, where n is the quotient of
x / y, rounded towards zero to an
integer.
frame () Return the current frame number. none See also: x.
from_byte Converts an sRGB pixel value to a color_component See also: to_sRGB, to_rec709,
(color linear value. from_rec709.
component)
from_rec709 Converts a rec709 byte value to a color_component See also: from_sRGB, to_
(color linear brightness. rec709.
component)
from_sRGB Converts an sRGB pixel value to a color_component See also: to_sRGB, to_rec709,
(color linear value. from_rec709.
component)
hypot (x, y) Returns the sqrt(x*x + y*y). This is x, y See also: atan2.
the length of the hypotenuse of a
right-angle triangle with sides of
length x and y.
int (x) Round x to the nearest integer x See also: ceil, floor, trunc,
not larger in absolute value. rint.
ldexp (x, Returns the result of multiplying x, exp See also: exponent.
exp) the floating-point number x by 2
raised to the power exp.
lerp (a, b, x) Returns a point on the line f(x) a, b, x See also: step, smoothstep.
where f(0)==a and f(1)==b.
Matches the lerp function in other

shading languages.
log (x) Returns the natural logarithm of x. x See also: log10, exp.
log10 (x) Returns the base-10 logarithm of x See also: log, exp.
REFERENCE GUIDE
300
Expression |
x.
logb (x) Same as exponent(). x See also: mantissa, exponent.
mantissa (x) Returns the normalized fraction. If x See also: exponent

the argument x is not zero, the
normalized fraction is x times a
power of two, and is always in the
range 1/2 (inclusive) to 1
(exclusive). If x is zero, then the
normalized fraction is zero and
exponent() returns zero.
max (x, y, ... ) Return the greatest of all values. x, y, (...) See also: min, clamp.
min (x, y, ... ) Return the smallest of all values. x, y, (...) See also: max, clamp
mix (a, b, x) Same as lerp(). a, b, x See also: step, smoothstep,

lerp
noise (x, y, z) Creates a 3D Perlin noise value - a x, optional y, optional z See also: random, fBm,
signed range centered on zero. turbulence
The absolute maximum range is
from -1.0 to 1.0.
This produces zero at all integers,

so you should rotate the
coordinates somewhat (add a
fraction of y and z to x, etc.) if you
want to use this for random
number generation.
pi () Returns the value for pi none none

(3.141592654...).
pow (x, y) Returns the value of x raised to x, y See also: log, exp, pow
the power of y.
pow2 (x) Returns the value of x raised to x, y See also: pow

the power of 2.
radians (x) Convert the angle x from degrees x See also: degrees
REFERENCE GUIDE
301
Expression |
into radians.
random (x, y, Creates a pseudo random value optional x, optional y, See also: noise, fBm,
z) between 0 and 1 - it always optional z turbulence
generates the same value for the
same x, y and z.
Calling random with no

arguments creates a different
value on every invocation.
rint (x) Round x to the nearest integer. x See also: ceil, floor, int, trunc
sin (x) Returns the sine of x. x in radians See also: asin, cos, tan, sinh
sinh (x) Returns the hyperbolic sine of x, x See also: sin, asin, cosh, tanh
which is defined mathematically
as:
(exp(x) - exp(-x)) / 2.
smoothstep Returns 0 if x is less than a, returns a, b, x See also: step, lerp

(a, b, x) 1 if x is greater or equal to b,
returns a smooth cubic
interpolation otherwise.
Matches the smoothstep function

in other shading languages.
sqrt (x) Returns the non-negative square x See also: pow, pow2
root of x.
step (a, x) Returns 0 if x is less than a, returns a, x See also: smoothstep, lerp
1 otherwise.
Matches the step function other

shading languages.
tan (x) Returns the tangent of x. x in radians See also: atan, cos, sin, tanh,
atan2
tanh (x) Returns the hyperbolic tangent of x See also: tan, atan, sinh, cosh
x, which is defined
REFERENCE GUIDE
302
Gamma |
mathematically as:
sinh(x) / cosh(x).
to_byte Converts a floating point pixel color_component See also: from_sRGB, to_
(color value to an 8-bit value that rec709, from_rec709
component) represents that number in sRGB
space.
to_rec709 Converts a floating point pixel color_component See also: from_sRGB, from_
(color value to an 8-bit value that rec709
component) represents that brightness in the
rec709 standard when that
standard is mapped to the 0-255
range.
to_sRGB Converts a floating point pixel color_component See also: from_sRGB, to_
(color value to an 8-bit value that rec709, from_rec709
component) represents that number in sRGB
space.
true () Always Returns 1. none See also: false
trunc (x) Round x to the nearest integer x See also: ceil, floor, int, rint
not larger in absolute value.
turbulence This is the same as fBm() except x, y, z, octaves, See also: fBm, noise, random
(x, y, z, the absolute value of the noise() lacunarity, gain
octaves, function is used.
lacunarity,
gain)
x () Return the current frame number. none See also: frame
y (frame) Evaluates the y value for an optional: frame, none

animation at the given frame. defaults to current
frame
Gamma
Applies a constant gamma value to a set of channels. This lightens or darkens the midtones.
REFERENCE GUIDE
303
Gamma |
The actual function is 1/pow(value).
Note: Increasing the gamma to lighten the midtones darkens any values above 1. Similarly,
decreasing the gamma to darken the midtones lightens any values below 0. If you don’t want this
to happen, you can instead use a Colorspace node, which allows you to adjust the gamma using a
color curve rather than a slider. Simply add points on the curve just below 0 and above 1 to
preserve the super blacks and whites.
Inputs and Controls

Type Name
Input unnamed The image sequence to which you want to apply a gamma value.
mask An optional image to use as a mask. By default, the gamma

adjustment is limited to the non-black areas of the mask.

Value
Gamma Tab
channels channels all The gamma adjustment is only applied to these

channels.

value value 1 The gamma value. The default value of 1 equals no

change.
Note that you can enter formulae into this field,
REFERENCE GUIDE
304
Gamma |

Value
making it easy to do quick calculations. For example,

you could type 1/2.2 here and press Return to get
0.45454545.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the gamma

adjustment is limited to the non-black areas of this
channel.
invert invert_mask disabled Inverts the use of the mask channel so that the gamma
adjustment is limited to the non-white areas of the
mask.
the mask.

mask.
none.


here. This will simulate doing the gamma adjustment
REFERENCE GUIDE
305
GenerateLUT |

Value

gamma adjustment at 1.
GenerateLUT
Generates lookup tables, or LUTs, from color management nodes downstream of a CMSTestPattern node.
LUTs are generated according to the size of CMSTestPattern cube size.
You can use the GenerateLUT node to create both 1D and 3D LUTs:
• 1D LUTs assume that the primary color channels (red, green, and blue) are independent, and the output
for a particular primary color (for example, red) only depends on the input value of that channel (rather
than the values of all three channels). For each input value, there is one output value. 1D LUTs are
represented by a curve. They are useful for quickly creating an image for viewing purposes, using less
processing power, but they cannot be used to adjust saturation independently of brightness.
• In 3D LUTs, the output values of each primary color channel are calculated based on the input values
from all three channels (red, green, and blue). For each input value, there are three output values. This is
represented by a 3D grid or cube. 3D LUTs can adjust saturation and brightness together, and are very
useful for emulating a particular known look, such as Cineon.
Tip: For good visual fidelity when using 3D LUTs, it is recommended that a log colorspace is
used for 3D LUT generation. For more information on how to generate a LUT in log colorspace,
please refer to Nuke's online help.
For more information, see CMSTestPattern and Vectorfield.
REFERENCE GUIDE
306
GenerateLUT |
Inputs and Controls

Type Name
Input unnamed An arrangement of nodes that do the color correction you want your
3D LUT to do, preceded by a CMSTestPattern node.

Value
GenerateLUT Tab
output file file none The path and file name of the LUT file you want to
generate.
file type file_type none The file format of the LUT file you want to generate.
If this is set to auto (the default), Nuke guesses the
format from the file name extension.
The following file extensions are recognized:

• .3dl - Autodesk and ASSIMILATE Scratch LUT
format (3D).
• .blut - Side Effects Houdini binary LUT format (3D).
• .cms - Light Illusion LightSpace CMS LUT format
(1D and 3D).
• .csp - Rising Sun Research CineSpace LUT format
(3D).
• .cub - FilmLight Truelight LUT format (1D and 3D).
• .cube - IRIDAS LUT format (1D and 3D).
• .vf - Nuke's native LUT format used by the
Vectorfield node (3D).
REFERENCE GUIDE
307
GenerateLUT |

Value
Note: FilmLight and Foundry have elected to no longer bundle Truelight as of Nuke 11. The
Truelight node has been superseded by the free version of Baselight for Nuke which itself can
import a BLG (Baselight Linked Grade) that can contain a Truelight strip, plus optionally a lot
more in terms of creative grading intent. Together with OpenColorIO and the native Nuke
Vectorfield node we feel all the Nuke color management issues for which the Truelight node
was originally intended are now addressed.
For further information on Baselight for Nuke, please contact [email protected]
1D preLUT
1D preLUT lut1d 1024 The size of the 1D LUT.

size
style1d logarithmic Some systems assume the LUT is designed to apply
to logarithmic images, others linear. Some LUT
formats are assumed to encode values as either
logarithmic or linear. Use this setting to get your
output LUT values encoded the way your use of the
LUT requires.
3D LUT (These controls are only visible when you’ve selected 3dl in the file type dropdown menu.)
input bit bitdepth_id 10 bit The bit depth to scale input values to for .3dl files.
depth (.3dl) For example, a 10-bit file has values between 0 and
1023.
output bit output_bitdepth_id input bit The bit depth to scale output values to for .3dl files.
depth (.3dl) depth For instance a 10-bit file has values between 0 and
1023. In general this should match the input bit
depth.
clamp ouput clamp_output enabled Clamp 3D LUT output values to the range of 0-1.
Turning this off allows superwhite values to be
represented which are allowed in some packages,
such as Nuke. This may cause the result to be
unreadable in other packages.
3dl file format_3dl_id autodesk Select the format you want to use for your .3dl file.
format flame/smoke The final cube size is determined by the
REFERENCE GUIDE
308
Grade |

Value
CMSTestPattern which generates the input.

• autodesk flame/smoke - use Autodesk preferred
cube sizes of 17x17, 33x33, and 65x65.
• autodesk lustre - use Autodesk preferred cube
sizes of 17x17, 33x33, and 65x65, and output a
3DMESH/Mesh header.
• assimilate scratch - scale and offset the 3D LUT.
This should be used only for legacy scratch files
prior to Flame compatibility.
Generate and generate N/A Generate the LUT file and save it to the output file
Write LUT location.
File

aborts.
the render aborts.

Grade
REFERENCE GUIDE
309
Grade |
Lets you define white and black points by sampling pixels from the Viewer. Setting the brightest parts of
the image to pure white and the darkest to pure black in this manner can help you add punch to
overexposed images, for example. You can also use this node for matching foreground plates to
background plates.
To sample pixels from the Viewer, click on a color swatch to the right of the control you want to set. The
eye dropper icon appears. In the Viewer:
• Ctrl/Cmd+click to sample a single pixel from the node’s output.
• Ctrl/Cmd+Shift+drag to sample a region of pixels from the node’s output.
• Ctrl/Cmd+Alt+click to sample a single pixel from the node’s input while viewing its output.
• Ctrl/Cmd+Alt+Shift+drag to sample a region of pixels from the node’s input while viewing its output.
• Crtl/Cmd+ right-click cancels pixel selections.
You can also create this node by pressing G on the Node Graph.
See also ColorCorrect and Multiply.
Inputs and Controls

Type Name
Input unnamed The image sequence whose tonal range you want to define.

or set to none.

Value
Grade Tab
REFERENCE GUIDE
310
Grade |

Value
blackpoint blackpoint 0 Sets the black point (typically the darkest pixel). Any
color is turned into pure black.
Typically, you would set this by sampling the darkest

pixels in the Viewer. To find these in your image, you
may want to temporarily increase the gain in the
Viewer controls until only the darkest pixels are
visible. Adjusting the Viewer gain does not affect the
colors you sample.
whitepoint whitepoint 1 Sets the white point (typically the lightest pixel). Any
color is turned into pure white.
Typically, you would set this by sampling the

brightest pixels in the Viewer. To find these in your
image, you may want to temporarily lower the
gamma in the Viewer controls until only the brightest
pixels are visible. Adjusting the Viewer gamma does
not affect the colors you sample.
lift black 0 Any pixels that are black are set to this color.
This allows you to, for example, match the tonal range
of one clip with that of another. First, set blackpoint
to the darkest pixel in the image whose colors you
want to change (pressing Alt to make sure you sample
from the input). Then, set this control to the darkest
pixel in the image whose tonal range you want to
match (not pressing Alt to make sure you sample
from the output).
gain white 1 Any pixels that are white are set to this color.
This allows you to, for example, match the tonal range
of one clip with that of another. First, set whitepoint
REFERENCE GUIDE
311
Grade |

Value
to the lightest pixel in the image whose colors you

want to change (pressing Alt to make sure you sample
from the input). Then, set this control to the lightest
pixel in the image whose tonal range you want to
match (not pressing Alt to make sure you sample
from the output).
multiply multiply 1 Multiplies the result of the Grade node by this factor.
This has the effect of lightening the result while
preserving the black point.
offset add 0 Offsets the result of the Grade node. Offsetting is to

add a fixed value to the result, which, in effect
lightens the whole image. You can also add a negative
value to the result, in which case the whole image
gets darker.
gamma gamma 1 Applies a constant gamma value to the result of the

Grade node. This lightens or darkens the midtones.
reverse reverse disabled Allows you to copy and paste this node and invert its
effect further downstream. This performs the
opposite gamma correction, followed by the
opposite linear ramp.
black clamp black_clamp enabled When enabled, any output values that are lower than 0
are set to 0.
white clamp white_clamp disabled When enabled, any output values that are higher than
1 are set to 1.
channel to none.

channel.
REFERENCE GUIDE
312
Grade |

Value

mask.
the mask.

mask.
none.



REFERENCE GUIDE
313
HistEQ |
HistEQ
Alters the histogram of the input image to be flat. This process is called histogram equalization. It can be
useful for increasing the contrast in an image, for example, if the background and foreground in the image
are both bright or both dark.
Inputs and Controls

Type Name
Input unnamed The image sequence to equalize.

Value
HistEQ Tab
Histogram histeqbins 4096 Adjust the number of entries in the histogram.

Entries
channel to none.

channel.
invert mask invert_mask disabled Inverts the use of the mask channel so that the color
mask.
REFERENCE GUIDE
314
Histogram |
Histogram
Histogram graphs the number of pixels at each brightness level. This is a useful gauge to see whether the
input image has a good distribution of shadows, midtones, and highlights.
You can also use this node to adjust the tonal range of the input image. Note, however, that Histogram
clamps superblacks and superwhites. If you don’t want this to happen, use the Grade node instead.
Tip: You can also use the histogram panel available in the content menu under Windows >
Histogram.
Inputs and Controls

Type Name
Input unnamed The image sequence whose tonal range you want to examine or
adjust.

or set to none.

Value
Histogram Tab
Histogram hist N/A The horizontal axis represents brightness (the

numerical value of a pixel), with the shadows on the
left, midtones in the middle, and highlights on the
REFERENCE GUIDE
315
Histogram |

Value
right. The vertical axis represents pixel count (the

percentage of pixels with that value).
input range in 0, 1, 1 Allows you to extend the image’s tonal range, which
has the effect of increasing contrast.
To set the black point, drag the leftmost input range

slider till it roughly lines up with the initial boundary
of the histogram. This makes shadows darker.
To set the white point, drag the rightmost input range

slider till it roughly lines up with the final boundary of
the histogram. This makes highlights lighter.
To set the midtone, or neutral, value, drag the middle

input slider. This applies a gamma correction (lightens
or darkens the midtones).
The histogram updates as you adjust these controls.
output range out 0, 1 Sets the range to which the input range is mapped.
This can be used to shorten the image’s tonal range,
which has the effect of removing pure black and
white and decreasing contrast.
Any pixels that are black are set to the value on the
left. This makes shadows lighter.
Any pixels that are white are set to the value on the
right. This makes highlights darker.
The histogram updates as you adjust these controls.
channel to none.

channel.
REFERENCE GUIDE
316
HSVTool |

Value
mask.
the mask.

mask.
none.



HSVTool
REFERENCE GUIDE
317
HSVTool |
HSVTool lets you simultaneously adjust hue, saturation, and value components from a single control panel.
It also features a color replacement tool and makes an excellent keyer.
The main strength of this node is the precision it offers in limiting corrections to a narrow sample of
colors. You can limit the correction to certain hues by sampling a few pixels and then shifting their values.
Because you limit the color range, the surrounding image is generally unaffected by the shift.
Inputs and Controls

Type Name
Input unnamed The image sequence whose HSV components you are adjusting.

(Scripting) Value
HSVTool Tab
Color Replacement
srccolor srccolor 0 Sets the source color you intend to adjust from.
dstcolor dstcolor 0 Sets the destination color you intend to adjust to.
Hue
Range huesrcs 0, 360
Rotation huerotation 0
Reset huereset N/A Click to reset all Hue controls to the default setting.
Range Rolloff huerolloffrange 0
HMask N/A disabled Enables the associated mask channel to the right.
channel to none.
HMask none The channel to use as a hue mask.
invert mask invert_hue_ disabled Inverts the use of the mask channel so that the
mask correction is limited to the non-white areas of the
mask.
REFERENCE GUIDE
318
HSVTool |

(Scripting) Value
Saturation
Range satsrcs 0, 1
Adjustment saturation 0
Reset satreset N/A Click to reset all Saturation controls to the default
setting.
Range Rolloff satrolloffrange 0
Force Value setsaturation disabled
SMask N/A disabled Enables the associated mask channel to the right.
channel to none.
SMask none The channel to use as a saturation mask.
invert mask invert_sat_mask disabled Inverts the use of the mask channel so that the
mask.
Brightness
Range brtsrcs 0, 1
Adjustment brightness 0
Reset brtreset N/A Click to reset all Brightness controls to the default
setting.
Range Rolloff brtrolloffrange 0
Force Value setbrightness disabled
BMask N/A disabled Enables the associated mask channel to the right.
channel to none.
BMask none The channel to use as a brightness mask.
invert mask invert_bright_ disabled Inverts the use of the mask channel so that the
REFERENCE GUIDE
319
HueCorrect |

(Scripting) Value
mask.
output N/A enabled Enables the associated mask channel to the right.
channel to none.
output rgba.alpha The channel to use for the output mask.
is Conversion hue
invert mask invert_alpha_ disabled Inverts the use of the mask channel so that the
mask.
HueCorrect
Lets you make precision adjustments to the levels of saturation in a range of hues. You do so via edits to a
series of suppression curves. The horizontal axis represents the original, or input, saturation, and the
vertical axis represents the new, or output, saturation.
By choosing which curve you edit and how much of that curve you alter, you can precisely limit the
influence of the effect.
For the compositor, HueCorrect is obviously of greatest use in diminishing green, blue, or redscreen spill.
For the times when you just want to correct the saturation component and don’t require limiting the
correction to any particular range of hues, you can also use the Saturation node.
REFERENCE GUIDE
320
HueCorrect |
Inputs and Controls

Type Name
Input unnamed The image sequence to correct.
mask An optional image to use as a mask. By default, the correction is

or set to none.

Value
HueCorrect Tab
channels channels rgb The correction is only applied to these channels.

curve editor hue N/A Use the curves to adjust the hue of the input image.
You can also look up color information for the current
pixel in the Viewer.
reset N/A N/A Click to reset the curves to the default values.
channel to none.

channel.
REFERENCE GUIDE
321
HueCorrect |

Value
mask.
the mask.

mask.
none.


here. This will simulate applying the correction before
the premultiplication was done. It is the same as
adding an Unpremult node before this node and a
you’re only using one HueShift node.

mix N/A enabled Enables the mix luminance slider. Disabling this
luminance checkbox is the same as setting mix luminance to 0.
mix_luminance 0 Controls how much of the original luminance is

preserved after the color correction. A value of 0
means the altered luminance is used in the output
image. A value of 1 produces a luminance value close
to that of the original input image.
REFERENCE GUIDE
322
HueShift |

Value
correction at 1.
HueShift
HueShift transforms the input image’s colorspace. It works by transforming the colors to CIE XYZ space and
then doing simple scales and rotates on the space. In XYZ space, X and Z control the color and Y is the
brightness.
Inputs and Controls

Type Name
Input unnamed The image sequence to hue shift.
mask An optional image to use as a mask. By default, the shift is limited to

or set to none.

Value
HueCorrect Tab
channels channels rgb The shift is only applied to these channels.

REFERENCE GUIDE
323
HueShift |

Value
input ingray 0.25 After converting to CIE XYZ space, the colors are
graypoint skewed so that this gray color is X = Z = 0.
output outgray 0.25 After all other operations are completed, the CIE XYZ
graypoint space is skewed to the output gray point so a fully
desaturated images is tinted the specified color.
overall saturation 1 Multiplies X and Z by the specified overall saturation

saturation value.
color axis color 1, 0, 0 Adjusts the saturation in one direction in the XZ plane
rgb by setting this color axis. Adjust the saturation along
axis control to affect the saturation.
saturation color_saturation 1 Adjusts the color saturation in the direction specifed

along axis by the color axis control.
hue rotation hue_rotation 0 Adjusts the rotation around the Y axis, which is mostly
useful for making psychedelic effects.
brightness brightness 1 Multiplies the image by the specified value.
channel to none.

channel.
mask.
the mask.
REFERENCE GUIDE
324
Invert |

Value
mask.
none.


here. This will simulate applying the correction before
you’re only using one HueShift node.

shift at 1.
Invert
Inverts a channel’s values. To invert a channel is to subtract its values from one, which causes its blacks to
become white and its whites to become black. In the course of building a script, you’ll have frequent need
to invert mattes in particular.
REFERENCE GUIDE
325
Invert |
Inputs and Controls

Type Name
Input unnamed The image sequence to invert.
mask An optional image to use as a mask. By default, the inversion is

or set to none.

Value
Invert Tab
channels channels all The inversion is only applied to these channels.

clamp clamp disabled Restricts the results of the inversion to values

between 0 and 1, clamping superblacks and
superwhites.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the inversion

REFERENCE GUIDE
326
Log2Lin |

Value
inversion is limited to the non-white areas of the

mask.
the mask.

mask.
none.


here. This will simulate doing the inversion before the

inversion at 1.
Log2Lin
Whenever you read a clip into a script, it is automatically converted to Nuke’s native color space, which is
32-bit per channel RGB, a linear format. This conversion takes place even if the clip you read in is in the
Kodak Cineon format, which is a logarithmic format. The reverse of this conversion, called a lin-to-log
conversion, also automatically takes place when you write the processed element back out of the script -
that is, Nuke automatically converts it back into a Cineon element.
REFERENCE GUIDE
327
Log2Lin |
Nuke uses the Kodak-recommended settings when making Cineon conversions in either direction. It’s rare
that you would want to override these settings, but if it becomes necessary you can use the Log2Lin node.
If you do, you should also check raw data (or set colorspace to linear) in the Read and Write node
controls to skip the automatic conversion.
Inputs and Controls

Type Name
Input unnamed The image sequence to convert.
mask An optional image to use as a mask. By default, the shift is conversion

or set to none.

Value
Log2Lin Tab
channels channels rgb The conversion is only applied to these channels.

operation operation log2lin Sets whether the conversion is from or to Cineon log
colorspace:
• log2lin - from Cineon colorspace.
• lin2log - to Cineon colorspace.
black black 95 Sets the value in the Cineon file that is converted into
0.
reset N/A N/A Click to reset the black value to the default setting.
ignore ignore_black disabled When enabled, the black offset is not subtracted so
REFERENCE GUIDE
328
Log2Lin |

Value
that the conversion is a pure log function.
white white 685 Sets the value in the Cineon file that is converted into
1.
reset N/A N/A Click to reset the white value to the default setting.
gamma gamma 0.6 Sets the exposed light level multiplier to calculate the
film density.
reset N/A N/A Click to reset the gamma value to the default setting.
channel to none.

conversion is limited to the non-black areas of this
channel.
conversion is limited to the non-white areas of the
mask.
the mask.

mask..
none.

REFERENCE GUIDE
329
MatchGrade |

Value

here. This will simulate applying the conversion
you’re only using one node.

conversion at 1.
MatchGrade
MatchGrade (NukeX and Nuke Studio only) allows you to automatically calculate a grade to match the
colors in the Source input to the colors in the Target input. You can use MatchGrade to:
• extract a baked-in grade if the Target clip that you want to match to contains the same frames as the
Source clip,
• match the grade between two different clips to create the same look,
• mask the grade to only match certain elements between the Source and Target clips, and
• export LUT or CDL files to re-use the calculated grade elsewhere.
Inputs and Controls

Type Name
Input Target The image sequence you want the Source clip to match.
REFERENCE GUIDE
330
MatchGrade |

Type Name
When Task is set to Match Graded Source, this should be a graded

version of the Source footage.
When Task is set to Match Different Clip, this should be a different

clip rather than a version of the Source footage.
Source The image sequence to which the color transform is applied.
MatchGrade Tab
Task matchGradeTask Match Graded Choose whether you want to match the Source
Source input with a graded version of the same clip or
a different clip:
• Match Graded Source - Match the Source
clip with a graded version of the same clip
(connected to the Target input). This allows
you to extract and export a baked-in grade
applied to the original clip.
In this mode, MatchGrade expects the same

frames in both its inputs.
• Match Different Clip - Match the colors of
one clip (Source) with those of another
(Target). This allows you to create a grade to
reproduce the look of the Target clip in the
Source clip.
In this mode, MatchGrade expects different

frames in the Source and Target inputs.
Analysis analysisType Analyze Set the type of analysis to perform:

Reference • Analyze Reference Frames - Calculate a
Frames single, global grade from selected reference
frames that cover the characteristic colors in
the sequence and apply that grade to every
frame of the Source sequence. You can use
REFERENCE GUIDE
331
MatchGrade |
the controls under Ref Frames to choose the

reference frames.
• Auto-analyze Per Frame - Calculate a local
grade frame-by-frame, so that the color
transform updates according to the current
frame.
Note: This control is only available

when Task is set to Match Graded
Source.
Mask mask None Set whether to use a mask channel to limit the
grade to certain areas of the image:
• None - Do not use a mask. The grade is
applied to the entire image.
• Alpha - Use the alpha channel as a mask. The
grade is limited to the non-black areas of the
alpha channel.
• Inverted Alpha - Invert the alpha channel and
use that as a mask. The grade is limited to the
non-white areas of the alpha channel.
Using a mask can be useful if you only want to

match certain elements between shots or
balance the color content when Task is set to
Match Different Clip.
Note: When Task is set to Match

Graded Source, only the alpha from
the Source input is used.
When Task is set to Match Different
Clip, the alpha from both inputs is
used.
Apply Grade to applyToMask enabled When enabled, the computed grade is only
REFERENCE GUIDE
332
MatchGrade |
Masked Region applied to the areas of the source controlled by

the Mask.
When disabled, the grade is applied to the

whole image instead, allowing you to compute
the grade from a selected region and apply it
to the whole image without having to export
the LUT.
Note: This control is only displayed

when the Mask control is set to
anything other than None.
Output output Matched Select what to display in the Viewer:

• Matched - View the color matched result.
• Source - View the Source input. This can be
useful when setting reference frames for the
Source input.
• Target - View the Target input. This can be
useful when setting reference frames for the
Target input.
Align Target to alignTargetToSource N/A Click this to add a Transform node and a
Source Reformat node upstream of the MatchGrade
node to automatically align the target with the
source clip.
You can use Align Target to Source if the clips

are not aligned in time and space, such as when
the target clip has been reformatted.

Source.
Crop Target autoCrop disabled When enabled, clicking Align Target to Source
REFERENCE GUIDE
333
MatchGrade |
generates a rectangular crop for the target

input.
You can use Crop Target if the target contains a

region that isn't present in the source, for
example a black border.

Source.
Ref Frames - These controls are only available when Analysis is set to Analyze Reference Frames.
Source sourceRefFrames 0 Displays the Source input frames that you have
set as reference frames. These should be frames
that cover the characteristic colors in the
Source sequence.
When Task is set to Match Graded Source, the

same frames are also used as reference frames
from the Target input.
addSourceRef- N/A Set the current Source input frame as a

Frame reference frame.
deleteSourceRef- N/A Remove the current Source input frame from

Frame the reference frames.
Target targetRefFrames 0 Displays the Target input frames that you have
set as reference frames. These should be frames
that cover similar content as the Source
reference frames (for example, the same
amount of grass, skin, and so on).
Note: You do not need to use the

same number of Source and Target
reference frames or set the same
frames as reference frames for both.
REFERENCE GUIDE
334
MatchGrade |

when Task is set to Match Different
Clip.
addTargetRefFrame N/A Set the current Target input frame as a

reference frame.

Clip.
deleteTargetRef- N/A Remove the current Target input frame from

Frame the reference frames.

Clip.
Analysis
Transform transform 3D LUT Set the transformation used to grade the

Source input:
• 3D LUT - Calculate the grade as a 3D look-up
table (LUT). If necessary, you can use the
Export controls to write this to a .csp format
and then click Create OCIOFileTransform to
create an OCIOFileTransform node that
applies the grade.
• CDL - Calculate the grade as a color decision
list (CDL). If necessary, you can click Create
OCIOCDLTransform to create an
OCIOCDLTransform node that applies the
grade.
REFERENCE GUIDE
335
MatchGrade |
Note: The CDL transform is limited

and cannot model all types of color
transformation. In most cases,
selecting 3D LUT gives the best
results.

when you have set at least one
reference frame.
Pre LUT prelut Auto Detect Specify a 1D shaper LUT to use for the analysis:
• Auto Detect - Automatically detect the best
pre-LUT to use.
• Linear - Use a linear pre-LUT.
• Logarithmic - Use a logarithmic pre-LUT.
LUT Resolution lutres 32 Set the resolution of the LUT in which

MatchGrade stores the color transformation.
A value of 16 speeds up processing. A value of

64 can improve the results, but also increases
processing time.
If two very similar colors in the Source image

are very different in the Target, try setting this
to 64, which is the maximum value.
Colorspace colorspace RGB Set the colorspace to use:

• RGB
• YUV
• LAB
The correct setting to use depends on the

nature of the transformation. Try each option to
see which works best with your footage.
REFERENCE GUIDE
336
MatchGrade |

Clip.
Iterations iterations 6 The number of refinement passes. Higher values

can produce a better color match, but also take
longer to process.

Clip.
Analyze analyzeRefFrames N/A Click this to calculate the grade.

Reference
Frames
when you have set at least one
reference frame.
slope slope N/A The slope value of the CDL transform. If you're
not happy with the value MatchGrade calculates
automatically, you can use this to fine-tune the
results.

when Transform is set to CDL and
you have clicked Analyze Reference
Frames.
lock lockSlope disabled Lock the slope value to prevent MatchGrade

from recalculating it when Recalculate CDL is
pressed.
REFERENCE GUIDE
337
MatchGrade |

Frames.
offset offset N/A The offset value of the CDL transform. If you're
results.

Frames.
lock lockOffset disabled Lock the offset value to prevent MatchGrade

pressed.

Frames.
power power N/A The power value of the CDL transform. If you're
results.

Frames.
REFERENCE GUIDE
338
MatchGrade |
lock lockPower disabled Lock the power value to prevent MatchGrade

pressed.

Frames.
saturation saturation N/A The saturation value of the CDL transform. If

you're not happy with the value MatchGrade
calculates automatically, you can use this to
fine-tune the results.

Frames.
lock lockSaturation disabled Lock the saturation value to prevent

MatchGrade from recalculating it when
Recalculate CDL is pressed.

Frames.
error cdlError N/A Displays the error for the CDL transform. The
smaller the value, the better the results.
This value cannot be edited.
REFERENCE GUIDE
339
MatchGrade |

Frames.
Recalculate calculateCDL N/A Click this to recalculate the CDL transform that
CDL best fits the 3D LUT. CDL Output values that
have lock enabled are not recalculated.
This allows you to manually set some values and

automatically estimate others. For example, you
can adjust and lock slope and offset and then
click this button to automatically recalculate
power and saturation.

Frames.
Export
LUT output file outfile N/A The file path and name of the 3D LUT file to
export. The file extension is .csp.

when Transform is set to 3D LUT and
Frames.
Write writelut N/A Click this to export the 3D LUT to the .csp file
specified in the LUT output file field.
REFERENCE GUIDE
340
MatchGrade |

Frames.
Create createOCIOFile- N/A Click this to create an OCIOFileTransform node

OCIOFile- Transform that you can use elsewhere in the script to apply
Transform the same grade.

Frames.
Create createOCIOCDL- N/A Click this to create an OCIOFileTransform node

OCIOCDL- Transform that you can use elsewhere in the script to apply
Transform the same grade. The button behaves differently
depending on the Transform control's setting:
• 3D LUT - Use the LUT output file and Write
controls to write the LUT to a .csp file and
then click Create OCIOFileTransform to
create an OCIOFileTransform node that
applies the grade.
• CDL - Click Create OCIOCDLTransform to
create an OCIOCDLTransform node that
applies the CDL grade.
before render beforeRender none These functions run prior to starting rendering
in execute(). If they throw an exception, the
render aborts.
before each beforeFrameRender none These functions run prior to starting rendering
frame of each individual frame. If they throw an
exception, the render aborts.
REFERENCE GUIDE
341
MinColor |
frame rendering. They are not called if the render
aborts. If they throw an exception, the render
aborts.
after render afterRender none These functions run after rendering of all frames
is finished. If they throw an error, the render
aborts.
render renderProgress none These functions run during rendering to

progress determine progress or failure.
MinColor
Calculates the difference between the darkest value in the input image and the target value, and then
adds the result to the image. This sets the darkest pixel to the target value.
To have MinColor calculate the darkest pixels and remember them, click Find Pixel Delta.
Inputs and Controls

Type Name
Input unnamed The image sequence to which the minimum is applied.

Value
MinColor Tab
search channels rgb Only these channels are searched for the minimum
value.
REFERENCE GUIDE
342
MinColor |

Value

modify modify all Sets the channels to which the calculated delta is
added.

target target 0.095 Sets the value to add to the calculated darkest pixel
from the input image. This sets the darkest pixel in
the image to the target value.
direction adjdir Up Sets the conditions under which the image is

modified:
• Up - only apply changes if the calculated delta is
positive.
• Down - only apply changes if the calculated delta
is negative.
• Both - apply changes if the calculated delta is
positive or negative.
delta pixeldelta 0 Displays the calculated pixel delta value.
Find Pixel N/A N/A Click to calculate the pixel delta for all or some
Delta frames.

aborts.
the render aborts.
REFERENCE GUIDE
343
Multiply |

Value


Multiply
Lets you multiply a channel’s values by a given factor, which has the effect of lightening the channel while
preserving the black point. This operation is also knows as gain.
See also ColorCorrect and Grade.
Inputs and Controls

Type Name
Input unnamed The image sequence whose channel values you want to multiply.
mask An optional image to use as a mask. By default, the multiplication is

REFERENCE GUIDE
344
Multiply |

Value
Multiply Tab
channels channels all The multiplication is only applied to these channels.

value value 0 The factor by which you want to multiply the values in
the above channels.
channel to none.

multiplication is limited to the non-black areas of this
channel.
multiplication is limited to the non-white areas of the
mask.
the mask.

mask.
none.

REFERENCE GUIDE
345
OCIOCDLTransform |

Value

here. This will simulate doing the gain adjustment

multiplication effect at 1.
OCIOCDLTransform
Applies an ASC CDL (American Society of Cinematographers Color Decision List) grade based on the
OpenColorIO Library. For more information, see http://opencolorio.org.
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the CDL transform.
mask An optional image to use as a mask. By default, the CDL transform is

or set to none.
REFERENCE GUIDE
346
OCIOCDLTransform |

Value
OCIOCDLTransform Tab
channels channels rgb The CDL transform is applied only to these channels.

slope slope 1 Adjusts the gain for the entire image.
offset offset 0 Adjusts the offset for the entire image.
power power 1 Adjusts the gamma for the entire image.
saturation saturation 1 Adjusts the saturation for the entire image.
direction direction forward Select Inverse to invert the transform.
working working_space linear Sets the color space in which the file is applied. The
space input is transformed from scene linear to the selected
working space, the file color transform is applied, and
then the result is transformed back to scene linear.
read from read_from_file disabled Enable this to load color correction information from
file a .cc or .ccc file.
file file none The ASC CDL file to use for the transform. This can be
a .cc or .ccc file.
cccid cccid none If the ASC CDL file is a .ccc (Color Correction
Collection) file, specify the cccid to look up here.
select cccid select_cccid N/A Click to open the cccid dialog and select the cccid
from the dropdown menu.
You can view the Slope, Offset, Power and

Saturation settings for the chosen .ccc from within
this dialog. These cannot be modified.
export grade export_cc N/A Exports the grade as a .cc (Color Correction) file.
as .cc
REFERENCE GUIDE
347
OCIOCDLTransform |

Value

channel to none.

areas of the mask
the mask.

mask.
none.



REFERENCE GUIDE
348
OCIOColorSpace |

Value
OCIOColorSpace
Much like the standard ColorSpace node, you can use the OCIOColorSpace node for converting an image
sequence from one colorspace to another. The OCIOColorSpace node is based on the OpenColorIO
library. For more information, see http://opencolorio.org.
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the colorspace transform.

transform is limited to the non-black areas of the mask.
or set to none.

Value
OCIOColorSpace Tab
channels channels rgb The ColorSpace transform is applied only to these

channels.
REFERENCE GUIDE
349
OCIOColorSpace |

Value

in in_colorspace linear Specifies the colorspace of your input image.
out out_colorspace linear Specifies the colorspace you want to convert the
image to.
channel to none.

areas of the mask.
the mask.

mask.
none.


REFERENCE GUIDE
350
OCIODisplay |

Value


OCIODisplay
Uses the OpenColorIO library to apply a colorspace conversion to an image sequence, so that it can be
accurately represented on a specific display device. The OCIODisplay node is based on the OpenColorIO
library. For more information, see http://opencolorio.org.
Inputs and Controls

Type Name

or set to none.
REFERENCE GUIDE
351
OCIODisplay |

(Scripting) Value
OCIODisplay Tab
input colorspace linear Specifies in the colorspace of your input sequence.

colorspace
display device display default Specifies the colorspace of the display device that will
be used to view the sequence.
view view sRGB Specifies the colorspace transform to apply to the scene
transform or image.
gain gain 1 Set the amount of exposure adjustment applied before

the display transform.
gamma gamma 1 Set the amount of gamma correction applied after the
display transform.
channel view channel_ RGB Specifies the channels to view before the display
selector transform.
layer layer rgba Specifies the layer to process. This should be a layer with
RGB data.
Context Tab
key1 key1 n/a Allows you to add variables to register certain viewer
processes by creating a custom config.ocio and
value1 value1 n/a
specifying the variable in the to_reference file
key2 key2 n/a transform.
value2 value2 n/a See Nuke's online help for more information.
key3 key3 n/a
value3 value3 n/a
key4 key4 n/a
value4 value4 n/a
REFERENCE GUIDE
352
OCIOFileTransform |
OCIOFileTransform
Uses the OpenColorIO library to load a colorspace conversion from a file (usually a 1D or 3D LUT) and
apply it. You can also load other file-based transformations, for example an ASC ColorCorrection XML. File
transforms are applied without any input or output colorspace handling. If the file expects log-encoded
pixels, but you apply the node to a linear image, you will get incorrect results. The OCIOFileTransform
node is based on the OpenColorIO library.
For more information, see: http://opencolorio.org
Inputs and Controls

Type Name


Value
OCIOFileTransform Tab
channels channels rgb The file transform is applied only to these channels.

file file none The file to use for the transform. Supported formats:
• .3dl (Flame)
REFERENCE GUIDE
353
OCIOFileTransform |

Value
• .3dl (Lustre)
• .ccc (ColorCorrectionCollection)
• .cc (ColorCorrection)
• .csp (Cinespace)
• .lut (Houdini)
• .itx (Iridas_itx)
• .cube (Iridas_cube)
• .mga (Pandora_mga)
• .m3d (Pandora_m3d)
• .spi1d (spi1d)
• .spi3d (spi3d)
• .spimtx (spimtx)
• .cub (Truelight)
• .vf (Nukevf)
direction direction forward Select Inverse to invert the transform.
interpolation interpolation linear Specifies the interpolation method. This is ignored if

the file used is not a LUT. The following interpolation
methods are listed from fastest to most accurate:
• nearest
• linear
• tetrahedral
• best
REFERENCE GUIDE
354
OCIOFileTransform |

Value
working working_space scene_linear Sets the color space in which the file is applied. The
space (linear) input is transformed from scene linear to the
specified working space, the file color transform is
applied, and then the result is transformed back to
scene linear.
channel to none.

areas of the mask.
the mask.

mask.
none.


REFERENCE GUIDE
355
OCIOLogConvert |

Value


OCIOLogConvert
Sometimes log data is required for certain processing operations, like plate resizing or pulling keys.
Usually a colorspace is provided specifically for these operations. The OCIOLogConvert node uses the
OpenColorIOlibrary to convert from compositing_log to scene_linear, or the reverse. For more
information, see http://opencolorio.org.
Inputs and Controls

Type Name

transform is limited to the non-black areas of the mask.
or set to none.
REFERENCE GUIDE
356
OCIOLogConvert |

Value
OCIOLogConvert Tab
channels channels rgb The colorspace conversion is applied only to these

channels.

operation operation log to lin Specifies the operation to perform:

• log to lin - Performs a log to linear colorspace
conversion.
• lin to log - Performs a linear to log colorspace
conversion.
channel to none.

areas of the mask.
the mask.

mask.
none.
REFERENCE GUIDE
357
PLogLin |

Value



PLogLin
Converts between logarithmic and linear encoding (colorspace) based on a single gray point rather than
the standard Cineon formula with a black point and white point (the formula that the Log2Lin node uses).
This method is frequently referred to as Josh Pines log conversion.
Like the Log2Lin node, you can use this node to override Nuke’s default Cineon-conversion. When doing
so, ensure that the Read or Write node is not applying an automatic colorspace conversion based on its
assumptions about the file format. You can do this by setting colorspace on the Read or Write to linear
or enabling its raw option. PLogLin uses the formula posted on the EXR list at:
http://lists.gnu.org/archive/html/openexr-devel/2005-03/msg00006.html
See also Log2Lin.
REFERENCE GUIDE
358
PLogLin |
Inputs and Controls

Type Name
Input unnamed • If you are using this node to convert your input file to linear, attach
the Read node here.
• If you are using this node to convert to logarithmic for output,
attach the node whose result you want to output here.

Output • If you are using this node to convert your input file to linear, attach
any nodes you want to use to process the image here.
• If you are using this node to convert to logarithmic for output,
attach the Write node here.

Value
PLogLin Tab
channels channels all The PLogLin effect is only applied to these channels.

operation operation log to lin Select the operation you want the node to perform:
• log to lin - convert the input from logarithmic to
linear colorspace.
• lin to log - convert the input from linear to
logarithmic colorspace.
REFERENCE GUIDE
359
PLogLin |

Value
linear linref 0.18 Set this to the linear value that corresponds with the
reference log reference value.
value
For example, you could leave this at 0.18 (mid-gray)
and enter the logarithmic value that corresponds with
0.18 in the log reference value field.
log reference logref 445 Set this to the logarithmic value that corresponds with
value the linear reference value.
negative ngamma 0.6 The film response gamma value.

gamma
density per density 0.002 The change in the negative gamma for each log space
code value code value.
channel to none.

areas of the mask.
the mask.

mask.
none.
REFERENCE GUIDE
360
Posterize |

Value


here. This will simulate doing the colorspace
conversion before the premultiplication was done. It
is the same as adding an Unpremult node before this
node and a Premult node after, but allows you to
work faster if you’re only using one color correct
node.

Posterize
Posterize reduces the color resolution of an image. This creates abrupt changes from one tone to another,
producing a dramatic graphic effect.
REFERENCE GUIDE
361
Posterize |
Inputs and Controls

Type Name
Input unnamed The image sequence to posterize.
mask An optional image to use as a mask. By default, the posterize effect is

or set to none.

Value
Posterize Tab
channels channels all The posterize effect is only applied to these channels.

Colors Colors 16 The bit depth to use. For example, setting this control
to 8 independently limits each affected channel to
256 colors.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the posterize

REFERENCE GUIDE
362
RolloffContrast |

Value
posterize effect is limited to the non-white areas of

the mask.
the mask.

mask.
none.


here. This will simulate doing the posterization before

posterize effect at 1.
RolloffContrast
Applies a contrast curve, with smooth falloff at the low and high ends, equally across all channels in the
input image. The center value controls the overall brightness of the colors and the contrast value
increases or decreases the contrast of the image. Use soft clip to control the amount of rolloff and check
override values to adjust the tension and extent of the soft clip curve.
REFERENCE GUIDE
363
RolloffContrast |
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the rolloff.
mask An optional image to use as a mask. By default, the contrast changes

are limited to the non-black areas of the mask.

Value
RolloffContrast Tab
channels channels rgba The effect is only applied to these channels.

range range 1 Values between 0 and the input value use a lookup
table to reduce processing time.
contrast contrast 1 Higher values increase the contrast. Values closer to 0

reduce the contrast.
center center 0.5 The center value is located at the center of the
contrast curve by default and controls the overall
brightness of the colors.
soft clip soft_clip 0 The amount of roll off from the linear curve near the
ends, softening the effect of the contrast.
override value_override disabled If enabled, custom values below override the

values computed values.
REFERENCE GUIDE
364
RolloffContrast |

Value
slope mag 1 slope_mag_low1 0.80000001 A value from 0 to 1 that controls where the interior
low shoulder control points are in the bezier.
slope mag 2 slope_mag_low2 0.0025 A value from 0 to 1 that controls where the interior
low shoulder control points are in the bezier.
blank point black_low 1 A value from 0 to 1 that controls the location of the
percentage black point. Computed as the distance between the
origin and the contrast line zero crossing.
slope mag 1 slope_mag_high1 1 A value from 0 to 1 that controls where the interior
upper control points are in the upper bezier.
shoulder
slope mag 2 slope_mag_high2 0.80000001 A value from 0 to 1 that controls where the interior
upper control points are in the upper bezier.
shoulder
white point white_high 1 A value from 0 to 1 that controls the location of the
percentage white point. Computed as the distance between the
1,1 point and the contrast line y = 0 crossing.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the contrast

inject inject disabled Copy the mask input to the predefined mask.a
invert invert_mask disabled Invert the use of the mask channel, so that the contrast
effect is is limited to the non-white areas of the mask.
fringe fringe disabled Blur the edges of the mask.
(un)premult N/A disabled Enable the associated channel to the right. Disabling
none.
REFERENCE GUIDE
365
Sampler |

Value


here. This will simulate adjusting the contrast before

invert invert_unpremult disabled Invert the use of the (un)premultiply channel.
mix mix 1 Dissolve between the original image at 0 and the full
effect at 1.
Sampler
This node converts a horizontal scanline of pixel data into a lookup curve. The Viewer sampler widget or
sampler y coordinate selects the scanline (sampler x is ignored).
The Sampler node can be useful for visualizing the effects of color corrections on the image.
Inputs and Controls

Type Name
Input unnamed The image sequence from which the sample is taken.
REFERENCE GUIDE
366
Saturation |

(Scripting) Value
Sampler Tab
channels channels rgb Loo kup curves are only generated for these channels.

channels.
sampler xy sampler 0, 0 Sets the position of the sampler widget that determines
the scanline to convert into a lookup curve.
Note: The sampler x coordinate is ignored.
normalize x normalize_x enabled When enabled, the lookup curves are normalized.
Sample sample_ N/A Click to sample the current frame and update the lookup
current frame current curve output for the current sampler point.
[lookup lut N/A Displays the scanline specified by the sampler y

curves] coordinate as lookup curves. You can also view
individual pixel values by mousing over the Viewer.
Saturation
For the times when you just want to correct an image’s saturation (color intensity) and don’t require
limiting the correction to any particular range of hues, use the Saturation node.
If you do want to adjust the levels of saturation in a range of hues (for example, to diminish green, blue, or
redscreen spill), use the HueCorrect node instead.
REFERENCE GUIDE
367
Saturation |
Inputs and Controls

Type Name
Input unnamed The image sequence whose saturation you want to adjust.
mask An optional image to use as a mask. By default, the saturation

adjustment is limited to the non-black areas of the mask.
or set to none.

Value
Saturation Tab
channels channels all The saturation adjustment is only applied to these

channels.

saturation saturation 1 Sets the saturation for the input image.
A value of 1 equals no change. A value of 0 produces

a grayscale image.
channel to none.

saturation adjustment is limited to the non-black
REFERENCE GUIDE
368
SoftClip |

Value
saturation adjustment is limited to the non-white
areas of the mask.
the mask.

mask.
none.


here. This will simulate adjusting the saturation before

saturation adjustment at 1.
SoftClip
Compresses high-dynamic-range (HDR) imagery into the 0-1 range. Because simple color compression
changes the apparent chroma and contrast of the image, the conversion control allows for several
compression modes that try to preserve colors.
REFERENCE GUIDE
369
SoftClip |
Inputs and Controls

Type Name
Input unnamed The HDR image sequence to compress into the 0-1 range.

(Scripting) Value
SoftClip Tab
conversion conversion none Sets the color compression mode:

• none - no compression is applied.
• preserve hue and brightness - if r, g, or b are greater
than softclip max, the saturation is reduced to get it
below softclip max while preserving the brightness.
• preserve hue and saturation - if r, g, or b are greater
than softclip max, the brightness is reduced to get it
below softclip max.
• logarithmic compress - compress all components
between softclip min and max through a logarithmic
curve between softclip min and 1.0.
softclip min softclip_min 0.8 Sets the lower clipping limit. All values above softclip
min are compressed.
Note: When using logarithmic compress, this

value should be set to 0 or 1 to avoid artefacts
for high softclip max values.
softclip max softclip_max 1 Sets the upper clipping limit. RGB values between
softclip min and this value are compressed to be
between softclip min and 1.
REFERENCE GUIDE
370
Toe |
Toe
Toe lifts the black level, in a similar way to gain controls in other color correction nodes, but with a rolloff
so that whites are mostly not affected. This makes the shadows in your image lighter.
Inputs and Controls

Type Name
Input unnamed The image sequence whose black levels you want to lift.


Value
Toe Tab

lift lift 0 Changes the black values to this value, without

affecting any original white values of the image.
channel to none.
REFERENCE GUIDE
371
Toe |

Value

channel.
invert invert disabled Inverts the use of the mask channel so that the color
mask.
the mask.

mask.
none.


here. This will simulate applying the color correction

REFERENCE GUIDE
372
Truelight |
Truelight
FilmLight and Foundry have elected to no longer bundle Truelight as of Nuke 11. The Truelight node has
been superseded by the free version of Baselight for Nuke which itself can import a BLG (Baselight Linked
Grade) that can contain a Truelight strip, plus optionally a lot more in terms of creative grading intent.
Together with OpenColorIO and the native Nuke Vectorfield node we feel all the Nuke color management
issues for which the Truelight node was originally intended are now addressed.
Vectorfield
Vectorfield applies color values from lookup tables (LUTs), allowing you to make complex color
corrections quickly using pre-calculated files. You can create LUT files using the GenerateLUT node or a
third-party application capable of producing a LUT format the Vectorfield node supports.
For more information, see CMSTestPattern and GenerateLUT.
Inputs and Controls

Type Name

(Scripting) Value
Vectorfield Tab
vectorfield vfield_file none The path and file name of the LUT file that describes the
file color correction to apply.
reload reload N/A Reloads the vectorfield file.
REFERENCE GUIDE
373
Vectorfield |

(Scripting) Value
file type file_type auto The file format of the LUT file to apply. The following
file extensions are recognized:
• auto - Nuke attempts to guess the file type from the
vectorfield file’s extension.
• .3dl - Autodesk and ASSIMILATE Scratch LUT format
(3D).
• .blut - Side Effects Houdini binary LUT format (3D).
• .cms - Light Illusion LightSpace CMS LUT format (1D
and 3D).
• .csp - Rising Sun Research CineSpace LUT format (3D).
• .cub - FilmLight Truelight LUT format (1D and 3D).
• .cube - IRIDAS LUT format (1D and 3D).
• .vf - Nuke's native LUT format used by the Vectorfield
node (3D).
• .vfz - Gzip compressed .vf files.
interpolation interpolation trilinear field If all possible input points and their values were stored
in a 3D LUT, the LUT would soon become too large. To
combat this, only some points and their values are
sampled and saved and the rest are interpolated. This
control lets you select the interpolation method to use:
• step field - This returns the value of the nearest voxel
with no interpolation. This option should only be used
for testing, as it is likely to produce artifacts.
• linear staggered vector field - Instead of having the
REFERENCE GUIDE
374
Vectorfield |

(Scripting) Value
color vectors stored in the center of each voxel, the

components of the color vectors are treated as being
stored on the faces of the voxels. This can be very
useful in fluid simulation. However, unless you
specifically generate the 3D LUT with this
interpolation in mind, you should not use this option.
• trilinear field - This is a smooth quasilinear
interpolation between the nearest voxel neighbors to
the queried point.
• cubic field - This option is like trilinear field but uses
a larger stencil and tries to interpolate more smoothly
between values using cubic interpolation.
GPU gpuExtrapolate enabled Controls whether the graphics processing unit (GPU)
extrapolate implementation extrapolates the grid values or clamps
to the maximum value of the LUT. In most cases, this
parameter should be on.
colorspace in colorspaceIn linear Determines how to convert from Nuke’s linear

colorspace to the input required by the 3D LUT:
• linear
• sRGB
• rec709
• Cineon
• Gamma 1.8
• Gamma 2.2
• PanaLog
• REDLog
• ViperLog
• AlexaV3LogC
• PLogLin
• SLog
• REDSpace
REFERENCE GUIDE
375
Vectorfield |

(Scripting) Value
colorspace colorspaceOut linear Determines how to convert from the 3D LUT's output to
out Nuke’s linear colorspace. See colorspace in for the
conversion type available.
3DL Settings (These controls are only visible when you’ve selected 3dl in the file type dropdown menu.)
output bit output_ auto Select the bit depth to scale output values for .3dl files.
depth (.3dl) bitdepth_id For instance, a 10 bit file will have values between 0 and
1023. In general this will match the input depth.
• auto - use header information to determine the
output bit depth and, failing that, use a heuristic to
guess. The heuristic assumes no values are outside the
range 0-1.
• inputbitdepth - use the input bit depth.
• 4 bit - use a 4 bit depth.
• 8 bit - use an 8 bit depth.
normalize normalize_ disabled Normalize 3dl output values to the range 0-1. This is
output output legacy behavior in Nuke which normalizes the output
range if it exceeds 1. Where the output scale is
incorrect, output bit depth should be used instead to
scale the data.
3dl file format_3dl_id auto Select the format you want to use for your .3dl file.
format • auto - use header information to determine the
format. Since it is not always possible to determine
the format from the header, you should select the
correct format where possible. The format can always
be determined for files generated by Nuke.
• autodesk flame/smoke - use Autodesk preferred
cube sizes of 17x17, 33x33, and 65x65.
• autodesk lustre - use Autodesk preferred cube sizes
of 17x17, 33x33, and 65x65, and output a
REFERENCE GUIDE
376
Vectorfield |

(Scripting) Value
3DMESH/Mesh header.
• assimilate scratch - scale and offset the 3D LUT. This
should be used only for legacy scratch files prior to
Flame compatibility.
REFERENCE GUIDE
377
Filter Nodes
Filter nodes contain convolve filters, such as blur, sharpen, edge detect, and erode.
Blur
Adds blur to an image or matte using box, triangle, quadratic, or gaussian filter algorithms. The blur
value is calculated for image pixels by examining their neighbors within the constraints of the size control
(in pixels), and applying the selected algorithms. The default, gaussian, produces the smoothest blur, but
takes longer to render.
You can also create this node by pressing B on the Node Graph.
Inputs and Controls

Type Name
Input unnamed The image sequence to blur.
mask An optional image to use as a mask. By default, the blur effect is


Value
Blur Tab
REFERENCE GUIDE
378
|

Value
channels channels all The blur effect is only applied to these channels.

size size 0 Sets the radius within which pixels are compared to
calculate the blur. Higher values widen the compare
area, producing more blur.
filter filter gaussian Select the filter algorithm to use:

• box
• triangle
• quadratic
• gaussian
Note: Box blur is the fastest to render,

whereas Gaussian blur is the smoothest.
quality quality 15 Controls the trade off between faster processing time
at lower values, and smoothness at higher values.
The image is scaled down so the blur filter is no larger

than this number, and then scaled up with linear
interpolation after filtering.
crop to crop enabled When enabled, the bounding box cannot grow past its
format incoming size. This avoids creating oversized
bounding boxes that can take longer to compute.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the filter

REFERENCE GUIDE
379
Bilateral |

Value

invert invert_mask disabled Inverts the use of the mask channel so that the filter
the mask.

mask.
blur effect at 1.
Bilateral
A bilateral filter is a smoothing filter that is particularly good at preserving edges, though it can be
computationally expensive.
Inputs and Controls

Type Name
Input guide An optional image to compute color similarity while filtering.
Connecting the guide input converts the output from the filter to
the same format as the guide image, resulting in a joint bilateral
resampling filter. If the source and guide images have the same
format, the node acts as a cross bilateral filter.
If the guide input is not connected, the output format is equal to

that of the source input and acts as a standard bilateral filter.
mask An optional image to use as a mask. By default, the filter effect is
REFERENCE GUIDE
380
Bilateral |

Type Name
unnamed The image sequence to filter.

Value
Bilateral Tab
when:
the Preferences.


effect.
REFERENCE GUIDE
381
Bilateral |

Value

gpu option.

system.
Channels channels all The filter is only applied to these channels.

Guide guide all When the guide input is connected, sets the channels
Channels in the guide image used to compute the filtering
weights.
The guide image should contain some prior

information on the source useful to the filtering, such
as a depth channel. If no information is available, the
result of an unguided C_Bilateral filter can be used as
guide input of a second filtering operation.
Size size 3 Sets the height and width of the area within which
pixels are compared to calculate the blur. Higher
values widen the compare area, producing greater
effect.
Positional positionalSigma 0.4 Controls the amount that nearby pixels affect the
Sigma overall blur.
Colour Sigma colourSigma 0.4 Controls the amount of blur that occurs over edges in
the image, lower values preserve edges more reliably.
REFERENCE GUIDE
382
BumpBoss |

Value
Filter filter gaussian Sets the type of color weighting applied to the
output:
• guassian - weighted to differences in the
surrounding pixel color and how close those pixels
are to one another.
• median - weighted according to the middle value
pixel.
Mask N/A disabled Enables the associated mask channel to the right.
channel to none.

Inject inject disabled Copies the mask input to the predefined mask.a
Invert invert_mask disabled Inverts the use of the mask channel so that the filter
Fringe fringe disabled When enabled, only apply the effect to the edge of
the mask.

mask.
Mix mix 1 Dissolves between the original image at 0 and the full
effect at 1.
BumpBoss
BumpBoss produces an emboss effect by overlaying input A over input B, rather than offsetting a single
input image like the Emboss node. The BumpBoss properties can be modified by the offset between the
center and lightposition indicators in the Viewer.
See also Emboss.
REFERENCE GUIDE
383
BumpBoss |
Inputs and Controls

Type Name
Input A The emboss texture to overlay input B.
B The image sequence to emboss with input A.


Value
BumpBoss Tab
channels channels all The emboss effect is only applied to these channels.

light position lightposition 480, 360 Sets the coordinates of the lightposition indicator in
xy the Viewer. The position of this indicator in relation to
the center indicator controls the shadow applied to
the emboss.
Note: You can also move the lightposition

indicator manually.
Bump Size bumpsize 1 Sets the size of the emboss applied to the image
connected to input B.
Light Height height 1 Controls the height of the lightposition indicator,
REFERENCE GUIDE
384
BumpBoss |

Value
where zero is level with the image, producing no

emboss.
Note: Negative values position the light ’behind’ the

image.
Light intensity 1 Sets the intensity of the simulated light.

Intensity
Minimum minshadow 0 Sets the minimum shadow intensity limit. Increasing

Shadow this value decreases the amount of shadow allowed,
Intensity and vice versa.
Image Center center 320, 240 Sets the coordinates of the center indicator in the
xy Viewer. The position of the lightposition indicator in
relation to this indicator controls the shadow applied
to the emboss.
Note: You can also move the center

indicator manually.
channel to none.

the mask.
REFERENCE GUIDE
385
Convolve |

Value
mask.
effect at 1.
Convolve
Convolve lets you create custom filter effects by supplying your own filter image. This image is used as
the convolution matrix. In other words, the new value of a pixel is calculated by centering the filter image
on the pixel, examining its neighbors, multiplying each pixel value by the corresponding pixel values in
the filter image, and then adding the results together. This allows you to defocus a clip and create lens
blur effects in the shape of the filter image, for example.
See also Blur and Defocus.
Inputs and Controls

Type Name
Input filter The filter image. This image is used as the convolution matrix.
Note that you don’t necessarily need to crop the filter image to a
smaller size, as Fast Fourier Transforms are used to speed up
convolutions with large filter images.
image The image sequence to receive the convolution filter effect.

REFERENCE GUIDE
386
Convolve |

Value
Convolve Tab
channels channels all The filter effect is only applied to these channels.

when:
the Preferences.


effect.
REFERENCE GUIDE
387
Convolve |

Value

gpu option.

system.
filter channel filter_channel rgba.alpha The channel to use as the convolution matrix from the
filter input.
This control is only available if use input channels is

disabled.
use input use_input_ disabled When enabled, Convolve uses the same channels from
channels channels the filter input as the image input (that is, whatever
channels is set to).
filter bounds filter_bounds shape Sets what to use as the filter bounds:
• shape - The filter input’s bounding box. In this case,
Convolve only uses the bounding box area, and the
center of the filter is the center of the bounding box.
This is the default value, and you may want to use it
if your filter input is a roto shape with a small
bounding box that doesn’t fill the entire format area,
for example.
• format - The filter input’s format. In this case,
Convolve uses the entire format area, allowing you
to offset the filter image within the format.
normalize normalize enabled When enabled, the filter input is divided by the sum
of all the pixels in it before using it. This ensures that
the convolution doesn’t change the overall brightness.
REFERENCE GUIDE
388
Defocus |

Value
channel to none.

For example, you can use a depth channel here to

simulate depth-of-field blurring.
the mask.

mask.
filter effect at 1.
Defocus
Defocuses the image using a disc filter. This allows you to simulate the defocusing effect of a circular lens
and create lens blur effects like ’bokeh’ (the blooming of highlights when defocused).
To create lens blur effects that aren’t circular, use the Convolve node.
To simply blur an image or a matte, use the Blur node, which is faster than Defocus.
REFERENCE GUIDE
389
Defocus |
Inputs and Controls

Type Name
Input unnamed The image sequence to defocus.

Value
Defocus Tab
channels channels all The defocus effect is only applied to these channels.

defocus defocus 1 Sets the size of the defocus disc used.
aspect ratio ratio 1 Adjusts the width of the defocus disc without altering
its height.
scaling scale 1 Scales the defocus disc simultaneously on the x and y

axes. You can also split the scaling into separate width
and height controls.
quality quality 20 Sets the interpolation used to smooth out the discs.
Higher values produce better results, but take longer
to render.
method method accelerated Sets the mode of operation:

• accelerated - faster than fullprecision mode,
especially for large defocus disks.
• fullprecision - delivers perfect defocus at the cost of
performance.
REFERENCE GUIDE
390
DegrainBlue |

Value
Note: Accelerated mode may generate

speckles in horizontal lines due to
imprecisions in the computer math.
Normally, those speckles are invisible to the
eye (delta < 0.00001), but they can cause
artifacts in nodes that use comparison math
(that is, if (src.red==0) ...). You can use
fullprecision to avoid this speckling, but
there is a significant render time increase.
channel to none.

the mask.

mask.
effect at 1.
DegrainBlue
This gizmo reduces grain in the blue channel, which often is where most grain occurs.
See also DegrainSimple and Denoise.
REFERENCE GUIDE
391
DegrainSimple |
Inputs and Controls

Connection Type Connection Name Function
Input unnamed The image sequence to degrain.

User Tab
Size size 8 The amount of grain to remove from the blue channel. Increasing this
value removes more grain and decreasing it leaves more in.
DegrainSimple
Lets you reduce grain in the red, green, and blue channels respectively, and allows you to mask the effect.
See also DegrainBlue and Denoise.
Inputs and Controls

Type Name
Input unnamed The image sequence to degrain.
mask An optional image to use as a mask. By default, the grain reduction is

limited to the non-black areas of the mask. At first, the mask input
appears as triangle on the right side of the node, but when you drag
it, it turns into an arrow labeled mask. If you cannot see the mask
input, ensure that the mask control is disabled or set to none.
REFERENCE GUIDE
392
Denoise |
Control Knob (Scripting) Default Function

(UI) Value
DegrainSimple Tab
channels channels all The channels to degrain. If you set this to something other
than all, rgba, or rgb, the R blur, G blur, and B blur sliders
below affect the first three channels in the selected layer.
R blur rVal 2 The amount of grain to remove from the red channel.
G blur gVal 2 The amount of grain to remove from the green channel.
B blur bVal 5 The amount of grain to remove from the blue channel, which
often is where most grain occurs.
mask N/A disabled Enable the associated mask channel to the right. Disabling this
checkbox is the same as setting the channel to none.
maskChannelInput none The channel to use as a mask. By default, the grain reduction is
inject inject disabled Copies the mask input to the predefined mask.a channel.
Injecting the mask allows you to use the same mask further
downstream.
invert invert_mask disabled Inverts the use of the mask channel so that the grain reduction
fringe fringe disabled When enabled, only apply the effect to the edge of the mask.
When disabled, the effect is applied to the entire mask.
mix mix 1 Dissolves between the original image at 0 and the full grain
reduction at 1.
Denoise
The Denoise node is an efficient tool for removing noise or grain from your footage. It uses spatial or
temporal filtering to remove noise without losing image quality.
REFERENCE GUIDE
393
Denoise |
Inputs and Controls

Type Name
Input Motion An optional input for attaching pre-calculated motion vectors, such as
those created by VectorGenerator.
Denoise can generate motion vectors internally, but connecting this

input may produce significantly faster results.
Noise An optional input for attaching a separate noise file. If you’re working
with complex footage that doesn’t have a convenient area for
analyzing noise, you can use the Noise input to analyze noise from
another image and apply the results to your footage.
Source The image sequence to denoise.

Value
Denoise Tab
available when:
the Preferences.

REFERENCE GUIDE
394
Denoise |

Value

effect.

gpu option.

system.
Source type Digital Choose Film or Digital according to what type of

footage you’re working with.
Noise Model model Modulated Choose what type of noise you’re dealing with:
• Constant - try this if you’re working on film footage
with lots of detail and not too much noise in dark
regions, or digital footage with lots of detail and
not too much noise in light regions.
• Modulated - for most cases, this is a good option.
Lift Blacks lift_blacks disabled When enabled, blacks in the image are lifted toward
white. This can be particularly useful if your image
contains a lot of sub-blacks, areas of black with a
value less than 0.
Profile profile Constant Choose the denoise profile to use:
REFERENCE GUIDE
395
Denoise |

Value
• Constant - let Denoise look at the analysis region

and remove the same amount of noise across all
intensities. Constant is nearly always a good choice,
but if you find that your denoise results are too
smooth and adjusting DenoiseAmount and RollOff
doesn’t help, you should try setting this to
Automatic.
• Automatic - let Denoise estimate a noise profile
automatically based on the entire ProfileFrame. This
way, Denoise removes different amounts of noise
from the shadow, midtone, and highlight areas of
the Source image. When you first switch to
automatic profiling, Denoise uses the current frame
to calculate the profile. If you’d like to use a
different frame, you need to set Profile Frame on
the Profile tab to that frame and click the
Recalculate Profile button.
Denoise always bases the noise profile on your

Source footage, even if you’ve attached another clip
to the Noise input.
You can use the TuneProfile controls to tweak both

Constant and Automatic profiles.
Output output Result Whether to output the denoised image or the noise
that was removed:
• Result - output the denoised Source image.
• Noise - output the noise that was removed from the
Source image. Only noise should be visible in this
image. If you can see a lot of picture detail, it means
the current settings are making Denoise work too
hard and remove too much of the image, which
leads to a soft result.
Denoise amount 1 Sets the overall amount of noise to remove. Increase

Amount this to remove more noise, or decrease to keep more
REFERENCE GUIDE
396
Denoise |

Value
detail. A value of 0 removes no noise.
Roll Off rolloff 2 Sets the smoothness of the denoise thresholding.
A value of 1 equals hard thresholding. Any other value

produces soft thresholding between the Denoise
Amount and the Roll Off multiplied by Denoise
Amount.
Smoothness smoothness 1 Sets the smoothness of the denoised image. This

affects the underlying image rather than the noise
detected.
In most cases, the default value of 1 works fine.

However, you can try using this control if you’re not
getting the correct smoothness level by adjusting the
Denoise Amount.
Luminance lumablend 0.7 Blends the denoised luminance with the image’s
Blend original luminance, bringing back some of the image
detail in the result.
You might want to have this set to 1 when you’re

working on denoising the footage, but for the final
result, you’ll want to decrease it.
Preserve preserve_edges disabled When enabled, Denoise attempts to sharpen the

Edges image at edges to prevent over-smoothing, though
this can emphasize noise in some cases.
Temporal Processing
Enable enable_temporal disabled When enabled, uses frames either side of the current
frame to perform temporal frame blending, which
may produce a better denoise result.
Note: The Motion input must be

connected to enable Temporal Processing.
REFERENCE GUIDE
397
Denoise |

Value
The number of frames used either side of the current

frame is determined by the Temporal Frame Offset
control.
When disabled, uses only the current frame to

produce the denoise result.
Frame frame_blending 0.1 Sets the variance allowed between regions in blended
Blending frames. Higher values attempt to blend more regions,
but can lose image detail.

Temporal Processing is enabled.
Frames to frames 1 Sets the number of frames, on either side of the

Blend current frame, used to calculate the denoise effect.

Temporal Processing is enabled.
Noise Analysis
Lock Noise lockNoiseAnalysis disabled When enabled, no analysis is performed. This is

Analysis activated by default if you use the Import button to
read a noise profile from an external file.
Analysis analysisRegion N/A Enter the coordinates for the region from which you
Region want to analyze noise.

the image and the left side of the analysis box.
• y - the distance (in pixels) between the bottom
edge of the image and the bottom edge of the
analysis box.
REFERENCE GUIDE
398
Denoise |

Value
the image and the right side of the analysis box.

• t - the distance (in pixels) between the bottom edge
of the image and the top edge of the analysis box.
• w - the width of the analysis box. This is only
available if you click the wh button.
• h - the height of the analysis box. This is only
available if you click the wh button.
You can also use the analysis box in the Viewer to

position the analysis region. This should be a flat area
free from image detail, so no textures, edges, or
shadows. The minimum size for the analysis region is
80x80 pixels. If the analysis region is too small,
Denoise doesn’t analyze the footage or remove any
noise.
By default, whenever the analysis box is altered, the

internal analysis of the noise in that region reoccurs.
Analysis analysisFrame 1 The frame at which to analyze the noise in the analysis
Frame region.
This field is automatically updated if you scrub to a

new frame in the Viewer and reposition the analysis
box.
Analyze analyze N/A Click this button to analyze the noise in your footage.
Noise This can be useful if you scrub to a new frame, don't
move the analysis box, and want to reanalyze the
noise from the new frame.
Analysis File analysisFile N/A The name and location of an external analysis file.
Click the Export button to save the analysis profile in
this file, or Import to read the analysis profile from
this file.
The analysis file can be any binary file - for example a

plain text file (.txt).
REFERENCE GUIDE
399
Denoise |

Value
Import import N/A Reads an analysis profile from an external analysis file
(specified in Analysis File). This disables any controls
that are read from the analysis file. To re-enable them,
you can uncheck Lock Noise Analysis.
Export export N/A Saves an analysis profile to an external analysis file

(specified in Analysis File).
Note: If you’ve set Profile to Constant,

only the controls that affect the analysis are
saved in this file. By contrast, if you’ve set
Profile to Automatic, both the analysis
profile and the automatically calculated
noise profile are exported.
Tune Frequencies
Process High processHigh enabled Enable this if you want to remove noise from high
Frequencies frequencies (fine detail).
High Gain highAmount 1 Scales the threshold for fine noise removal.
Increase this value to remove more fine noise, or

decrease it to keep more detail and fine noise. A
value of 0 means no fine noise is removed, whereas 1
is the estimated threshold.
Process processMedium enabled Enable this if you want to remove noise from medium
Medium frequencies.
Frequencies
Medium Gain mediumAmount 1 Scales the threshold for medium noise removal.
Increase this value to remove more medium noise, or

decrease it to keep more detail and medium noise. A
value of 0 means no medium noise is removed,
whereas 1 is the estimated threshold.
Process Low processLow enabled Enable this if you want to remove noise from low
REFERENCE GUIDE
400
Denoise |

Value
Frequencies frequencies (coarse detail).
Normally, most of the noise occurs in the high and

medium frequencies, so often you can disable the low
frequency altogether.
Low Gain lowAmount 1 Scales the threshold for low noise removal.
Increase this value to remove more low noise, or

decrease it to keep more detail and low noise. A value
of 0 means no low noise is removed, whereas 1 is the
estimated threshold.
Process Very processVeryLow disabled Enable this if you want to remove noise from very low
Low frequencies (very coarse detail).
Frequencies
Normally, most of the noise occurs in the high and
medium frequencies, so often you can disable the
very low frequency altogether.
Very Low veryLowAmount 1 Scales the threshold for very low noise removal.
Gain
Increase this value to remove more very low noise, or
decrease it to keep more detail and very low noise. A
value of 0 means no very low noise is removed,
whereas 1 is the estimated threshold.
Tune Channels
Luminance lumaAmount 1 Scales the denoising threshold for the luminance

Gain channel. Increase this value to remove more noise, or
decrease it to keep more detail and noise.
Chrominance chromaAmount 5 Scales the denoising threshold for the chrominance

Gain channel. Increase this value to remove more noise, or
decrease it to keep more detail and noise.
Profile Tab
curve editor profileCurve N/A The noise profile curve. The x axis represents image
intensity, from dark areas on the left to lighter areas
REFERENCE GUIDE
401
Denoise |

Value
on the right. The y axis represents the relative amount

of noise removed.
You can adjust the curve manually by dragging the

points on the curve to a new location. To add more
points to the curve, Ctrl/Cmd+Alt+click on the curve.
If you are not happy with your changes, click Reset

Profile to reset the curve to its original shape.
reset N/A N/A Resets any changes you’ve made to the noise curve.
This does not affect changes made using the Low
Gain, Mid Gain, and High Gain controls.
This button has the same functionality as the Reset

Profile button below.
Tune Profile tuneProfile disabled Enable this to tune the noise profile by adjusting Low
Gain, Mid Gain, or High Gain. This works in both the
Constant and Automatic profiling mode.
Low Gain lowGain 1 Scales the denoising threshold in the low light areas
of the Source image. For example, a value of 2
multiplies the threshold by 2. Everything below the
threshold is considered noise and removed, while
everything above the threshold is kept.
This control is only available when Tune Profile is

enabled.
Mid Gain midGain 1 Scales the denoising threshold in the midtone areas

enabled.
High Gain highGain 1 Scales the denoising threshold in the highlight areas
REFERENCE GUIDE
402
DirBlur |

Value


enabled.
Profile Frame profileFrame 1 The frame at which to estimate the noise profile when
Profile is set to Automatic.
This control is read only. If you want to change the

profile frame, you need to scrub to a new frame in the
Viewer and click Recalculate Profile.
Recalculate reprofile N/A If you’ve set Profile to Automatic, Denoise first uses
Profile the current frame to calculate the profile. If you’d like
to use a different frame, you need to scrub to that
frame and click this button.
DirBlur
This node applies three types of directional blur to the input image: zoom, radial, and linear.
REFERENCE GUIDE
403
DirBlur |
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the blur effect.


Value
DirBlurWrapper Tab
channels channels all The blur effect is only applied to these channels. You
should leave this set to all and select the channels you
want to blur using BlurLayer.
BlurType BlurType zoom Select the blur type to apply:

• zoom - smears pixels in all directions from a center
point. You can use this to simulate light rays, or to
transition between one clip and another.
• radial - smears pixels around a center point. You can
use this to simulate motion blur on a rotating object.
• linear - smears pixels in a given direction. You can
use this to simulate motion blur on a moving object.
BlurLayer BlurLayer rgb The blur effect is only applied to these channels. This
control works together with the channels control. For
a channel to be blurred, it needs to be selected from
both the channels and the BlurLayer dropdown
menu.
REFERENCE GUIDE
404
DirBlur |

Value
Holdout N/A disabled Enables the associated holdout channel to the right.
Channel Disabling this checkbox is the same as setting the
channel to none.
This control only has an effect when BlurType is set to

zoom.
holdout none Pixels corresponding to non-black pixels in the

holdout channel are NOT averaged into the final pixel
value.
This control only has an effect when BlurType is set to

zoom.
BlurCenter xy BlurCenter 320, 240 Sets the blur origin or center on the xy axes, or drag
the BlurCenter handle in the Viewer.
When using linear blur, BlurCenter only has an effect

if UseTarget is enabled.
BlurLength BlurLength 5 The length (in pixels) of the blur applied.
When using zoom blur, negative values reverse the

blur direction.
BlurAngle BlurAngle 45 Determines the direction of linear blur when

UseTarget is disabled.
Linear Target target 480, 240 When using linear blur with UseTarget enabled, a
xy vector is created between this and BlurCenter. This
vector then determines the direction of linear blur.
You can also adjust Linear Target by dragging the

target handle in the Viewer.
UseTarget UseTarget disabled When enabled, the direction of linear blur is

determined by a vector between BlurCenter and Blur
REFERENCE GUIDE
405
DirBlur |

Value
Target.
If this is disabled, the direction of linear blur is

determined by Blur Angle.
Samples Samples 1 The number of samples. You can increase this value for
smoother blur, but note that it takes much longer to
render.
PixelAspect PixelAspect 1 Pixel aspect ratio. If you’re working on non-square

pixel or anamorphic footage, you need to set this
value correctly to avoid unwanted distortions.
Quality Quality 1 The scaling factor of the blur. The minimum value is 1
and produces the highest quality.
Pixel Offset pixeloffset 0 When BlurType is set to zoom, lets you offset pixels
along the direction of the blur.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the blur is

invert invert_mask disabled Inverts the use of the mask channel so that the blur is
the mask.

mask.
blur effect at 1.
REFERENCE GUIDE
406
DropShadow |
DropShadow
This gizmo creates a drop shadow for any input image that has an alpha channel with values greater than 0.
Inputs and Controls

Type Name
Input input The input image from which shadows are cast.
bg The background image to merge over using the specified merge

operation.

Value
User Tab
enable drop enable_ enabled When enabled, the drop shadow effect is applied.
shadow dropshadow_
effect
inherit input inherit_input_ disabled When enabled, the drop shadow uses the input
color color color as the shadow and ignores the color control.
When disabled, the shadow color is determined by

the color control.
color color 0, 0, 0 When inherit input color is disabled, controls the

color used for the drop shadow effect.
opacity opacity 0.7 Controls the opacity of the shadow effect.
angle dropshadow_ 225 Controls the direction in which the shadow is cast,
angle where 0/360° is equal to left and 180° is right.
REFERENCE GUIDE
407
EdgeBlur |

Value
distance dropshadow_ 5 Determines the distance from the input that the
distance shadow falls, in pixels.
softness softness 0 Controls how much blur/feather is applied to the

shadow effect.
shrink/expand size 0 Controls the size of the effect by eroding (negative

values) or dilating (positive values) the shadow.
input operation over Sets the merge operation used when comping the
input over the background.
EdgeBlur
Blurs detected edges within the matte specified in the matte channel.
Inputs and Controls

Type Name
Input unnamed The image sequence to blur.

REFERENCE GUIDE
408
EdgeBlur |

Value
EdgeBlur Tab
channels channels all The blur effect is only applied to these channels.

matte input N/A enabled Enables the associated mask channel to the right.
channel to none.
control channel rgba.alpha Sets the channel containing the alpha used to detect
the edges.
size size 3 Sets the size of the blur applied to the detected
edges.
filter filter gaussian Select the filter algorithm to use:

• box
• triangle
• quadratic
• gaussian
Note: Box is the fastest to render, whereas

Gaussian blur is the smoothest.
quality quality 15 Sets the quality of the applied filter. Lower values
produce faster renders at the expense of smoothness,
and vice-versa.
Note: The image is scaled down so the blur

filter is no larger than this number, and then
scaled up using linear interpolation after
filtering.
crop to crop enabled When enabled, if the bounding box equal to or less
REFERENCE GUIDE
409
EdgeBlur |

Value
format than the format, don't enlarge it past the format. This
avoids creating oversized images that you don't use.
edge mult edge_mult 2 Controls the sharpness of the edge blur. Lower values
produce sharper edges and vice versa.
tint tint 1 Sets the multiplier for the rgb channels in the edge
blur.
brightness brightness 1 Sets the multiplier for all channels in the edge blur.
channel to none.
output none Sets the output channel for any blur applied.
channel to none.

invert invert_mask disabled Inverts the use of the mask channel so that the blur is
the mask.

mask.
blur effect at 1.
REFERENCE GUIDE
410
EdgeDetect |
EdgeDetect
A basic edge detection node with options for a blur before the edge detection and an erode after.
Inputs and Controls

Type Name
Input unnamed The image sequence to detect edges in.

(Scripting) Value
EdgeDetectWrapper Tab
Edge edgedetector sobel Choose an edge detection method:

Detector • simple
• sobel
• prewitt

channels.
Threshold threshold 0
channel to none.
output none Sets the output channel for the edge detect applied.
channel to none.
REFERENCE GUIDE
411
EdgeExtend |

(Scripting) Value
maskedgedetect none The channel to use as a mask. By default, the blur is

PreBlur/PostErode
Erode size erodesize 0
Blur size blursize 0 Lower numbers make the calculations faster but the
result might not be as smooth. The image size is scaled
down so that the blur filter is no larger than this
number, and then scaled up with linear interpolation
after filtering.
Blur quality blurquality 15 Set the blur quality. The lower the number, the faster
the calculation.
EdgeExtend
EdgeExtend allows you to correct unpremultiplied foreground color at the edge of soft mattes by eroding
or dilating the sample region, pulling pixels from deeper inside or outside your matte. The algorithm
behind the node works independent of the size of the fill regions you are trying to extend. You can also
output and edge mask to allow you to work on the edges separate from the rest of the image. A typical
use case might be adjusting a matte to account for motion blur.
Inputs and Controls

Type Name
Input Matte Controls which areas of the image are affected using the alpha
channel or luminance.
Source The image sequence to which the effect is applied.
REFERENCE GUIDE
412
EdgeExtend |

Value
EdgeExtend Tab
Not available when:
in the Preferences.
there is not enough free memory available on the
GPU.

dropdown.

takes effect.

gpu option.
REFERENCE GUIDE
413
EdgeExtend |

Value

operating system.
Channels channels rgb The effect is only applied to these channels.

Matte matte Source Where the foreground matte is taken from to limit
Alpha the effect when eroding or dilating the edges:
• Matte Luminance - use the luminance of the
Matte input.
the Matte input.
Source Is sourceIsPremultiplied disabled When enabled, EdgeExtend assumes that the image
Premultiplied connected to the Source input is premultiplied.
When disabled, no premultiplication is assumed.
Erode erode 0 Controls the amount of erosion or dilation applied

to the Matte area. Negative values dilate the matte
and positive values erode the matte.
Detail detailAmount 1 Controls how much of the original high frequency

Amount detail and noise is recovered after eroding or
dilating the matte region.
Output
Edge Mask edgeMaskChannel disabled When enabled, output an edge mask to the
REFERENCE GUIDE
414
Emboss |

Value
selected channel. This allows you to work on the

mask separately to the rest of the image and then
merge the result back into the pipeline.
Premultiply premultiply enabled When enabled, EdgeExtend premultiplies the

output image.
When disabled, no premultiplication is applied.
Emboss
Produces an emboss effect by offsetting the original input using the Angle and Width controls, rather
than overlaying a second input like the BumpBoss node.
See also BumpBoss.
Inputs and Controls

Type Name
Input unnamed The image sequence to emboss.

REFERENCE GUIDE
415
Emboss |

(Scripting) Value
Emboss Tab

Emboss Type optype traditional Sets the type of emboss output you want:
• traditional - applies the emboss effect to the input
image.
• effect - outputs the effect only, making it easy for
you to composite with other footage.
Angle Angle 0 Sets the angle of the emboss effect in degrees.
Width Width 1 Set the width of the emboss outlines. A low value gives
a more subtle, finely detailed result.
channel to none.
maskedgedetect none The channel to use as a mask. By default, the effect is

Edge edgedetector sobel Choose an edge detection method:

Detector • simple
• sobel
• prewitt
edgechannels edgechannels rgb The edge detection is only applied to these channels.

Threshold threshold 0
REFERENCE GUIDE
416
Emboss |

(Scripting) Value
channel to none.
output none Sets the output channel for the edge detect applied.
channel to none.

PreBlur/PostErode
Erode size erodesize 0
Blur size blursize 0 Set the blur size. Lower values make the calculations
faster but the result might not be as smooth. The
image size is scaled down so that the blur filter is no
larger than this number, and then scaled up with linear
interpolation after filtering.
Blur quality blurquality 15 Set the blur quality. The lower the number, the faster
the calculation.
channel to none.

invert invert_mask disabled Inverts the use of the mask channel so that the effect is
fringe fringe disabled When enabled, only apply the effect to the edge of the
mask.
When disabled, the effect is applied to the entire mask.
REFERENCE GUIDE
417
Erode (blur) |

(Scripting) Value
effect at 1.
Erode (blur)
Similar to Erode (filter), but smoother, input pixels are filtered relative to the size control. Negative values
cause brighter areas to expand into darker areas and positive values cause darker areas to expand into
lighter areas - particularly useful with mattes. Additionally, you can add blur to the input using the blur and
quality controls.
Note: Solid areas of any color other than 0 or 1 change their color as they are treated as the
anti-aliased edge. This can be corrected by using a different algorithm, such as the Erode (filter)
node.
Inputs and Controls

Type Name
Input unnamed The image or matte to erode/blur.

Value
Erode Tab
channels channels none The effect is only applied to these channels.

REFERENCE GUIDE
418
Erode (fast) |

Value
size size -1 Adjusts the size of pixels within the channel defined
by the channels control. Negative values increase the
light pixels and vice versa.
blur blur 0 Blurs the edges of the generated mask.
quality quality 15 Quality of the initial blur. Smaller values take less time
to process.
channel to none.

invert invert_mask disabled Inverts the use of the mask channel, so that the effect
the mask.

mask.
effect at 1.
Erode (fast)
Also known as the Dilate node, Erode (fast) controls the relative amount of dark (negative size values) and
light (positive size values) pixels in the input. This can be used to grow or shrink mattes, for example, an
alpha matte’s pixels can be given a positive size value to cover more area.
Erode (fast) is similar to Erode (filter), but computationally less expensive.
REFERENCE GUIDE
419
Erode (fast) |
Inputs and Controls

Type Name
Input unnamed The image or matte to erode.

Value
Dilate Tab

size size 0 Adjusts the size of pixels within the channel defined
by the channels control. Positive values increase the
light pixels and vice versa.
channel to none.

the mask.

mask.
effect at 1.
REFERENCE GUIDE
420
Erode (filter) |
Erode (filter)
Filters input pixels relative to the size control, and is particularly useful with mattes. Negative values cause
brighter areas to expand into darker areas and positive values cause darker areas to expand into lighter
areas. Use the filter dropdown to control the erode computation speed (box) versus quality (gaussian).
Erode (filter) is similar to Erode (fast), but can be computationally more expensive because you can select
the filter type to improve the erode quality.
Inputs and Controls

Type Name
Input unnamed The image or matte to erode.

Value
FilterErode Tab

by the channels control. Values less than 1 increase
the light pixels and vice versa.
filter filter box Select the filter algorithm to use:

• box
• triangle
• quadratic
• gaussian
REFERENCE GUIDE
421
Glow |

Value

channel to none.

the mask.

mask.
effect at 1.
Glow
Causes bright areas in an image to appear brighter by adding glow using a blur filter.
REFERENCE GUIDE
422
Glow |
Inputs and Controls

Type Name
Input unnamed The image sequence you want to apply glow to.

Value
Glow Tab
channels channels rgb The effect is only applied to these channels.

channels.
tint tint 1 Sets the color of the glow effect.
tolerance tolerance 0 Sets the low threshold below which no glow is

applied.
brightness brightness 1 Sets the brightness of the glow effect.
saturation saturation 1 Sets the color saturation value for any glow effect.
Width N/A disabled

channel
W none
non linear nonlinear disabled
effect only effect_only disabled When enabled, only the effect is passed downstream
allowing you to use the effect on other footage.
size size 15 Adjusts the size of pixels within the channels defined
the dark pixels and vice versa.
filter filter gaussian Sets the filter algorithm to use:

• box
• triangle
REFERENCE GUIDE
423
Glow |

Value
• quadratic
• gaussian

n/a quality 15 Sets the quality of blur to apply. The image is scaled
down, so that the blur filter is no larger than this
number, and then scaled up using linear interpolation
after filtering.
Note: Lower quality values process more

quickly, but the result is not as smooth.
crop to crop enabled When enabled, if the bounding box is at or inside the
format format, don't enlarge the effect past the format. This
avoids creating oversized images that aren’t used.
channel to none.

the mask.

mask.
REFERENCE GUIDE
424
GodRays |

Value
effect at 1.
GodRays
Averages together many samples of the image between its initial position and the translate and center
indicators in the Viewer to produce lighting and blur effects.
• The translate indicator controls the direction of the rays created.
• The center indicator controls the center point for rotate, scale, and skew functions to produce motion
blur.
• Add color and gamma changes, or use a mask input to control the output.
Note: Currently translate and skew do not correctly merge with other controls, such as scale -
they must be done individually.
Inputs and Controls

Type Name
Input unnamed The image sequence to which you want to apply godrays.

Value
Godrays Tab

REFERENCE GUIDE
425
GodRays |

Value
translate xy translate 0,0 Sets the position of the translate widget in the
Viewer which controls the direction of the rays or blur.
Straight-line motion blur can be simulated by

positioning the translate widget along the direction of
motion.
rotate rotate 0 Controls the direction of rotational rays or blur in

relation to the center widget. Negative values
produce clockwise rotation and vice versa.
You can also use this in conjunction with the scale

control to create spirals.
scale scale 1 Controls the direction of zoom in relation to the

center widget. Values less than one zoom out and
vice versa.
You can also use this in conjunction with the rotate

control to create spirals.
skew skew 0 Controls the direction of skew in relation to the

center widget
center center 1024, 778 Sets the coordinates for the center widget in the
Viewer. This is the center for rotation, scale, and
skew.
from color from_color 1 Sets the color by which the initial image is multiplied.
to color to_color 1 Sets the color by which the final image is multiplied.
gamma gamma 1 Sets the gamma space in which colors are

interpolated.
steps steps 5 Sets the number of steps used to create the rays. The
more steps you use and the shorter, and therefore
smoother, the rays appear.
REFERENCE GUIDE
426
Inpaint |

Value
max max disabled When enabled, only the brightest result of each image
is output rather than the average.
channel to none.

the mask.

mask.
effect at 1.
Inpaint
Nuke's Inpaint is a time saving node for removing unwanted elements, such as tracking markers, blemishes,
or wires. Inpaint uses surrounding pixels to fill an area marked in the alpha channel of the source image or
Matte input. The Stretch controls bias the inpainting in a defined direction and the Detail controls allow
for greater control and cloning of high frequency textures from another part of the source image, or even
from a different image using the Detail input. Inpaint also benefits from GPU acceleration to provide fast
results.
REFERENCE GUIDE
427
Inpaint |
Inputs and Controls

Type Name
Input Detail The image sequence from which detail is applied to the inpainted
area.
Matte Controls which areas of the image are affected using the alpha
channel or luminance.
Source The image sequence to which the effect is applied.

Value
Inpaint Tab
when:
the Preferences.


effect.
REFERENCE GUIDE
428
Inpaint |

Value

gpu option.

system.
Channels channels rgb The effect is only applied to these channels.

Fill Region fillRegion Source Sets the Matte channel used to inpaint the Fill Region.
Alpha Source pixels are sampled from outside this region
and extended into the Fill Region.
the Source input.
input.
the Matte input.
Smoothness smoothness 1 Controls the smoothness of the inpainted area. Higher

values can help to reduce flickering between frames,
but at the expense of local detail.
REFERENCE GUIDE
429
Inpaint |

Value
L*a*b* LABColourSpace disabled When enabled, converts the inpainting to the LAB
Colorspace colorspace, which can improve the result.
Stretch
Amount stretchAmount 0 Sets the amount of stretch applied to the edge pixels
in the direction indicated by the Direction control.
Direction stretchDirection 0 Sets the direction of stretch in degrees when Amount

is set to any value greater than 0. You can use the
direction to align linear features in the inpainted area,
such as road markings or brick work.
Detail
Source detailSource Source Sets where detail is recovered from, the Source input
or the Detail input. This allows you to recover detail
from another image entirely, such as a grain sample.
Amount detailAmount 0 Controls the amount of high-frequency detail to add

back into the inpainted area, where 0 is none and 1 is
all the detail from the detail source. Values greater
than 1 multiply the detail.
An example of detail you might want to recover is

grain, which can be erased by inpainting.
Translate translate 0,0 Controls the xy coordinates from which detail is

recovered. If the Detail input is a different format to
the Source input, you might not get the results you
expect, but you can use the Detail Center to correct
the offset of the Viewer widget.
Rotate rotate 0 Sets the rotation of the detail recovered, in degrees,

relative to the Detail Center controls. Adjusting the
Detail Center changes the relative rotation.
Scale scale 1 Sets the scale of the detail recovered relative to the
Detail Center controls. Adjusting the Detail Center
changes the relative scale.
REFERENCE GUIDE
430
Laplacian |

Value
Center center dependent Sets the center of the detail so that changes to the
on input translation, rotation, and scale are relative to the
format or Detail Center.
compositing
environment This control defaults to the center of the Source
Project format or the Project Settings > full size format
Settings control if no Source is connected.
Laplacian
The Laplacian node takes the input image, applies a blur, and then subtracts the original from the blurred
input to produce an image useful for edge detection or motion estimation. Apply different smoothing
filters to the output to trade off between speed (box) and quality (gaussian).
Inputs and Controls

Type Name
Input unnamed The image sequence to which you want to apply the filter.

Value
Laplacian Tab

REFERENCE GUIDE
431
Laplacian |

Value

• box
• triangle
• quadratic
• gaussian

n/a quailty 15 Controls the level of filtering quality. Entering a low

number runs the calculation more quickly, but the end
result isn’t quite as smooth.
The image is scaled down during the calculation, so

that the blur filter is no larger than this number, and
then scaled up with linear interpolation after filtering.
crop to crop enabled When enabled, crop the resulting image to the format.
format If the bounding box is at or inside the format, don’t
enlarge it past the format. This avoids creating
oversized images that you don’t need.
channel to none.

REFERENCE GUIDE
432
LevelSet |

Value
the mask.

mask.
effect at 1.
LevelSet
Within a single channel determined by the channel control, LevelSet outputs the distance from a pixel in
the Viewer to the nearest pixel in the input channel with the value set in the threshold control.
LevelSet can be used to make high-quality dilate or unpremultiply operations. When gradient out is set
to rgba, LevelSet outputs the motion vectors of pixels in the Viewer using the red (x axis) and green (y
axis) values.
Inputs and Controls

Type Name
Input unnamed The input sequence to dilate or unpremultiply.

(Scripting) Value
LevelSet Tab
channel channel rgba.alpha The levelset is only calculated for this channel.
threshold threshold 0.5 Sets the pixel value that the Viewer pixel distance is
measured from.
REFERENCE GUIDE
433
Matrix |

(Scripting) Value
level set out N/A enabled Enables the associated channel to the right. Disabling
this checkbox is the same as setting the channel to none.
output rgba.alpha Write distance from the threshold crossing to this

channel.
gradient out gradient motion Write gradient vectors to these channels. These vectors
point toward the threshold crossing.
create matte create_matte disabled Output a matte channel rather than a level set.
dilate matt_limit 0 Output matte is the level set thresholded at this value.
This is a good way to grow a matte. Negative values
shrink the matte.
extrapolate extrapolated none If the creatematte button is turned on and

channels extrapolatechannels is set to something other than
none, the colors at the edge of the original matte are
extrapolated as the matte is expanded.
Unless gradientextrapolate is enabled, this function

simply copy uses the color of the closest edge pixel.
gradient gradient_ disabled Instead of copying the nearest pixel, colors are
extrapolate extrapolate extrapolated as a gradient to produce a ramp.
Matrix
Adds a user defined matrix to the input allowing you to multiply all or specific channels by the matrix
values. You can copy and paste an existing matrix, using the Animation menu, or adjust individual fields
manually.
The matrix controls how each pixel in the source image is affected by those around it, for example,
-1 -1 -1
-1 8 -1
-1 -1 -1
REFERENCE GUIDE
434
Matrix |
produces an edge detection filter (see EdgeDetect) by multiplying the center pixel by 8 and the
surrounding pixels by -1, and then adding the nine values together to calculate the new value of the center
pixel. This process is repeated for every pixel in the source image to create the output image.
Note: Matrix results that fall outside the 0-1 range are clipped.
Inputs and Controls

Type Name
Input unnamed The image to pass through the matrix.

Value
Matrix Tab

matrix matrix N/A Defines the matrix used to multiply all or specific
channels. You can copy and paste an existing matrix,
using the Animation menu, or adjust individual fields
manually.
normalize normalize disabled When enabled, the matrix is divided by a constant so

that it’s sum is 1 (where possible). This causes the
output image to have the same brightness as the
source image.
channel to none.
REFERENCE GUIDE
435
Median |

Value
the mask.

mask.
effect at 1.
Median
Input image pixels within the filter area, defined by the size control, are sorted into order and the median
pixel is output to the Viewer. Median filter are very good at removing single pixel noise, with only a slight
loss of sharpness as a side effect.
Inputs and Controls

Type Name
Input unnamed The image to apply the median filter to.
REFERENCE GUIDE
436
Median |

Value
Median Tab

size size 3 Defines the filter area, in pixels, to apply the median
across.
ignore top ignore_top_line disabled When enabled, the top line of pixels is ignored within
line the filter box to account for different area calculation
methods.
Note: Enable this control for backward

compatibility to pre 6.2v1 versions of Nuke.
channel to none.

the mask.

mask.
effect at 1.
REFERENCE GUIDE
437
MotionBlur |
MotionBlur
MotionBlur (NukeX and Nuke Studio only) uses the same techniques and technology as the motion blur
found in Kronos to add realistic motion blur to a sequence, but presents the controls in a less complex,
more user friendly way.
Inputs and Controls

Type Name
Inputs FgVecs If the motion in your input sequence has been estimated before (for
example, using FurnaceCore’s F_VectorGenerator or third-party
BgVecs
software), you can supply one or more vector sequences to MotionBlur
to save processing time.
If you have separate vectors for the background and foreground, you
should connect them to the appropriate inputs and supply the matte
that was used to generate them to the Matte input. If you have a single
set of vectors, you should connect it to the FgVecs input.
Matte An optional matte of the foreground, which may improve the motion
estimation by reducing the dragging of pixels that can occur between
foreground and background objects.
Source The sequence to receive the motion blur effect.

Value
MotionBlur Tab
when:
REFERENCE GUIDE
438
MotionBlur |

Value

the Preferences.


effect.

gpu option.

system.

REFERENCE GUIDE
439
MotionBlur |

Value
Shutter Time shutterTime 0.75 Sets the equivalent shutter time of the retimed
1/48th of a second.

horizontally across the screen. The figures below show
how Shutter Time affects the retimed rectangle.
Method motionEstimation Dependent Sets the method of calculating motion estimation

on script vectors:
• Local - uses local block matching to estimate motion
vectors. This method is faster to process, but can
lead to artifacts in the output.
• Regularized - uses semi-global motion estimation
to produce more consistent vectors between
regions.
REFERENCE GUIDE
440
MotionBlur |

Value
Note: Scripts loaded from previous versions

of Nuke default to Local motion estimation
for backward compatibility. Adding a new
MotionBlur node to the Node Graph defaults
the Method to Regularized motion
estimation.
Vector Detail vectorDetail 0.2 Varies the density of the vector field. The larger vector
detail is, the greater the processing time, but the more
detailed the vectors should be.
A value of 1 generates a vector at each pixel, whereas

a value of 0.5 generates a vector at every other pixel.
• Bilinear - the default filter. Faster to process, but can
produce poor results at higher zoom levels. You can
use Bilinear to preview a motion blur before using
one of the other resampling types to produce your
output.
Matte matteChannel None Where to get the (optional) foreground mask to use
Channel for motion estimation:
the Source input.
input.
the Matte input.
REFERENCE GUIDE
441
MotionBlur2D |
MotionBlur2D
MotionBlur2D does not produce motion blur independently. It actually collates UV motion information
from the 2D transf input (for example, a Transform node) to output UV vectors down the node tree to a
VectorBlur node, producing motion blur based on these vectors.
MotionBlur3D does a similar job, but for camera moves instead of 2D transforms.
See also MotionBlur, Transform, Tracker, and VectorBlur.
Inputs and Controls

Type Name
Input 2D transf UV transform information, for example a Transform or Tracker node.
unnamed The image sequence to receive the motion blur.

Value
MotionBlur2D Tab
Output UV uv motion Sets the 2D transf input channel(s) from which the
motion vector information is derived.
shutter shutter 0.5 Enter the number of frames the shutter stays open
REFERENCE GUIDE
442
MotionBlur3D |

Value

frame 30 to 31.
frame 29 to 30.
the current frame.

field is used to set the time that the shutter opens by
adding it to the current frame. Values are in frames,
so -0.5 would open the shutter half a frame before
the current frame.
MotionBlur3D
While similar to the MotionBlur2D node in that it does not produce motion blur independently,
MotionBlur3D is designed specifically for camera moves rather than 2D transforms. MotionBlur3D collates
UV motion information from the cam input to output UV vectors down the node tree to a VectorBlur
node, producing motion blur based on these vectors.
See also Camera and VectorBlur.
REFERENCE GUIDE
443
MotionBlur3D |
Inputs and Controls

Type Name
Input cam Connect a camera to provide UV transform information.
unnamed The image sequence to receive the motion blur.

Value
MotionBlur3D Tab
Output UV uv motion Sets the cam input channel(s) from which the motion
vector information is derived.
Z input N/A disabled Enables the associated channel to the right. Disabling
none.
Z none The channel to use for Z depth information.
distance distance 1 Sets the distance to the object to blur. Higher values
cause camera translations to contribute less to
motion blur and vice versa. You can set distance to
inf to use rotations only.
Note: If a Z input channel is provided,

distance is a multiplier for the distance.
REFERENCE GUIDE
444
Sharpen |

Value

frame 30 to 31.
frame 29 to 30.
the current frame.

field is used to set the time that the shutter opens by
adding it to the current frame. Values are in frames,
so -0.5 would open the shutter half a frame before
the current frame.
Sharpen
Uses Laplacian to sharpen specific channels from the input image. Apply different smoothing filters to the
output to trade off between speed (box) and quality (gaussian).
See also Soften.
REFERENCE GUIDE
445
Sharpen |
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the sharpen effect.
mask An optional image to use as a mask. By default, the blur is limited to


Value
Sharpen Tab

minimum minimum 0
maximum maximum 1
amount amount 1

• box
• triangle
• quadratic
• gaussian
REFERENCE GUIDE
446
Sharpen |

Value



channel to none.

the mask.

mask.
effect at 1.
REFERENCE GUIDE
447
Soften |
Soften
Uses Laplacian to soften specific channels from the input image. Apply different smoothing filters to the
output to trade off between speed (box) and quality (gaussian).
See also Sharpen.
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the soften effect.


Value
Soften Tab

minimum minimum 0
maximum maximum 1
amount amount 1
REFERENCE GUIDE
448
Soften |

Value


• box
• triangle
• quadratic
• gaussian

gaussian blur is the smoothest.


channel to none.

REFERENCE GUIDE
449
VectorBlur |

Value
the mask.

mask.
effect at 1.
VectorBlur
Generates motion blur by blurring each pixel into a straight line, using the values from the motion vector
channels (u and v channels) to determine the direction of the blur. You can create motion vectors using the
VectorGenerator, MotionBlur2D, MotionBlur3D, ScanlineRender, or RayRender nodes. Many third-party
applications can also produce motion vector information as two-channel, floating point images that you
can use with VectorBlur. If possible, you should unpremultiply these images.
You can use the constant uv offset values to add motion blur for camera movement and shake. Subtract
the xy translation or camera rotation at the current frame from the same values at the next frame to get the
correct values to use.
See also VectorGenerator, MotionBlur2D, MotionBlur3D, ScanlineRender, and RayRender.
REFERENCE GUIDE
450
VectorBlur |
Inputs and Controls

Type Name
Input unnamed The image sequence to receive the blur effect.

VectorBlur Tab
If you set this to something other than all or

none, you can use the checkboxes on the right to
select individual channels.
Not available when:
in the Preferences.
the GPU.

dropdown.
REFERENCE GUIDE
451
VectorBlur |

requires you to restart Nuke before the
change takes effect.
Use GPU if useGPUIfAvailable enabled When enabled, rendering occurs on the Local

the GPU whenever the script is opened
on a machine that does have a GPU
available.
You should also select this if you wish
to render from the command line with
the --gpu option.

uv channels uv none Sets the channel(s) to control the horizontal (u)

and vertical (v) size of the blur vector.
Note: If the uv values are read from a

file, enable the Read node’s colorspace
raw data checkbox. See Read.
mv presets mv_presets Nuke Scanline Sets which renderer format to apply to the
Render motion vectors. Each preset contains a scale and
offset specific to the renderer, so that the vectors
are in the format expected by Nuke.
For example, if VectorBlur is downstream of a

PRMan node, select PRMan from the dropdown.
REFERENCE GUIDE
452
VectorBlur |
vector preprocessing
output output result Sets the output of the VectorBlur node:

• result - the final result of the blur is displayed
in the Viewer.
• uvs - the uv map used to generate the motion
vectors.
blur uv blur_uv none When set to uniform or linear, applies the

motion blur filter to the motion vectors
themselves, before using them to blur the image.
This option can smooth out gradients in the

motion vectors, resulting in a softer blur between
regions with very different motion.
• none - don't blur the UVs.
• uniform - apply a small uniform blur to soften
edges in the UV map. This is how blur was
calculated pre-Nuke 10.
• linear - apply a linearly-weighted blur to blend
between regions with different motion. This has
a stronger effect than the uniform option.
invert uv invert_uv disabled When enabled, invert the motion vectors before
blurring them.
Note: The effect is only visible when

the shutter is not centred at the current
frame (for Nuke vectors, this means
when the shutter offset is not equal to -
0.5).
uv offset uv_offset 0 Sets a constant value to add to the uv channels

values.
REFERENCE GUIDE
453
VectorBlur |
Note: If the uv values are read from a

file, this should be -0.5 or -127/255.
uv alpha N/A disabled Enables the associated alpha channel to the right.
channel to none.
alpha none If your motion vectors were premultiplied, set

this channel to the channel used to multiply the
vectors to improve the results.
Tip: Where possible, use vectors that

have not been premultiplied.
motion blur
blur type blur_type gaussian Sets the profile to use along the direction of
blur:
• uniform - a box-like profile with uniform
weighting along the blur vector.
• gaussian - a gaussian profile with a softer fall-
off towards the ends of the blur vector.
motion scale 0 Sets the value by which the u and v channels are
amount multiplied. Set this to 0.5 for motion blur from a
standard shutter.
motion motion_falloff 0.33 Controls the steepness of the gaussian fall-off on

falloff the blur. 0 gives a very sharp fall-off, while 1
gives a shallow fall-off.
motion offset -0.5 Sets the shutter offset, causing the shutter to
offset open at frame +/- motion offset.
expand blur grow_bbox 0 Sets the number of pixels added to the outside
bbox edge of image. Set this manually for cropped
images with non-zero uv channels.
soft lines soft_lines disabled When enabled, increase the softness of the
REFERENCE GUIDE
454
VectorBlur |
motion blur by drawing wider lines, at the

expense of slower processing.
blur inside blur_inside enabled When soft lines is enabled, apply a small blur
inside regions with zero motion, to give a
smoother transition to areas with motion blur.
normalize normalize enabled When enabled, normalize the results to

compensate for tearing (holes appearing behind
moving objects) and collisions (objects moving
on top of one another) in the rendered output.
Normalizing can exacerbate the effects of tearing

and collisions, but it tends to give a softer blur
than not normalizing.
channel to none.
maskChannelInput none The channel to use as the alpha channel.
effect is limited to the non-white areas of the
mask.
fringe fringe disabled When enabled, only apply the effect to the edge
of the mask.

mask.
mix mix 1 Dissolves between the original image at 0 and

the full effect at 1.
REFERENCE GUIDE
455
VolumeRays |
VolumeRays
The VolumeRays node is actually a gizmo (a group of nodes) designed to create ray lighting effects
radiating from the vol_pos indicator in the Viewer.
Inputs and Controls

Type Name
Input img The image sequence to receive the VolumeRays effect.
mask An optional image to use as a mask. By default, the effect is limited


(Scripting) Value
User Tab
Volumetric style RGB Sets the style of rays to create, dependent on the
Creation Luminance available input channels:
• RGB Luminance - creates rays based on the input
image luminance.
• Alpha Edges - creates rays based on the input image
alpha's edges. This setting has no visible effect on full
frame alphas because the rays are created outside the
format.
• Alpha Solid - creates rays based on the input image's
entire alpha.
• Shadow Making - creates rays using the input image's
REFERENCE GUIDE
456
VolumeRays |

(Scripting) Value
alpha to cast shadows.
RGB Options
Luma luma_tol 0 Sets the base luminance below which rays are not
Tolerance generated. Increasing this value eliminates rays produced
from low luminance areas of the image.
Alpha Options
Edge Size edge_size 0 Adjusts the thickness of the alpha’s edges, controlling the
amount of rays affected by the alpha. Positive values
decrease the number of rays allowed past the edges and
vice versa.
Edge Blur blur_size 3 Controls the amount of blur applied to rays at the alpha’s
edges.
Volume Options
Volumetrics vol_pos 100,100 Sets the x and y positions for the vol_pos widget in the
Center Viewer.
Ray Length raylength 20 Sets the length of the generated rays.
Pre-Ray Blur pre_blur 0 Sets the amount of blurring applied to the rays.
Quality quality Medium Controls the quality of the rays produced:

• Low
• Medium
• High
• Very High
REFERENCE GUIDE
457
VolumeRays |

(Scripting) Value
Note: Higher quality rays take longer to

render.
Add on Top comp_me disabled When enabled, the effect overlay and the input image
are passed from the node.
When disabled, only the effect is passed from the node

for compositing purposes.
Flicker Options
Use Flickering chk_flicker enabled When enabled, the Flicker Options are applied to the
rays.
Flicker Speed flicker_speed 2 Controls the speed of ray flickering. Higher values
produce more flicker.
Flicker Size flicker_size 40 Controls the size of flicker to generate artificially if the
vol_pos widget is static.
Note: If Transform Noise with Volume

Center is enabled, this control has no effect on
the rays.
Transform xform_flicker disabled When enabled, flicker noise moves with the vol_pos
Noise with widget so that flickering is the result of the FlickerSpeed
Volume only.
Center
When disabled, flicker is artificially generated using the
Flicker Size value.
Color Options
Desaturate chk_desat disabled When enabled, the input image is fully desaturated so
before that generated rays do not pickup color from the input.
Coloring
REFERENCE GUIDE
458
VolumeRays |

(Scripting) Value
Volume CCorrect1_ 1 Controls the overall amount of Gamma applied to the

Gamma gamma generated rays.
Volume Gain CCorrect1_ 1 Controls the overall amount of Gain applied to the
gain generated rays.
Initial Volume initcolor 1

Color
End Volume volume_end_ 0,0,0,0

Color color
Radial Falloff
Enable Radial chk_radial enabled When enabled, the Radial Falloff controls are applied to
Falloff the rays.
Radial Size radial_size 500
Radial Falloff rad_softness 1 Controls the amount of falloff as rays move away from
the vol_pos widget in the Viewer. Higher values cause
the rays to fade more rapidly.
Mask Options
Enable Mask chk_use_mask disabled When enabled, any image connected to the Mask input
is used as a mask.
Note: The mask must contain an alpha channel.
Mask Blur mask_blur 0 Adjusts the blur applied to the edges of the mask.
Transform chk_xform_ disabled When enabled, the mask and vol_pos widget’s relative
Mask with mask positions are maintained - any transform applied to the
Volume vol_pos widget is applied to the mask.
Center
When disabled, the mask is not transformed by vol_pos
widget.
REFERENCE GUIDE
459
ZDefocus |
ZDefocus
Blurs the image according to a depth map channel. This allows you to simulate depth-of-field (DOF)
blurring.
In order to defocus the image, ZDefocus splits the image up into layers, each of which is assigned the
same depth value everywhere and processed with a single blur size. After ZDefocus has processed all the
layers, it blends them together from the back to the front of the image, with each new layer going over
the top of the previous ones. This allows it to preserve the ordering of objects in the image.
Inputs and Controls

Type Name
Input filter This image is used as the blur kernel. It represents the shape and size
of the camera aperture used to shoot the input footage. As the clip
in the image input is blurred, any out-of-focus highlights (’bokeh’) in
the clip assume the shape of the filter image.
You can create a filter image using the Roto node (Draw > Roto) or
the Flare node (Draw > Flare), for example. The filter image can also
be a color image. For example, if you want to add color fringing to
your out-of-focus highlights to simulate chromatic aberration, you
can use the Flare node to easily create a suitable filter image.
You don’t necessarily need to crop the filter image to a smaller size,
as Fast Fourier Transforms are used to speed up convolutions with
large filter images.
image The image sequence to receive the blur effect. This should also
contain the depth map channel.

REFERENCE GUIDE
460
ZDefocus |

Type Name

Value
ZDefocus Tab

when:
the Preferences.


effect.
REFERENCE GUIDE
461
ZDefocus |

Value

gpu option.

system.
depth z_channel depth.Z Specifies the input channel containing the depth map
channel information.
Note: The depth map should not be anti-

aliased. If it is, pixels along an edge between
two objects can be assigned a depth that is
in-between the depth of the front and back
objects. This looks wrong, as it suggests that
those edge pixels are floating somewhere
between the objects.
math math depth Specifies how the depthchannel is used to calculate

the distance between the camera and an object. For
example, some programs use higher values to denote
further away, while in others they mean closer to the
camera:
• direct - The Z value in the depth channel directly
controls blur. For example, if Z is 0.5, then the blur
size will be 0.5 times the value of the size control
(unless this is bigger than maximum, in which case it
will be clamped to maximum).
• depth - The Z value in the depth channel is the
REFERENCE GUIDE
462
ZDefocus |

Value
distance between the camera and whatever is in the

image at that pixel.
• far = 0 - The Z value in the depth channel is equal
to 1/distance. The values are expected to decrease
from large positive values close to the camera to
zero at infinity. This is compatible with depth maps
generated by Nuke and RenderMan.
• far = 1 - Near plane = 0, far plane = 1. This is
compatible with depth maps generated by OpenGL.
• -direct - As with the direct mode, the Z value in the
depth channel directly controls blur. In other words,
each layer is blurred by the same amount as in the
direct mode. However, in this mode, the layers are
interpreted as being in the opposite order, so a
higher depth value places a layer in front of another
rather than behind it.
• -depth - The Z value in the depth channel is -
distance in front of the camera. This is the same as
depth, but the distances are negative to start with.
• far = -0 - The Z value in the depth channel is equal
to -1/distance. The values are expected to increase
from large negative values close to the camera to
zero at infinity. This is compatible with depth maps
generated by Maya.
• far = -1 - Near plane = 0, far plane = -1.
output output result Sets the output type:

• result - displays the input image and the result of
the blur controls.
• focal plane setup - displays depth-of-field (DOF)
information in the rgb channels:
• red - Less than DOF (in front of the area that’s in
focus).
• green - Inside DOF (in focus). Note that if depth
of field is set to 0, nothing is displayed in green.
REFERENCE GUIDE
463
ZDefocus |

Value
• blue - Greater than DOF (behind the area that’s in

focus).
• layer setup - like focal plane setup, but displays
depth-of-field (DOF) information after the depth has
been divided into layers. Pixels assigned to the same
layer have the same amount of blur applied to them.
• filter shape setup - displays the filter shape in rbg,
allowing you to adjust the filter more accurately.
show image show_image enabled When output is set to focal plane setup, enabling
this shows the depth-of-field (DOF) information
overlaid upon the input image.
fill fill_foreground enabled When enabled, Nuke attempts to compensate for

foreground missing information by filling regions in the
foreground which are revealed when the foreground
goes out of focus.
You can try enabling this control if you see sharp edge
artefacts in blurred objects in front of the focal point
(nearer to the camera). However, because the true
image information isn't available in these regions,
enabling fill foreground can sometimes introduce
undesirable artefacts by adding things which aren't
there. If you see blurry artefacts in the foreground,
rather than sharp edge artefacts, try disabling this
control.
focus plane center 0 Sets the Z depth of areas in the image that are entirely
(C) in focus.
focal point xy focal_point 200, 200 Controls the position of the focal point widget in the
Viewer. Adjusting the Viewer widget updates the
focusplane and focalpoint fields automatically.
REFERENCE GUIDE
464
ZDefocus |

Value
depth of dof 0 Sets a depth slice around the focus plane that is
field entirely in focus.
Note: True theoretical depth of field would

set this to zero.
blur inside blur_dof enabled When enabled, a small amount of blur is applied to
the in-focus region. This gives a smoother transition
between the in-focus region and the out-of-focus
regions around it.
size size 25 Sets the size of the blur at infinite depth. Blur nearer
the camera than the focus plane may be larger.
If you have set math to direct, the size is multiplied

by the depth to give the blur size at that depth.
Setting size to 1 allows you to use the values in the
depth map as the blur size directly.
maximum max_size 50 The filter size is clipped at this maximum value. No

blurring greater than this value is generated no matter
where the object is in relation to the camera.
Set this value as low as possible for maximum

processing speed.
automatic autoLayerSpacing enabled When enabled, ZDefocus automatically works out how
layer spacing many depth layers to use, based on the maximum blur
size (maximum). In this mode, the layers are closer
together near to the focal plane, where a small change
in the blur amount is more obvious, and increasingly
more widely-spaced further away.
When disabled, you can control the depth layers and

their spacing manually using depth layers and layer
curve.
depth layers nLayers 20 The number of depth layers to use for the blur. Use a
REFERENCE GUIDE
465
ZDefocus |

Value
small number of layers for maximum speed.
The maximum number of blur sizes that are used

between 0 and maximum is 256. This means you can
have up to 256 layers behind the focal plane, and up
to 256 in front of it as well.
layer curve layerCurve 1 The curve to apply to the layer spacing.
A value of 0 gives evenly spaced layers. Positive values

concentrate the layers closer to the in-focus region.
Negative values mean the layers are concentrated far
from the focal plane, towards the maximum blur size.
filter type filter_type disc Sets the blur filter applied to the image:
• disc - applies a round disc filter to the image.
• bladed - applies a bladed filter to the image
(simulates the iris blades that can make up a camera's
diaphragm).
• image - uses the image in the filter input as the blur
kernel.
The following controls, up to and including aspect ratio, are only available when filter type is set to
disc.
filter shape shape 0 Dissolves the filter shape between Gaussian at 0 and
disc at 1.
aspect ratio aspect 1 Sets the filter aspect ratio, which is 1:1 by default.
Values less than 1 squeeze the filter on the x axis, and
values larger than 1 squeeze it on the y axis.
This allows you to simulate the cat's eye effect, caused

by vignetting inherent within some lens designs.
The following controls, up to and including catadioptric, are only available when filter type is set to
bladed.
aspect ratio aspect 1 Sets the filter aspect ratio, which is 1:1 by default.
Values less than 1 squeeze the filter on the x axis, and
REFERENCE GUIDE
466
ZDefocus |

Value
values larger than 1 squeeze it on the y axis.
This allows you to simulate the cat's eye effect, caused

by vignetting inherent within some lens designs.
blades blades 5 Sets the number of iris blades that make up the
camera's diaphragm. A value of 3 produces a triangle,
4 a square, 5 a pentagon, 6 a hexagon, and so on.
Note: This field only accepts integers larger

than 1.
roundness roundness 0.2 Controls the rounding of the filter polygon’s sides,
where zero is equal to no rounding.
rotation rotation 0 Controls filter rotation in degrees. Positive values

produce counter-clockwise rotation and vice-versa.
inner size inner_size 0.8 Controls the size of the inner polygon, as a percentage
of the outer polygon.
inner feather inner_feather 1 Adds outward or inward feathering around the inner
polygon. With values larger than 0.5, your feather
effect is outward and, respectively, if your values are
smaller than 0.5, the feather effect is inward. A value of
0.5 produces no feathering.
inner inner_brightness 0.8 Controls the brightness of the inner polygon, where 0
brightness is equal to black and 1 to white.
catadioptric catadioptric_size 0.3 When catadioptric is enabled, controls the size of the
size catadioptric hole in the filter.
catadioptric catadioptric disabled When enabled, ZDefocus simulates catadioptric lenses.

This means the defocused areas of the image are
annular, producing donut-shaped bokeh.
You can use the catadioptric size to control the hole

in the center of the filter.
REFERENCE GUIDE
467
ZDefocus |

Value
The following controls, up to and including clamp image filter, are only available when filter type is set
to image.
legacy resize legacy_resize_ N/A Loading scripts from pre-Nuke 8.0v7 enables the
mode mode legacy resize mode checkbox automatically, for
backward compatibilty, and uses the filter bounds
dropdown to determine how images used in filtering
are resized.
Adding new ZDefocus nodes hides the legacy resize

mode checkbox and allows you to use the image
filter dropdown to give you more flexibility when
calculating blur.
filter channel filter_channel rgba.alpha The channel to use as the convolution matrix from the
filter input.
This control is only available if use input channels is

disabled.
use input use_input_ disabled When enabled, the same channels are used for both
channels channels the filter and image inputs.
filter bounds filter_bounds shape Sets what to use as the filter bounds when legacy
resize mode is enabled:
• shape - The filter input’s bounding box. In this case,
ZDefocus only uses the bounding box area, and the
center of the filter is the center of the bounding box.
This is the default value, and you may want to use it
if your filter input is a roto shape with a small
bounding box that doesn’t fill the entire format area,
for example.
• format - The filter input’s format. In this case,
ZDefocus uses the entire format area, allowing you
to offset the filter image within the format.
REFERENCE GUIDE
468
ZDefocus |

Value

legacy resize mode is enabled.
image filter image_filter Cubic Sets which of Nuke's filtering algorithms to use when
remapping pixels from their original positions to new
positions, when legacy resize mode is disabled:
values.
Sinc4 the most
REFERENCE GUIDE
469
ZDefocus |

Value

legacy resize mode is disabled.
clamp image clamp_image_ disabled When using filters that employ sharpening, such as
filter filter Rifman and Lanczos, you may see a haloing effect. If
necessary, check clamp image filter to correct this
problem.

legacy resize mode is disabled.
gamma bloom_gamma disabled When enabled, a gamma curve of 2.2 is applied

correction before blurring and then reversed for the final output.
This is useful for making bokeh lens shape effects

warmer and more visible.
bloom bloom disabled When enabled, highlights over the bloom threshold
are boosted to make lens shape effects more visible.
bloom bloom_threshold 0.8 When bloom is enabled, highlights above this value
threshold are multiplied by the bloom gain value to make lens
shape effects more visible.
bloom gain bloom_gain 2 When bloom is enabled, highlights above the bloom
threshold are multiplied by this value.
channel to none.

REFERENCE GUIDE
470
ZSlice |

Value
invert invert_mask disabled Inverts the use of the mask channel, so that the blur is
the mask.

mask.
effect at 1.
ZSlice
This node uses the depth channel of the input to highlight a slice of the image characterized by the same
Z depth. You can modify the slice using the Center of Slice and Field Width controls.
Inputs and Controls

Type Name
Input unnamed The input to select the Z slice from.
Note: The input must contain a Z depth channel.
mask An optional image to use as a mask. By default, the slice is limited to

REFERENCE GUIDE
471
ZSlice |

Value
ZSlice Tab
channels channels all The effect is only applied these channels.

Z N/A disabled Enables the associated Z channel to the right.

channel to none.
Zchan none The channel containing the Z depth information.
Center of center 0.5 Defines the center, or depth, of the slice within the
Slice image. High values are closer to the camera, and lower
values further away.
Field Width dof 0.1 Defines the width of the slice. Higher values highlight
a thicker slice around the specified depth, and vice
versa.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the slice is

invert invert_mask disabled Inverts the use of the mask channel, so that the slice is
the mask.

mask.
REFERENCE GUIDE
472
ZSlice |

Value
effect at 1.
REFERENCE GUIDE
473
Keyer Nodes
Keyer nodes deal with extracting procedural mattes from image sequences using luma keying, chroma
keying, and difference keying.
ChromaKeyer
ChromaKeyer is a green and bluescreen keyer that can take advantage of the local GPU. Use the screen
color selector to choose a color from the Source input to use as the blue/green screen color. To remove
blue/green spill from the foreground object, use the despill bias control to pick skin tones from the
source. Use the matte parameters to improve the matte.
Inputs and Controls

Type Name
Input Source The foreground image to key.

Value
ChromaKeyer Tab
Local GPU gpuName N/A Displays the GPU used for rendering when Use GPU
if available is enabled. Local GPU displays Not
available when:
the Preferences.
REFERENCE GUIDE
474
|

Value


effect.

gpu option.

operating system.
view _view final result Select what you want to render into the output.
• intermediate result - view the intermediate result.
Use this option on shots that can only be keyed
using several different keys on different parts of the
image (multipass keying). This renders the original
source image with the matte generated in this
node. In ChromaKeyer nodes down the tree, you
should set the inside mask control to source alpha.
• final result - this creates a premultiplied RGBA
foreground that can be composited later. You can
REFERENCE GUIDE
475
|

Value
produce unpremultiplied output by disabling the

premultiply control.
screen color _screenColor 0, 0, 0 Sets the screen color to become transparent, usually
blue or green background.
Click the color swatch next to screen color to

activate the eye dropper and then
Ctrl/Cmd+Shift+click and drag a rectangular area
over the green pixels in the Viewer. This averages the
pixels in the area selected to produce a better key.
screen gain _screenGain 1 Adjusts how much of the screen color is removed to
make the screen matte. Increasing this value keys
more.
screen _screenBalance 0.5 Sets the balance point for the image to key. If the key
balance is not working too well with the default setting, try
setting the balance to about 0.05, 0.5 and 0.95 and
see what works best.
matte
chroma gain _chromaGain 0 Controls how much of the chroma difference

between the source image and the screen color is
used to correct the matte.
Increasing chroma gain generates a matte that has

less transparent areas in the foreground, but can
produce harder edges.
white point _whitePoint 1 Sets the error threshold used for correcting the
foreground. All alpha values above this point are set
to 1.
black point _blackPoint 0 Sets the error threshold used for correcting the
background. All alpha values below this point are set
to 0.
alpha bias _alphaBias 0.5, 0.5, 0.5 Adjusts the alpha bias in case your screen color isn’t
REFERENCE GUIDE
476
|

Value
purely blue or green and is causing parts of the

foreground image to become transparent. Pick the
color from the part of the foreground that is
affected.
inside mask _insideMask ignore If you're using multiple ChromaKeyer nodes to pull
several different keys on different parts of the image
(multipass keying), set inside mask to source alpha
to add up the results of previous ChromaKeyer
nodes.
When set to ignore, any upstream keys are not taken

into account.
despill
Use custom _ disabled Enables the use of the despill bias control.
despill bias overrideDespillBias
despill bias despillBias 0.5, 0.5, 0.5 Adjusts the despill bias to remove any remaining spill
from around the foreground image. The most useful
colors to pick for despill bias are often hair colors
and skin tones.
Note: This control is disabled unless Use

custom despill bias is enabled.
replace _replace ignore Sets the color method to use for pixels where the
mode alpha of the matte has been modified:
• ignore - the despilled image is left untouched if
the alpha is modified.
• edge hard color - the despilled image has a
corresponding amount of the replace color added
for any increase in alpha.
• edge linear color - the image has a corresponding
amount of the original pixel (screen color and all)
REFERENCE GUIDE
477
Difference |

Value
reintroduced/removed if the alpha is changed.

• edge soft color - the despilled image has a
corresponding amount of the replace color added
for any increase in alpha, however, it attempts to
modulate the luminance of the resulting pixel so
that it matches the original pixel. This produces a
more subtle result than the edge hard color option.
add-in matte _replaceAlphaDiff disabled When enabled, add the replace color into areas
fix corrected using the matte black point and white
point controls to enhance the despill effect.
replace color _replaceColor 1, 1, 1 When replace mode is set to edge hard color or
edge soft color, sets the color to add where the
alpha of the matte has been increased by the mask.
replace _replaceAmount 1 Controls the amount of replace color to add in the

amount final result.
premultiply _preMultiply enabled When enabled, premultiply the matte generated by

ChromaKeyer with the foreground.
Premultiplying the matte can remove noise from the

background, but it can also be too aggressive in
some areas, such as hair.
Difference
This node produces the difference between two images as a matte. It takes two inputs:
• input B contains the subject with the background, and
• input A contains the background without the subject (a clean plate).
You can use this node to produce a rough matte for a moving subject shot in front of a static background,
for example. However, because of lighting discrepancies, grain, and tiny movements of the camera or
background, you may need to use other nodes to refine the results.
REFERENCE GUIDE
478
HueKeyer |
Inputs and Controls

Type Name
Input A The background without the subject (a clean plate).
B The subject with the background.

(Scripting) Value
Difference Tab
offset offset 0 Subtract this value from each pixel of the output.
gain gain 1 Any pixels that are white are set to this color.
output output enabled Enables the associated output channel to the right.
channel to none.
rgba.alpha Render the resulting image into this output channel.
HueKeyer
This node is a chroma keyer. It calculates hue from the incoming red, green, and blue channels, and uses
the amount curve to determine the value to output in the alpha channel. In the graph, the x axis represents
hue and the y axis the alpha value for that hue. To key out a particular hue, find the hue on the x axis and
Ctrl/Cmd+Alt+click to create a point on the curve at that location. Then drag the point up to a value of 1
on the y axis to set that hue to black in the alpha channel.
REFERENCE GUIDE
479
IBKColor |
Inputs and Controls

Type Name
Input unnamed The image sequence to key.

(Scripting) Value
HueKeyer Tab
channels channels rgb The hue is calculated from these channels.
lut lut N/A Use the amount curve to control the amount of alpha
that is output from the hue. With the sat_thrsh curve
you can adjust the saturation threshold. Increase this
threshold to reject hues with low saturation.
output output enabled Enables the associated output channel to the right.
channel to none.
rgba.alpha Render the resulting image into this output channel.
invert invert enabled Inverts the use of the alpha channel so that the result is a
matte.
IBKColor
The IBK keyer differs from many other keyers in that, instead of using a single color picker, it uses an input
image (a clean plate with just the color variations of the background) to drive the key. This generally gives
you good results when working with uneven blue- or greenscreens.
The IBK keyer consists of two nodes: IBKColor and IBKGizmo. IBKColor creates the clean plate from the
blue- or greenscreen image, and IBKGizmo pulls the key.
REFERENCE GUIDE
480
IBKColor |
Note: Pre-rendering output from IBKColor to some formats can result in loss of color
information and incorrect results when the image is used as input to IBKGizmo. We
recommended that the .exr format is used for pre-rendered images as it supports full float data,
reducing loss of information.

1. Connect the output from IBKColor to the c input of IBKGizmo.
2. Attach the input of IBKColor, along with the fg input of IBKGizmo, to the original blue- or greenscreen
image.
3. Attach the bg input of IBKGizmo to your background image.
4. Create two Viewers and view the output from both nodes side-by-side.
5. Adjust the controls of IBKColor until you get a clean plate with nothing but the background colors.
6. Adjust the controls of IBKGizmo until you are happy with the key.
7. Connect the output from IBKGizmo to a compositing node, such as Merge, to composite the
foreground over the background.
Inputs and Controls

Type Name
Input 1 The blue- or greenscreen image.
Output unnamed Connect this to the IBKGizmo node’s c input.

(Scripting) Value
Parameters Tab
screen type screen_type blue Select green or blue depending on which one you have
in the foreground image.
size Size 10 Adjust the amount of color expansion.
darks off 0, 0, 0 Adjust the color values to get the best separation
between black and the screen type color. You want to be
left with shades of the screen color and black only. Start
REFERENCE GUIDE
481
IBKGizmo |

(Scripting) Value
by bringing down the value for blue if you’re using a

blue screen, and the value for green if you have a green
screen in the image. The erode and patchblack sliders
should be set to 0 when you’re adjusting these values.
As a rule of thumb, if you have a dark green discolored

area, increase darks g. Similarly, if you have a light red
discolored area, increase the lights r value.
lights mult 1, 1, 1 Adjust the color values to get the best separation
between black and the screen type color. You want to be
left with shades of the screen color and black only. Start
by bringing down the value for blue if you’re using a
blue screen, and the value for green if you have a green
screen in the image. The erode and patchblack sliders
should be set to 0 when you’re adjusting these values.
As a rule of thumb, if you have a dark green discolored

area, increase darks g. Similarly, if you have a light red
discolored area, increase the lights r value.
erode erode 0 Increase this value if you still see traces of the
foreground edge color in the output. This might be
especially useful if, after adjusting the lights and the
darks values, you’re still left with areas of discoloration.
This is likely if your screen does not have a very
saturated hue.
patch black multi 0 After adjusting the darks and the lights, you can increase
this value to remove all the black from the output
image, if you want. This might be useful, for example, if
you get blue/green artifacts in your composite.
IBKGizmo
REFERENCE GUIDE
482
IBKGizmo |
The IBK keyer differs from many other keyers in that, instead of using a single color picker, it uses an input
image (a clean plate with just the color variations of the background) to drive the key. This generally gives
you good results when working with uneven blue- or greenscreens.
The IBK keyer consists of two nodes: IBKColor and IBKGizmo. IBKColor creates the clean plate from the
blue- or greenscreen image, and IBKGizmo pulls the key.
Note: The output from IBKGizmo is a premultiplied image with an alpha channel - it does not
composite the foreground over the background. To perform the composite, connect IBKGizmo
to a compositing node, such as Merge.
Note: Pre-rendering output from IBKColor to some formats can result in loss of color
information and incorrect results when the image is used as input to IBKGizmo. We
recommended that the .exr format is used for pre-rendered images as it supports full float data,
reducing loss of information.
Inputs and Controls

Type Name
Input bg The background image. This is used in calculating fine edge detail.
c Either:
• An IBKColor node. This creates the clean plate (that is, takes the
color that is keyed out and creates a smoother representation of
that).
• A clean plate that was shot on location. This is more accurate than
using an IBKColor node, but in most cases a clean plate isn’t
available.
If you have set screen type to pick, you do not need this input.
fg The blue- or greenscreen image.
REFERENCE GUIDE
483
IBKGizmo |

(Scripting) Value
IBK Tab
screen type st C-blue Set this according to the screen type in your foreground
image:
• C-blue - select this if your foreground image is a
bluescreen image and you want to use the c input (a
clean plate) to drive the key.
• C-green - select this if your foreground image is a
greenscreen image and you want to use the c input (a
clean plate) to drive the key.
• pick - use the color picker below to replace the c input
with a single color. This way, the IBK acts more like a
traditional keyer, such as Primatte.
color color 0, 0, 1 If you’ve set screen type to pick, use this control to pick
a color that best represents the area you are trying to
key.
red weight red_weight 0.5 Sets how the red channel is weighted in the keying
calculation. This affects the hardness of the matte in
primarily red areas.
If you check screen subtraction and view the output,

you often see discolored edges because the background
hasn’t been fully removed from the original plate. This is
not spill but the result of the matte being too strong.
Lowering one of the weights will correct the edge of the
foreground object in that color. For example, if a red
foreground object has an edge problem, you should
lower the red weight. This may affect other edges, so
it’s often a good idea to use multiple IBKGizmos with
different weights split with Keymix nodes (Merge >
Keymix).
blue/green blue_green_ 0.5 Sets how the blue or green channel is weighted in the
weight weight keying calculation. This affects the hardness of the matte
in primarily blue or green areas.
REFERENCE GUIDE
484
IBKGizmo |

(Scripting) Value
If you’re using a bluescreen, this controls the green

weight.
If you’re using a greenscreen, this controls the blue

weight.
If your foreground and background are opposite

extremes in color (for example, yellow on saturated
blue), you need to take the weight all the way down,
making everything transparent.
luminance lm_enable disabled Adds a luminance factor to the keying and helps to
match capture transparent foreground areas that are brighter
than the background.
You can also use luminance match to lessen some of

the screen area noise if you lower the screen range.
However, pushing this control too far will eat into your
foreground blacks.
You can use this to firm up the alpha channel in the

lighter areas.
screen range level 1 Lower this until it stops changing the background. If
there’s too much noise left in the backing region, you
may want to degrain the footage before pulling the key.
This may clear the noise from the backing region, but
you may also end up hardening the edges of the matte.
luminance luma 0 Allows you to control the strength of the overall effect.
level
This is usually only effective in a small number of cases,
generally you don’t have to adjust this.
enable ll_enable disabled Check this to activate the luminance level control.
autolevels autolevels disabled Can be used to reduce any hard edges from a
foreground object with saturated colors. The same can
be achieved with the weights, but the autolevels control
REFERENCE GUIDE
485
IBKGizmo |

(Scripting) Value
affects only the saturated colors whereas the weights

affect the entire image.
When using this control, it’s best to have this as a

separate node that you can then split with other
IBKGizmos as the weights controls will no longer work
as expected.
If you actually have foreground colors you want to keep,

you can check the equivalent color box to keep them.
For example, you may have a saturated red subject
against a bluescreen, which results in a magenta
transition area. Autolevels eliminates this. However, if
you have a magenta foreground object, autolevels
makes the magenta more red. To keep the magenta, you
need to check the magenta box.
This control is especially useful for when you have

supersaturated colors in your image.
yellow yellow disabled Check this to prevent autolevels from changing

saturated yellow in your foreground elements.
cyan cyan disabled Check this to prevent autolevels from changing

saturated cyan in your foreground elements.
magenta magenta disabled Check this to prevent autolevels from changing

saturated magenta in your foreground elements.
screen ss enabled To have the keyer subtract the foreground from the RGB,
subtraction check this control.
To premultiply the original foreground with the

generated matte, uncheck this control.
use bkg ublu disabled To have the bg input affect the brightness of the edge,
luminnance check use bkg luminance.
These controls are best used with the luminance match

enable slider. They can also help with screens that
REFERENCE GUIDE
486
Keyer |

(Scripting) Value
exhibit some form of fringing artifact - usually a

darkening or lightening of an edge on one of the color
channels on the screen. To offset the effect, grade the
bg input up or down with a Grade node (Color > Grade)
just before input. If it’s just an area that needs help, draw
a Bezier shape (Draw > RotoPaint) around the area and
grade the bg input up or down locally to remove the
artifact.
use bkg ubcr disabled To have the bg input affect the color of the edge, check
chroma use bkg chroma.
These controls are best used with the luminance match

enable slider. They can also help with screens that
exhibit some form of fringing artifact - usually a
darkening or lightening of an edge on one of the color
channels on the screen. To offset the effect, grade the
bg input up or down with a Grade node (Color > Grade)
just before input. If it’s just an area that needs help, draw
a Bezier shape (Draw > RotoPaint) around the area and
grade the bg input up or down locally to remove the
artifact.
Keyer
This node gives you access to several simple keyers. You can use it to pull keys based on the input image’s:
• red, green, or blue channel,
• redscreen, greenscreen, or bluescreen,
• luminance (the default),
• saturation, or
• maximum and minimum values.
To use this node, view the alpha channel in the Viewer. Then, use the range graph to adjust the low and
high pixel values of the matte. The A handle determines the low or transparent values of the key: any
REFERENCE GUIDE
487
Keyer |
pixels that fall below this value are clipped to black. The B handle determines your high or opaque values:
pixel values above this setting are clipped to white.
The default positions let you control the low and high values, assuming your image has distinct light and
dark areas. However, sometimes the subject of the matte falls into the “middle-gray” area; the C and D
handles let you shift the center for the high values of the key.
Inputs and Controls

Type Name
Input unnamed The image sequence to key.

(Scripting) Value
Keyer Tab
input input rgb Select the layer you want to use for keying.
output N/A enabled Enables the associated output channel to the right.
channel to none.
output rgba.alpha Render the resulting image into this output channel.
replace combine replace Select how you want to combine the matte with the
existing output channels.
• replace - replace the existing output channels with the
matte
• intersect - create an intersection of the matte and the
existing channels.
• union - create a combination of the matte and the
existing channels.
invert invert disabled Inverts the use of the alpha channel so that the result is a
non-white matte.
operation operation luminance key Select which keying tool you want to use for keying your
sequence:
REFERENCE GUIDE
488
Keyer |

(Scripting) Value
• red keyer - use the red channel to pull the key.

• green keyer - use the green channel to pull the key.
• blue keyer - use the blue channel to pull the key.
• redscreen - subtract the red channel from other
channels.
• greenscreen - subtract the green channel from other
channels.
• bluescreen - subtract the blue channel from other
channels.
• luminance key - use the luminance value to pull the
key.
• saturation key - use the saturation value to pull the
key.
• max keyer - use the maximum values of the red, green
and blue channels to pull the key.
• min keyer - use the minimum values of the red, green
and blue channels to pull the key.
range range N/A Drag the A, B, C, and D delimiter handles in the graph to
adjust your keying result. The A delimiter marks the
value where you want the keying to start, the distance
between B and C mark the length of the full effect and
delimiter D indicates where the effect stops.
A range 1 Enter the value for delimiter A in the range graph. You
can also drag the handle in the graph.
B range 1 Enter the value for delimiter B in the range graph. You
C range 1 Enter the value for delimiter C in the range graph. You
D range 1 Enter the value for delimiter D in the range graph. You
REFERENCE GUIDE
489
Keylight |
Keylight
Keylight is an industry-proven color difference keyer. Use the Screen Color selector to choose a color
from the Source input to use as the blue/green screen color and the View dropdown menu to judge the
key. To remove blue/green spill from the foreground object, use the Despill Bias control to pick skin
tones from the source. Use the Screen Matte parameters to improve the matte.
Keylight’s core algorithm was developed by the Computer Film Company (now Framestore) and has been
further developed by Foundry.
Copyright © 2020 Foundry & Framestore. All rights reserved.
Inputs and Controls

Type Name
Input bg The background image to replace the blue/green screen in the

foreground.
OutM The Outside Mask, or garbage matte, used to remove unwanted

objects (lighting rigs, and so on) from the foreground.
InM The Inside Mask, or holdout matte, used to confirm areas that are
definitely foreground.
Source The foreground image to key.

Value
Keylight Tab
View show Final Result Select what you want to render into the output.
• Source - view the blue/green screen
foreground image.
• Source Alpha - view the foreground image’s
REFERENCE GUIDE
490
Keylight |

Value
alpha channel.
• Screen Matte - view the matte created from
picking the Screen Color. It does not include any
inside or outside masks.
• Inside Mask - view the inside input. This is used
to firm up the foreground matte to stop print
through.
• Outside Mask - view the outside input. The
outside mask is used as a garbage mask to
reveal the background.
• Combined Matte - view the screen matte,
inside mask, and outside masks added together.
• Status - view an exaggerated view of the key so
that minor problems are shown clearly.
• Intermediate Result - view the intermediate
result. Use this option on shots that can only be
keyed using several different keys on different
parts of the image (multipass keying). This
renders the original source image with the
Screen Matte generated in this Keylight node. In
Keylight nodes down the tree, you should set
the Source Alpha in the Inside Mask folder to
Add To Inside Mask.
• Final Result - this creates a premultiplied RGBA
foreground that can be composited later.
There’s an Unpremultiply Result checkbox you
can use if you wish.
• Composite - this renders the foreground
composited over the background using all
mattes, spill and color corrections.
Unpremultiply unPreMultiply enabled When disabled, the key result is not

Result unpremultiplied. If you’re using premultiplied
images, you should leave this enabled.
REFERENCE GUIDE
491
Keylight |

Value
Screen Color screenColor 0, 0, 0 Sets the screen color to become transparent,

usually blue or green background.
Screen Gain screenGain 1 Adjusts how much of the screen color is removed
to make the screen matte. Increasing this value
keys more.
Screen screenBalance 0.5 Sets the balance point for the image to key.
Balance Generally speaking, blue screens tend to work
best with a balance of around 0.95 and green
screens with a balance of around 0.5. If the key is
not working too well with these settings, try
setting the balance to about 0.05, 0.5 and 0.95
and see what works best.
Alpha Bias alphaBias 0.5, 0.5, 0.5 Adjusts the AlphaBias in case your screen color
isn’t purely blue or green and is causing parts of
the foreground image to become transparent.
Pick the color from the part of the foreground
that is affected.
Despill Bias despillBias 0.5, 0.5, 0.5 Adjusts the DespillBias to remove any remaining
spill from around the foreground image. The
most useful colors to pick for Despill Bias are
often hair colors and skin tones.
Note: This control is disabled unless

Use Alpha Bias for Despill is disabled.
Use Alpha Bias gangBiases enabled When enabled, use the AlphaBias color as the
for Despill DespillBias color.
When disabled, set the DespillBias color

separately.
Screen PreBlur preBlur 0 Adjusts how much you want to soften the
foreground image used to generate the key. If
you’re keying digital video or grainy footage, you
REFERENCE GUIDE
492
Keylight |

Value
might get better results if you adjust this control

slightly.
Screen Matte
Clip Black screenClipMin 0 Sets the cut off point below which alpha values in
the Screen Matte are set to zero.
Note: Adjusting this control too

aggressively can cause hard edges
between the foreground and
background.
Clip White screenClipMax 1 Sets the cut off point above which alpha values in
the Screen Matte are set to the white point.
Note: Adjusting this control too

aggressively can cause hard edges
between the foreground and
background.
Clip Rollback screenClipRollback 0 Allows edge detail to be restored after clipping

the Screen Matte too aggressively using the
ClipBlack and ClipWhite controls.
Screen Dilate screenGrowShrink 0 Adjusts alpha values in the Screen Matte. Either
grow (if greater than zero) or shrink (if less than
zero) the alpha in the Screen Matte at a sub-pixel
level.
Note: This control should not normally

be used as eroding edges can produce a
very poor key.
Screen screenSoftness 0 Controls the amount of blur applied to the matte.
REFERENCE GUIDE
493
Keylight |

Value
Softness For example, softness could be used on a very

harsh matte pulled for use as an inside matte
further down the tree.
Screen screenDespotBlack 0 Removes isolated spots of black in a white matte

Despot Black by simplifying the matte. Black specks in the
white matte are absorbed by the surrounding
white areas.
Screen screenDespotWhite 0 Removes isolated spots of white in a black matte

Despot White by simplifying the matte. White specks in the
black matte are absorbed by the surrounding
black areas.
Screen screenReplaceMethod Soft Color Sets the color method to use for pixels where the
Replace alpha of the Screen Matte has been modified:
• None - the despilled image is left untouched if
• Source - the image will have a corresponding
amount of the original pixel (screen color and
all) reintroduced/removed if the alpha is
changed.
• Hard Color - the despilled image has a
corresponding amount of the Screen Replace
Color added for any increase in alpha.
• Soft Color - the despilled image has a
Color added for any increase in alpha, however,
it attempts to modulate the luminance of the
resulting pixel so that it matches the original
pixel. This produces a more subtle result than
the Hard Color option.
Screen screenReplaceColour 0.5, 0.5, 0.5 Sets the color to add where the alpha of the
Replace Color Screen Matte has been increased by the mask
when Inside Mask > Inside Replace is set to
HardColor or SoftColor.
REFERENCE GUIDE
494
Keylight |

Value
Tuning
Midtones At midPoint 0.5 Sets the base midtone level used by the Gain
controls below.
Shadow Gain lowGain 1 Adjusts the strength of the shadows in the Screen
Matte.
Midtones midGain 1 Adjusts the strength of the midtones in the

Gain Screen Matte.
Highlights highGain 1 Adjusts the strength of the highlights in the

Gain Screen Matte.
Inside Mask
Source Alpha sourceAlphaHandling Ignore This parameter determines what to do with any
embedded alpha in the original source image.
You will need this if you are doing multiple keys
on different parts of the image with the View
output set to Intermediate Result:
• Ignore - this does not add any embedded alpha
to the screen matte.
• Add To Inside Mask - the embedded alpha is
added to the inside mask. You should select this
when multipass keying with View set to
Intermediate Result.
• Normal - the embedded alpha is used to
composite the image.
Inside insideReplaceMethod Soft Color Sets the color method to use for pixels where the
Replace alpha of the Inside Matte has been modified:
• None - the despilled image is left untouched if
• Source - the image will have a corresponding
amount of the original pixel (screen color and
all) reintroduced/removed if the alpha is
changed.
REFERENCE GUIDE
495
Keylight |

Value
• Hard Color - the despilled image has a

Color added for any increase in alpha.
• Soft Color - the despilled image has a
Color added for any increase in alpha, however,
it attempts to modulate the luminance of the
resulting pixel so that it matches the original
pixel. This produces a more subtle result than
the Hard Color option.
Inside insideReplaceColour 0.5, 0.5, 0.5 Sets the color to add where the alpha of the
Replace Color Inside Matte has been increased by the mask
when Inside Mask > Inside Replace is set to
HardColor or SoftColor.
Crops
X Method SourceXMethod Color When the crop controls are in use, sets the fill
method to use on the X axis at the edges of the
clip:
• Color - uses the Edge Color to fill the cropped
area.
• Repeat - the edge pixels are used to fill the
cropped area.
• Reflect - an equal number of pixels are
reflected back from the image to fill the
cropped area.
• Wrap - an equal number of pixels are wrapped
from the opposite edge to fill the cropped area.
Y Method SourceYMethod Color When the crop controls are in use, sets the fill
method to use on the Y axis at the edges of the
clip:
• Color - uses the Edge Color to fill the cropped
area.
• Repeat - the edge pixels are used to fill the
REFERENCE GUIDE
496
Keylight |

Value
cropped area.
• Reflect - an equal number of pixels are
reflected back from the image to fill the
cropped area.
• Wrap - an equal number of pixels are wrapped
from the opposite edge to fill the cropped area.
Edge Color SourceEdgeColour 0 When an XY Method is set to Color, sets the

color used to fill the cropped area.
Left SourceCropL 0 Sets the crop amount from the left edge.
Right SourceCropR 1 Sets the crop amount from the right edge.
Bottom SourceCropB 0 Sets the crop amount from the bottom edge.
Top SourceCropT 1 Sets the crop amount from the top edge.
Keylight
InM insideComponent None Sets the component to use from any Inside Mask
Component attached to the InM input:
• Luminance - the luminance of the image
connected to the InM input is used.
• Inverted Luminance - the luminance of the
image connected to the InM input is inverted
before use as the mask.
• Alpha - the alpha of the image connected to
the InM input is used.
• Inverted Alpha - the alpha of the image
connected to the InM input is inverted before
use as the mask.
OutM outsideComponent None Sets the component to use from any Outside
Component Mask attached to the InM input:
• Luminance - the luminance of the image
connected to the OutM input is used.
• InvertedLuminance - the luminance of the
REFERENCE GUIDE
497
Primatte |

Value
image connected to the OutM input is inverted

before use as the mask.
• Alpha - the alpha of the image connected to
the OutM input is used.
• InvertedAlpha - the alpha of the image
connected to the OutM input is inverted before
use as the mask.
About about N/A Click to display a Keylight informational box.
Primatte
Primatte keys are created incrementally by sampling single pixels, or a range of pixels, from a blue or
greenscreen image. This controls the polyhedron in colorspace that determines what pixels are in or
outside the matte. To use Primatte, click Auto-Compute to automatically sense the backing screen color,
eliminate it, and even get rid of some of the foreground and background noise. Alternatively, you can set
operation to how you want to use the color samples, or use the left and right arrow buttons to step
through all the operations. Then, use Ctrl+click or Ctrl+Shift+drag to select a single color or a range of
colors from the Viewer. Primatte then uses these to adjust the polyhedron. Nuke provides unlimited Undo
during this process.
Primatte is a trademark of IMAGICA Corp. Primatte patent is held by IMAGICA Corp.
Inputs and Controls

Type Name
Input bg The background image to replace the blue/green screen in the

foreground.
fg The foreground, or subject, image to key.
REFERENCE GUIDE
498
Primatte |

Type Name
mask An optional image to use as a mask. By default, the Primatte effect is

or set to none.

(Scripting) Value
Primatte Tab
foreground foreground rgb Channels from the fg input containing the image to
key and composite.
background background rgb Channels from the bg input to composite over, and to
calculate the spill suppression.
crop crop N/A The area of the input image you want to keep.
Anything outside this box is cropped.

the original image and the left side of the crop box.
of the original image and the bottom edge of the
crop box.
the original image and the right side of the crop box.
• t - the distance (in pixels) between the bottom edge
of the original image and the top edge of the crop
box.
• w - the width of the crop box. This is only available if
you click the wh button.
• h - the height of the crop box. This is only available
if you click the wh button.
REFERENCE GUIDE
499
Primatte |

(Scripting) Value
You can also adjust the crop box in the Viewer by

dragging its edges.
channel to none.
mask none The channel to use as a mask. By default, the Primatte

Primatte effect is limited to the non-white areas of the
mask.
Initialize
algorithm algorithm Primatte Choose which algorithm Primatte uses to calculate

your keying result:
• Primatte - The Primatte algorithm delivers the best
results and supports both the solid color and the
complement color spill suppression methods. It is
the algorithm that uses three multi-faceted
polyhedrons (as described further down in the this
chapter) to separate the 3D RGB colorspace. It is also
the default algorithm mode and, because it is
computationally intensive, it may take longer to
render.
REFERENCE GUIDE
500
Primatte |

(Scripting) Value
• PrimatteRT - is the simplest algorithm and

therefore, the fastest. It uses only a single planar
surface to separate the 3D RGB colorspace (as
described further down in this chapter) and, as a
result, does not have the ability to separate out the
foreground from the backing screen as carefully as
the above Primatte algorithm. Other disadvantages
of the Primatte RT algorithm is that it does not work
well with less saturated backing screen colors and it
does not support the complement color spill
suppression method.
• Primatte RT+ - this is in between the above two
options. It uses a six planar surface color separation
algorithm (as described further down in this
document) and delivers results in between the other
two in both quality and performance. Other
disadvantages of the Primatte RT+ algorithm is that
it does not work well with less saturated backing
screen colors and it does not support the
complement color spill suppression method.
Reset Reset N/A Resets all of the Primatte properties to their initial
values.
Auto- Auto-Compute N/A This can be used as the first step in the Primatte
Compute operation. Its purpose is to automatically detect the
backing screen color, remove it, and do some clean-
up on the foreground and background noise. If the
clip was shot with an evenly lit, well-saturated backing
screen, the Auto-Compute button leaves you with an
image that may only need some spill removal to
complete your keying operation.
viewer viewer disabled Opens a Primatte Viewer that displays a graphical

representation of the Primatte algorithms and allows
you to see what is happening as the various Primatte
tools are used. It is a passive feature that has no
REFERENCE GUIDE
501
Primatte |

(Scripting) Value
adjustment capabilities, but it may prove useful in

evaluating an image as you operate on it.
Degrain
type graintype none Select the type of degrain to use:

• none - you get the color of the exact pixel sampled.
• small - gets the average color of a small region of
the area around the sampled pixel. Use this when the
grain is very dense.
• medium - gets the average color of a medium-sized
region of the area around the sampled pixel. Use this
when the grain is less dense.
• large - gets the average color of a larger region of
the area around the sampled pixel. Use this when the
grain is very loose.
tolerance graintolerance 0.2 Increases the effect of the Clean BG Noise tool
without changing the edge of the foreground object.
This allows you to tell the Primatte algorithm what
brightness of pixels you think represents grain. You
should try not to use too high of a value; otherwise, it
affects the overall matte.
Actions
operation mode Smart Select What the colors you sample in the Viewer do:
BG Color • Smart Select BG Color - gets the sampled backing
screen color, analyzes the original foreground image,
and determines the foreground areas using the
Primatte foreground detection routine. Then, using
the newly determined foreground areas, performs a
Clean FG Noise operation internally and renders the
composite image.
• Clean BG Noise - removes noise from the
background. Any pixels you sample in the Viewer are
REFERENCE GUIDE
502
Primatte |

(Scripting) Value
known to be 100% background. White noisy areas in

the 100% background region become black.
• Clean FG Noise - removes noise from the
foreground. Any pixels you sample in the Viewer are
known to be 100% foreground. The color of the
sampled pixels is registered by Primatte to be the
same color as in the original foreground image. This
makes dark gray areas in the 100% foreground
region white.
operation • Matte Sponge - sets the sampled color in the image

(continued) to be 100% foreground. However, if the sampled
color is already keyed out and removed, it leaves the
current suppressed color. It only affects the key or
matte information. This tool is usually used to quickly
remove stray transparent pixels that have appeared
during the chroma keying procedure.
• Make FG Trans. - suppresses spill in foreground
transparency. When this mode is selected, the
opaque foreground color region sampled in the
Viewer becomes slightly translucent. This operation
is useful for the subtle tuning of foreground objects
that are otherwise completely covered with smoke
or clouds.
• Restore Detail - brings back detail to the
foreground. The completely transparent background
region sampled in the Viewer becomes translucent.
This operation is useful for restoring lost hair details,
thin wisps of smoke and so on.
• Spill Sponge - removes spill from a composite
image. When this mode is selected, the background
color component in the pixels sampled in the Viewer
is keyed out and removed for the color region
selected.
operation • Spill(-) - removes color spill from the sampled pixel
REFERENCE GUIDE
503
Primatte |

(Scripting) Value
(continued) color (and all colors like it) in the amount of one
Primatte increment. If spill color remains, another
click using this tool removes more of the color spill.
• Spill(+) - returns color spill to the sampled pixel
color (and all colors like it) in the amount of one
Primatte increment. This tool is used to move the
sampled color more in the direction of the color in
the original foreground image. It can be used to
nullify a Spill (-) step.
• Matte(-) - reduces the opacity of the matte for the
sampled pixel color (and all colors like it) in the
amount of one Primatte increment. If the matte is
still too opaque, another click using this operational
mode tool makes the sampled color region even
more translucent. This can be used to thin out smoke
or make a shadow thinner to match shadows in the
background imagery.
• Matte(+) - increases the opacity of the matte for the
amount of one Primatte increment. If the matte is
still too translucent or thin, another click using this
operational mode tool makes the sampled color
region even more opaque. This can be used to
thicken smoke or make a shadow darker to match
shadows in the background imagery. It can only
make these adjustments to the density of the color
region on the original foreground image. It can be
used to nullify a Matte (-) step.
operation • Detail(-) - reduces foreground detail for the

(continued) sampled pixel color (and all colors like it) in the
amount of one Primatte increment. If detail is still
missing, another click using this operational mode
tool makes detail more visible. This can be used to
restore lost smoke or wisps of hair.
REFERENCE GUIDE
504
Primatte |

(Scripting) Value
• Detail(+) - increases foreground detail for the

amount of one Primatte increment. If there is still too
much detail, another click using this operational
mode tool makes more of it disappear. This can be
used to remove smoke or wisps of hair from the
composite. Sample where detail is visible and it
disappears. This is for moving color regions into the
100% background region. It can be used to nullify a
Detail (-) step.
• Fine Tuning Sliders - provides finer control over the
removal of spilled background color. The color of
the sampled pixel is registered as a reference color
for fine tuning. To perform the tuning operation,
select a Fine Tuning slider and move it to achieve the
desired effect.
• 3D Sample - When viewer is enabled in the Primatte
properties, the sampled colors are displayed as a
spray of white pixels in the Primatte Viewer. This
allows you to see where the selected backing screen
colors reside within the 3D RGB colorspace.
• Simple Select BG Color - gets the sampled backing
screen color, projecting a line in the opposite
direction on the hue wheel and generating artificial
pixels that may represent the FG object. Then, using
the artificially generated foreground pixels, it
performs a Clean FG Noise operation internally and
renders the composite image.
currentColor black To sample colors from the Viewer, make sure this is
pushed in (displaying an eyedropper), then
Ctrl/Cmd+click in the Viewer to pick single pixels, or
Ctrl/Cmd+Shift+drag to pick a rectangle of pixels.
adjust adjustLighting disabled When enabled, based on the currently selected BG

lighting color, Primatte generates a clean, evenly lit backing
REFERENCE GUIDE
505
Primatte |

(Scripting) Value
screen to use in the keying operation. This can

improve the results if you have a very unevenly lit
backing screen.
hybrid render hybridRender disabled When enabled, Primatte internally creates two keys
from the same image:
• core - This matte has the transparency removed, but
suffers from the bad edges on the foreground
object.
• edge - This matte has a clean edge on the
foreground, but suffers from transparency within the
foreground object.
The core matte with the bad edges is then blurred and
eroded before it is composited over the edge matte
that has the transparency, resulting in a composite with
the best of both options.
This can help you deal with foregrounds that contain a

color that is close to the backing screen color.
Adjust Lighting
threshold lightingSPoly 10 When adjust lighting is enabled, determines if a grid

pixel should be treated as a pure background sample,
a simulated background sample, or a foreground
sample. Increasing the value brings more of the
foreground into the adjusted lighting.
grid size lightingGridSize 12 When adjust lighting is enabled, sets the grid size
used. Increasing this value increases the grid resolution
used in the adjusted lighting calculation.
Hybrid Matte
erode hybridErode 5 When hybrid render is enabled, sets the amount of

erosion performed on the core matte. To view the
results, set output mode to hybrid core and view the
alpha channel.
REFERENCE GUIDE
506
Primatte |

(Scripting) Value
blur radius hybridBlurRadius 0.5 When hybrid render is enabled, sets the blur radius
used when blurring the core matte. To view the
results, set output mode to hybrid edge and view the
alpha channel.
Fine Tuning
L-poly (spill) spill 0 Adjust boundary between spill-suppressed opaque

foreground and no-change opaque foreground. The
more to the right the slider moves, the more spill is
removed. The more to the left the slider moves, the
closer the color component of the selected region is
to the color in the original foreground image.
M-poly matte 0 Adjust boundary between partially transparent

(trans) foreground and fully opaque foreground. Moving this
slider to the right makes the registered color region
more transparent. Moving the slider to the left makes
the matte more opaque.
S-poly detail 0 Adjust boundary between completely transparent

(detail) background and partially translucent foreground.
Moving this slider to the right makes the registered
color region more visible. Moving the slider to the left
makes the color region less visible.
Spill Process
replace with spillProcess complement The color spill replacement method to use:
• no suppression - no spill suppression is applied.
• complement - replaces the spill color with the
complement of the backing screen color.
• solid color - replaces the spill color with a solid
color of your choice. Use replace color below to
choose the color.
• defocused background - replaces the spill color
with colors from a defocused version of the
background image. Use defocus to adjust the
REFERENCE GUIDE
507
Primatte |

(Scripting) Value
amount of defocus applied to the background buffer

image.
replace color replaceColor 0.18 When replace with is set to solid color, allows you to
select a solid color to use to replace the spill.
defocus defocus 10 When replace with is set to defocused background,

sets the amount to defocus the background for spill
process.
Output
output mode output_mode composite Determines what the output from the node should be:
• composite - premultiplies the foreground with the
background based on the matte created. This is your
final result.
• premultiplied - outputs the premultiplied result of
the Primatte node, along with the calculated matte.
This can be useful if you want to do your
compositing using a Merge node (with operation
set to over) rather than Primatte. This allows you to
color correct, transform, and otherwise process your
image before compositing it over the background.
Note, however, that Primatte works within the sRGB
colorspace, whereas Nuke works within a linear
colorspace. This means you need to add a
Colorspace node after both Primatte and your
original background image to convert their
colorspaces to sRGB, then do your color corrections
or transforms, merge the images together, and,
finally, use another Colorspace node to convert the
result back to linear.
REFERENCE GUIDE
508
Primatte |

(Scripting) Value
• unpremultiplied - outputs the unpremultiplied

result of the Primatte node, along with the calculated
matte. This can be useful if you want to do your
compositing using a Merge node (with operation
set to matte) rather than Primatte. This allows you to
color correct, transform, and otherwise process your
image before compositing it over the background.
Note, however, that Primatte works within the sRGB
colorspace, whereas Nuke works within a linear
colorspace. This means you need to add a
Colorspace node after both Primatte and your
original background image to convert their
colorspaces to sRGB, then do your color corrections
or transforms, merge the images together, and,
finally, use another Colorspace node to convert the
result back to linear.
output mode • notpremultiplied - outputs the original foreground

(continued) pixels (instead of the result of the Primatte node),
along with the calculated matte.
• adjust lighting FG - outputs the light-adjusted
foreground that the adjust lighting mode creates
(this has a more even shade of the backing color with
the foreground object). If adjust lighting is disabled,
this option simply displays the un-optimized,
original foreground image.
• adjust lighting BG - outputs the optimized artificial
backing screen that the adjust lighting mode creates
(a clean backing screen that has no foreground
object).
• hybrid core - outputs the internally generated core
matte, used when hybrid render is enabled.
• hybrid edge - outputs the internally generated edge
matte, used when hybrid render is enabled.
output output rgba The channels to write the output image to.
REFERENCE GUIDE
509
Ultimatte |
Ultimatte
Ultimatte AdvantEdge technology is a color difference keyer. It is the culmination of over 25 years of
bluescreen imaging experience by Ultimatte Corporation. The final result is a composite that preserves
fine details like hair, smoke, and motion blurred edges with a minimum of compromises.

1. Use the screen control in the Viewer to sample the screen (backing) color near detail that you want to
retain.
2. Activate the overlay display and use the overlay + and - pickers to get the overlay to cover as much of
the screen as possible without intruding on the foreground subject area. The overlay indicates areas
that will definitely not be in the composite.
3. Turn the overlay view off, view the alpha channel, and use the matte + and - pickers to bring areas
more solidly into the matte or remove noise or other aberrations from the matte.
4. Use the hold shadow dropper (only available when screen correct and shadow are enabled) to scrub
on the shadows that you’d like to preserve. These shadows may best be seen in the foreground image.
This allows the shadows to be handled separately (and more accurately) from foreground subject areas
5. Finally, you can use the spill and match pickers to improve the color correction applied to the RGB
channels.
Ultimatte is a trademark of Ultimatte Corporation. Ultimatte patent is held by Ultimatte Corporation.
Inputs and Controls

Type Name
Input hm The holdout matte used to confirm areas that are definitely
foreground.
At first, the holdout mask input appears as triangle on the left side of
the node, but when you drag it, it turns into an arrow labeled hm.
gm The garbage matte used to remove unwanted objects (lighting rigs,

and so on) from the foreground.
REFERENCE GUIDE
510
Ultimatte |

Type Name
bg The background, or screen, image to replace the blue/green screen in

the foreground.
cp An optional clean plate input.
fg The foreground, or subject, image to key.

Value
Ultimatte Tab
screen color screen_color 0 Sets the primary color on which the screen extraction
is based. Choose an area on the blue/green screen
near important subject detail that is not obscured in
any way.
Film film disabled When enabled, the controls on the Film tab are
activated.
screen screen_correct enabled When enabled, the controls on the Screen Correct
correct tab are activated.
shadow shadow disabled When enabled, the controls on the Shadow tab are
activated.
spill spill_suppression enabled When enabled, the controls on the Spill tab are
suppression activated.
cleanup cleanup disabled When enabled, the controls on the Cleanup tab are
activated.
color color_ disabled When enabled, the controls on the Color tab are
conformance conformance activated.
overlay overlay_mode off Select overlay rendering mode:

• off - do not display an overlay.
• screen - leave the subject unchanged and display
the preliminary matte area blended with the overlay
color.
• subject - view the subject blended with overlay
REFERENCE GUIDE
511
Ultimatte |

Value
color, and leave the preliminary matte area

unchanged.
• Use the add and remove overlay tools to select
pixels that should be included in, or excluded from,
the screen correct region.
• Note: The add and remove overlay tools
accumulate scrubbed pixels in keep and remove
lists. If the lists contain equivalent values, a conflict
may occur, resulting in no visible change to the
overlay when picking. To correct this, try using Undo
to revert recent selections or the Reset button on
the ScreenCorrect tab.
show image overlay_mono disabled When overlay is set to screen or subject, enable this
as control to make the input image grayscale so that the
monochrome overlay areas are more easily distinguishable.
overlay color overlay_color 0.57999998, When overlay is set to screen or subject, sets the
0.07, 0.22, overlay color. Adjust the alpha channel to modify the
0.5 opacity of the overlay.
output mode output_mode Composite Sets the type of output to produce:

• composite - displays a composite of the screen and
subject output.
• premultiplied - displays the premultiplied screen
and subject.
• unpremultiplied - displays the unpremultiplied
screen and subject.
Density Tab
brights brights 100 Adjusts the matte density in bright foreground areas.
Low values are more transparent, high values are
more opaque.
REFERENCE GUIDE
512
Ultimatte |

Value
Note: Hard, dark edges may occur around

foreground subjects if this control is set too
high.
darks darks 0 Adjusts the matte density in black glossy or dark

foreground objects. Low values are more transparent,
high values are more opaque.
edge kernel edge_kernel 0 Sets the number of pixels to use as a kernel to reduce
dark edges that may exist in transition areas due to an
over-dense matte.
Note: Settings this control too high may

cause excessive print-through from the
background at the edges of the matte.
warm warm 0 Adjusts the matte density in warm colors (like flesh
tones). Low values are more transparent, high values
are more opaque.
Note: Print-through may occur in red-hued

foreground objects if this control is set too
low.
cool cool 0 Adjusts the matte density in cool colors. Low values
are more transparent, high values are more opaque.
Note: Print-through may occur in blue-

hued foreground objects if this control is
set too low.
Screen Correct Tab - these controls are primarily for use on green screen imperfections
enable screen_correct enabled When enabled, the overlay Screen Correction
REFERENCE GUIDE
513
Ultimatte |

Value
controls are activated.
There are a few cases where you can get better results
by disabling screen correction:
• When there is no clear view of the screen. For
instance it is entirely covered by smoke, or is shot
through a window with lots of reflections.
• When you happen to have a combination of
noise/grain and screen color that make it impractical
or impossible to find settings for the controls that
produce a usable result.
The screen correct algorithm generates a synthetic

clean plate internally, but for best results, also provide
a clean plate as input if you have one.
Note: It is important to pre-crop out any

non-image areas, such as letterbox blanking
areas, so that they are not interpolated into
the screen area.
screen screen_tolerance 0.50999999, Adjusts the color range, or tolerance, per channel to
tolerance 0.50999999, be included or excluded from the screen overlay.
0.51999998
shrink screen_shrink 2 Adjusts the screen overlay. Increasing the shrink value
erodes the overlay and decreasing it dilates the
overlay.
darks (red screen_darks_red_ 0 Excludes or includes dark areas from the screen
smaller) smaller overlay. This control only affects areas where the blue
value (when using green screen) is greater than the
red value in the foreground image.
darks (red screen_darks_red_ 0 Excludes or includes dark areas from the screen
larger) larger overlay. This control only affects areas where the red
REFERENCE GUIDE
514
Ultimatte |

Value
blue value in the foreground image.
brights (red screen_brights_ 100 Excludes or includes bright areas from the screen
smaller) red_smaller overlay. This control only affects areas where the blue
red value in the foreground image.
brights (red screen_brights_ 100 Excludes or includes bright areas from the screen
larger) red_larger overlay. This control only affects areas where the red
blue value in the foreground image.
orphans screen_orphans coarse Controls the level at which rogue orphan pixels are
excluded from the screen overlay:
• off - no orphan pixels are removed automatically.
• coarse - low level exclusion.
• medium - medium level exclusion.
• fine - high level exclusion.
reset reset N/A Click to reset the ScreenCorrect settings to their

default values.
Shadows Tab
enable shadow disabled When enabled, the Shadows controls are activated.
These are used to adjust the opacity, sharpness, and

color of shadows with minimal effect on the
foreground subject. Shadow processing relies on the
internal synthetic clean plate and thus is only active if
screen correct is also enabled.
The effects of the Shadows controls can be seen in

both the matte and the color channels.
high shadows_high 1 Reduces or eliminates unwanted shadows that are

lighter than those that are retained.
Note: All retained shadows are lightened by this
REFERENCE GUIDE
515
Ultimatte |

Value
process.
low shadows_low 0 Restores the density of the darkest part of retained

shadows.
density shadows_density 100 Adjusts the density of retained shadows. Higher

values increase and lower values decrease the density.
blur shadows_blur 0 Blurs any retained shadows.
tint shadows_tint 0 Tints the color of any retained shadows.
Spill Tab
enable spill_suppression enabled When enabled, the Spill controls are activated.
The Ultimatte algorithms automatically suppresses

spill if spill suppression processing is enabled. Use
the Spill controls to eliminate excessive spill and to
restore color similar to the backing that has been
inappropriately removed from the foreground.
The effects of the Spill controls can be seen in the

processed foreground, but are best judged using the
composite output mode.
cool spill_cool 50 Adjusts the amount of spill in cool colored

foreground objects. Used to reproduce blue, green,
or cyan colors that changed through the spill
suppression algorithms.
warm spill_warm 50 Adjusts the amount of spill in warm colored

foreground objects. Used to reproduce pink, purple,
and magenta colors for bluescreen, or yellow and
orange colors for greenscreen that changed through
the spill suppression algorithms.
midtones spill_midtones 50 Adjusts the amount of spill on midrange foreground

objects.
brights spill_brights 100 Adjusts the amount of spill on bright foreground
REFERENCE GUIDE
516
Ultimatte |

Value
objects.
darks spill_darks 0 Adjusts the amount of spill on dark foreground

objects.
ambience spill_ambience 0.5 Sets a color to subtly influence the foreground

objects in areas that may have contained spill.
strength spill_strength 40 Adjusts the intensity of the ambient color.
background spill_ 0.4 Sets the background veiling color used to override
veiling backgroundveiling the automatic suppression of the backing color, which
attempts to suppress the backing color to black.
Increasing the value increases the amount of veiling

(residue) left in the backing area. Decreasing the value
increases suppression of that color in the screen area
and can cause foreground edges and transparencies
to shift color towards the channel's complementary
color (cyan for red, magenta for green, and yellow for
blue).
Note: In most cases this control can be left

at the default value.
reset reset N/A Click to reset all Spill settings to their default values.
Cleanup Tab
enable cleanup disabled When enabled, the Cleanup controls are activated.
These controls adjust the black and gray areas of the

matte. This dramatically affects the foreground edges,
the opacity of transparent objects, and the noise in
the foreground image. They result in loss of
foreground detail, so use sparingly, and whenever
possible solve the problem using the ScreenCorrect
controls instead.
REFERENCE GUIDE
517
Ultimatte |

Value
The effects of the Cleanup controls can be seen in

both the matte and the color channels.
cleanup cleanup_cleanup 100 Reduces imperfections or small amounts of noise in

the screen area. Lower values are sharper, higher
values reduce more noise.
Note: Raising this setting too high results

in a cut and paste look with a hard, unnatural
edge.
shrink cleanup_shrink 0 Reduces the size of the cleaned-up matte.
blur cleanup_blur 0 Softens the cleaned-up matte.
recover cleanup_recover 0 Sets a threshold below which the Cleanup controls

have no effect.
Color Tab
enable color_ disabled When enabled, the Color controls are activated.
conformance
These controls apply a color correction to the
processed foreground to better match the
background. Use the Viewer toolbar match pickers to
choose colors that should match and, if necessary,
adjust the overall correction with the Color controls.
The effects of the Color controls are best judged
using the composite output mode.
darks color_darks 0.5 Adjusts the darkest parts of the image. This is a global
control, affecting the entire image, but the greatest
effects are seen in the darkest areas.
midtones color_midtones 0.5 Adjusts the midtones parts of the image. This is a
REFERENCE GUIDE
518
Ultimatte |

Value
global control, affecting the entire image, but the

greatest effects are seen in the midtone areas.
brights color_brights 0.5 Adjusts the brightest parts of the image. This is a
global control, affecting the entire image, but the
greatest effects are seen in the lightest areas.
hue color_hue 0 Adjusts the color contents of the image without

changing its brightness or color intensity (purity)
values. At default setting (0), the image hue is not
altered.
The range of the control extends from -300 to +300.
saturation color_saturation 0 Adjusts the color intensity or purity values of the

image without altering its color contents or
brightness values. At default setting (0), the image
saturation is not altered.
At the minimum setting (-200), the color intensity is

reduced to zero and the image is monochrome, or
shades of gray.
brightness color_brightness 0 Adjusts the overall intensity of the image. There is no

change at the default setting of 0.
Film Tab
enable film disabled When enabled, the Film controls are activated.
These controls attempt to reduce the effects of cyan

undercutting, a phenomenon the occurs due to the
nature of film's emulsion layers. Cyan undercutting is
typically not a problem until bluescreen keying
techniques are applied.
transparency film_transparency 100 Adjusts the amount of film correction in partially

transparent foreground areas (such as hair detail).
Transparency is most easily judged using the

REFERENCE GUIDE
519
Ultimatte |

Value
correction film_correction 0.2, 0, 0 Adjusts the amount of correction per-channel in the

foreground image.
Corrections are most easily judged using the

strength film_strength 0 Adjusts the overall amount of film correction applied

to the foreground image.
shrink film_shrink 2 Erodes or dilates the subject overlay.
brights film_brights 200 Includes or eliminates bright areas from the subject
overlay.
darks film_darks 5 Includes or eliminates dark areas from the subject

overlay.
REFERENCE GUIDE
520
Merge Nodes
Merge nodes deal with layering multiple images together. The default mode is over, which simply places
one image on top of another.
Absminus
Absminus is the same as the Merge node, only with operation set to difference by default. It layers
images together using the difference compositing algorithm: abs(A-B). This algorithm calculates how
much the pixels in input A differ from the pixels in input B. It can be useful for comparing two very similar
images.
See also Merge.
AddMix
Using AddMix, the alpha of input A is used to index two color correction lookup curves (LUTs). The first is
used to multiply input A, the second to multiply input B, and the results are added together. AddMix
performs a similar operation to the Merge node's over mode, but also premultiplies your image. If you
have a CG image that has been rendered, you're probably better off not using AddMix because you might
lose detail by having your image premultiplied twice.
AddMix can be useful if your input contains a mask that covers a specific area and you want to only merge
in that particular area, for instance if you have rotoscoped a particular shape and want to mix it using the
roto shape as an alpha.
REFERENCE GUIDE
521
|
Inputs and Controls

Type Name
Input A The first input sequence to mix (must include an alpha channel).
B The second input sequence to mix.
mask An optional image to use as a mask. By default, merge is limited to

or set to none.

(Scripting) Value
AddMix Tab
A A rgba Select the channels to use on the A input.

B B rgba Select the channels to use on the B input.

output output rgba Render into selected output channels.

premultiplied premultiplied Disabled Compensate for the A input being premultiplied. Colors
in the A input are divided by their alpha before being
multiplied by the A lookup curve.
nonlinear nonlinear Disabled Remap the zero to infinity range down into 0-1
REFERENCE GUIDE
522
Blend |

(Scripting) Value
logarithmically, so as to fit high dynamic range image

details into the lookup curve range. This may improve
results when merging high dynamic range images.
[A and B LUTs] lut N/A View and adjust the color lookup curves for A and B
inputs.
reset N/A N/A Reset the A and B LUT curves back to the original values.
mix mix 1 Dissolve between the B input only and the merged
result.
Blend
Lets you blend images together by creating the weighted average of all of the inputs. You can continue to
add more inputs, and more weight parameters will be added accordingly. This allows you to easily
average together three, four, or more images and adjust how much each contributes to the result.
This node is particularly useful if you have multiple images to combine, allowing you to specify the
amount to mix them by for each image. It is similar to using the Merge node, except if you wanted to
merge more than two images with Merge nodes, you would have to chain multiple nodes.
Inputs and Controls

Type Name
Input numbered inputs The images you want to average together. Connect a minimum of
two images to the numbered inputs.
mask An optional image to use as a mask. By default, the blend effect is

or set to none.
REFERENCE GUIDE
523
ContactSheet |

Value
Blend Tab
channels channels all The blend effect is only applied to these channels.

[input weight0 1 The weight of each of the numbered inputs. The

number] larger the value, the more the input in question
contributes to the result.
normalize weight1 enabled When enabled, the inputs are summed up and each of
them divided by the total, so that the total doesn't
exceed 1.0. This could help you avoid overexposed
areas.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the blend

invert invert_mask disabled Inverts the use of the mask channel so that the blend
the mask.

mask.
ContactSheet
REFERENCE GUIDE
524
ContactSheet |
Generates a contact sheet that shows the different inputs or the frames of one input lined up next to each
other. This can help you demonstrate, document, or manage what you are doing for a project.
See also LayerContactSheet.
Inputs and Controls

Type Name
Input numbered The images you want to appear on the Contact Sheet.
inputs

(Scripting) Value
ContactSheet Tab
Resolution width 3072 Defines the size of the Contact Sheet in pixels.
height 2048
rows/columns rows 3 Defines the number or rows and columns in the Contact
Sheet.
columns 4
gap gap 0 Sets the gap, in pixels, around each input or frame.
center center disabled When enabled, the input images are arranged so that
the specified number of rows are centered vertically
within the Resolution bounds.
Row Order roworder BottomTop Sets the row display order:

• TopBottom - rows are populated from the top
downward.
• BottomTop - rows are populated from the bottom
upward.
Column Order colorder LeftRight Sets the column display order:

• LeftRight - columns are populated from left to right.
• RightLeft - columns are populated from right to left.
• Snake - columns are populated in alternate directions,
REFERENCE GUIDE
525
CopyBBox |

(Scripting) Value
starting from left to right.
Use frames splitinputs disabled When enabled, the FrameRange specified is read from
instead of each input and displayed in the Contact Sheet.
inputs
For example, with 4 inputs and FrameRange set to 1-2,
you’d see 8 plates displayed in the Contact Sheet (2
frames from each input).
Frame Range startframe -1 Sets the frame range to extract from each input when
Use frames instead of inputs is enabled.
endframe -1
CopyBBox
Copies the bounding box from the A input onto the B stream. The bounding box defines the area of the
frame that Nuke sees as having valid image data. The larger the bounding box is, the longer it takes Nuke
to process and render the images.
Some Nuke operations, such as a merge, can cause an expansion of the bounding box area because Nuke
does not know that the extra area is going to be black or another constant color. Often, you can fix this by
copying the bounding box from one of the inputs to the resulting image, thus cutting off the extra area.
See also AdjustBBox.
CopyRectangle
With the CopyRectangle node, you can copy a rectangular area from one input on top of another.
The CopyRectangle node can also be used to limit effects, such as color corrections, to a small rectangular
region of an image. To do so, you need to use the same image in both input A and B and only perform
the color correction on one input.
REFERENCE GUIDE
526
Dissolve |
Inputs and Controls

Type Name
Input A The image from which the rectangle is copied.
B Them image to which the rectangle is copied.

(Scripting) Value
CopyRectangle Tab
channels channels all Sets the channels to copy from input A within the
specified rectangle.

channels.
area xyrt area 512 Sets the bounds of the copy rectangle on the xy axes
and the rectangle position from the right and top.
389
1536
1167
softness softness 0 Allows you to to vignette the edges of the copied

rectangle. The larger the value, the more of the area
around the edges is faded to input B.
A value of 0 produces no vignetting.
effect at 1.
Dissolve
Dissolve creates a weighted average of two inputs.
REFERENCE GUIDE
527
Dissolve |
Inputs and Controls

Type Name
Input 0-9999 The image sequences to mix between.
mask An optional image to use as a mask. By default, the dissolve effect is

or set to none.

Value
Dissolve Tab
channels channels all The dissolve effect is only applied to these channels.

which which 0 Select the inputs to dissolve between.
Set this to an integer (whole number) to show that

input, or a fractional value to dissolve between the
two nearest inputs. For example, if you set this to 3,
Nuke displays input 3. If you set this to 3.5, Nuke
displays a mix between inputs 3 and 4.
channel to none.
maskChannelInput none The channel to use as a mask. By default, the dissolve

REFERENCE GUIDE
528
In |

Value
dissolve is limited to the non-white areas of the mask.
the mask.

mask.
In
In is the same as the Merge node, only with operation set to in by default. It layers images together using
the in compositing algorithm: Ab. This algorithm only shows the areas of image A that overlap with the
alpha of B. It can be useful for combining mattes.
See also Merge.
Keymix
Keymix layers two images together using a specified Roto shape or image as a mask. It copies input A to
input B where the mask is non-black. This follows the formula Aa + B(1-a). It is similar to the Over node,
but for use with unpremultiplied images. The resulting image contains no alpha channel.
See also Over within the Merge node table.
REFERENCE GUIDE
529
Keymix |
Inputs and Controls

Type Name
Input A The image sequence to merge with input B. By default, this input is
copied to input B only where the mask is non-black.
B The image sequence to merge with input A.
mask The image to use as a mask. By default, the effect of input A is limited

(Scripting) Value
Keymix Tab
channels channels all The channels to copy from input A.
Any other channels are copied unchanged from input B.
mask channel N/A enabled Enables the associated mask channel to the right.
channel to none.
maskChannel rgba.alpha The channel to use as a mask. By default, the effect of

input A is limited to the non-black areas of this channel.
invert invertMask disabled Inverts the use of the mask channel so that the effect of
input A is limited to the non-white areas of the mask.
mix mix 1 Dissolves between the original B input at 0 and the full
Keymix effect at 1.

• B side - use the bounding box from input B. Any data
from the A input that’s outside this box is clipped.
• A side - use the bounding box from input A. Any data
from the B input that’s outside this box is clipped.
REFERENCE GUIDE
530
LayerContactSheet |
LayerContactSheet
Generates a contact sheet that shows all the layers in the input lined up next to each other. This can help
you demonstrate, document, or manage what you are doing for a project.
See also ContactSheet.
Inputs and Controls

Type Name
Input numbered The images you want to appear on the Contact Sheet.
inputs

Value
LayerContactSheet Tab
Resolution width 3072 Defines the size of the Contact Sheet in pixels.
height 2048
rows/columns rows 3 Defines the number or rows and columns in the

Contact Sheet when automatic rows/columns is
columns 4 disabled.
automatic autodims enabled When enabled, the number of rows and columns is
rows/columns auto-calculated.
gap gap 0 Sets the gap, in pixels, around each input.
center center disabled When enabled, the input images are arranged so that
the specified number of rows are centered vertically
within the Resolution bounds.
Row Order roworder BottomTop Sets the row display order:

• BottomTop - rows are populated from the bottom
REFERENCE GUIDE
531
Matte |

Value
upward.
• TopBottom - rows are populated from the top
downward.
Column Order colorder LeftRight Sets the column display order:

• LeftRight - columns are populated from left to
right.
• RightLeft - columns are populated from right to
left.
Show layer showLayerNames disabled When enabled, the layers on the Contact Sheet are
names labeled with the associated channel name.
Matte
Matte is the same as the Merge node, only with operation set to matte by default. It layers images
together using the matte compositing algorithm: Aa+B(1-a). This algorithm is the same as a
premultiplied over. You should use unpremultiplied images with this operation.
Max
Max is the same as the Merge node, only with operation set to max by default. It layers images together
using the max compositing algorithm: max(A,B). This algorithm outputs the maximum value for each
channel of each pixel. It can be useful for combining mattes and bringing aspects like bright hair detail
through.
Merge
REFERENCE GUIDE
532
Merge |
Merge allows you to layer multiple images together. When using this node, you need to select a
compositing algorithm that determines how the pixel values from one input are calculated with the pixel
values from the other to create the new pixel values that are output as the merged image. The operation
dropdown menu houses a large number of different compositing algorithms, giving you great flexibility
when building your composite.
When using most of the available merge algorithms, Nuke expects premultiplied input images. However,
with the matte operation you should use unpremultiplied images.
You can also create this node by pressing M on the Node Graph.
Inputs and Controls

Type Name
Input A The image sequence to merge with input B. When you connect the A
input, a new input A1, A2, etc. is spawned allowing you to connect
multiple image.

cannot see the mask input, ensure that the mask control isn’t

Value
Merge Tab
operation operation over Sets the Merge algorithm to use:

• atop - Shows the shape of image B, with A covering
B where the images overlap. Algorithm: Ab+B(1-a)
• average - The average of the two images. The result
is darker than the original images. Algorithm:
(A+B)/2
REFERENCE GUIDE
533
Merge |

Value
• color-burn - Image B gets darker based on the

luminance of A. Algorithm: darken B towards A
• color-dodge - Image B gets brighter based on the
luminance of A. Algorithm: brighten B towards A
• conjoint-over - Similar to the over operation,
except that if a pixel is partially covered by both a
and b, conjoint-over assumes a completely hides b.
For instance, two polygons where a and b share
some edges but a completely overlaps b. Normal
over produces a slightly transparent seam here.
Algorithm: A+B(1-a)/b, A if a>b
• copy - Only shows image A. This is useful if you also
set the mix or mask controls so that some of B can
still be seen. Algorithm: A
• difference - How much the pixels differ. See also
Absminus. Algorithm: abs(A-B)
• disjoint-over - Similar to the over operation, except
that if a pixel is partially covered by both a and b,
disjoint-over assumes the two objects do not
overlap. For instance, two polygons that touch and
share an edge. Normal over produces a slightly
transparent seam here. Algorithm: A+B(1-a)/b, A+B
if a+b<1
• divide - Divides the values but stops two negative
values from becoming a positive number.
Algorithm: A/B, 0 if A<0 and B<0
• exclusion - A more photographic form of
difference.
Algorithm: A+B-2AB
• from - Image A is subtracted from B. Algorithm: B-A
• geometric - Another way of averaging two images.
Algorithm: 2AB/(A+B)
• hard-light - Image B is lit up by a very bright and
sharp light in the shape of image A. Algorithm:
REFERENCE GUIDE
534
Merge |

Value
multiply if A<.5, screen if A>.5

• hypot - Resembles the plus and screen operations.
The result is not as bright as plus, but brighter than
screen. Hypot works with values above 1. Algorithm:
diagonal sqrt(A*A+B*B)
• in - Only shows the areas of image A that overlap
with the alpha of B. See also In. Algorithm: Ab
• mask - This is the reverse of the in operation. Only
shows the areas of image B that overlap with the
alpha of A. Algorithm: Ba
• matte - Premultiplied over. Use unpremultiplied
images with this operation. See also Matte.
Algorithm: Aa+B(1-a) (unpremultiplied over)
• max - Takes the maximum values of both images.
See also Max. Algorithm: max(A,B)
• min - Takes the minimum values of both images.
See also Min. Algorithm: min(A,B)
• minus - Subtracts B from A. Algorithm: A-B
• multiply - Multiplies the values but stops two
negative values from becoming a positive number.
See also Multiply.
Algorithm: AB, A if A<0 and B<0
• out - Only shows the areas of image A that do not
overlap with the alpha of B. See also Out. Algorithm:
A(1-b)
• over - This is the default operation. Layers image A
over B according to the alpha of image A.
Algorithm: A+B(1-a)
• overlay - Image A brightens image B.
Algorithm: multiply if B<.5, screen if B>.5
• plus - The sum of image A and B. Note that the plus
algorithm may result in pixel values higher than 1.0.
See also Plus. Algorithm: A+B
REFERENCE GUIDE
535
Merge |

Value
• screen - If A or B is less than or equal to 1 the

screen, else use the maximum example. Similar to
plus. See also Screen. Algorithm: A+B-AB if A and B
between 0-1, else A if A>B else B
• soft-light - Image B is lit up. Similar to hard-light,
but not as extreme. Algorithm: B(2A+(B(1-AB))) if
AB<1, 2AB otherwise
• stencil - This is the reverse of the out operation.
Only shows the areas of image B that do not overlap
with the alpha of A.
Algorithm: B(1-a)
• under - This is the reverse of the over operation.
Layers image B over A according to the matte of
image B. Algorithm: A(1-b)+B
• xor - Shows both image A and B where the images
do not overlap. Algorithm: A(1-b)+B(1-a)
Video sRGB disabled When enabled, colors are converted to the default 8-
colorspace bit colorspace (ProjectSettings > LUT > 8-bit files)
before doing the composite and results are converted
back to linear afterwards.
alpha screen_alpha disabled When enabled, the input images are unchanged
masking where the other image has zero alpha, and the output
alpha is set to a+b-a*b.
When disabled, the same math is applied to the alpha

as is applied to the other channels.
Note: This option is disabled on operations

where it makes no difference or PDF/SVG
specifications says it should not apply.
set bbox to bbox union Sets the output bounding box type and any data
outside this region is clipped off:
• union - Resize the output bbox to fit both input
REFERENCE GUIDE
536
Merge |

Value
bboxes completely.
• intersection - Use only those parts of the image
where the input bboxes overlap.
• A - Use input A’s bbox.
• B - Use input B’s bbox.
metadata metainput B Sets which input’s metadata is passed down the node
from tree.
Note: When metadata from is set to All

and there are keys with the same name in
both inputs, keys in B override keys in A.
range from rangeinput B Sets which input's range is passed down the node
tree.
A channels Achannels rgba Sets the channels from the A input(s) to merge with
the B channels, and which channel to treat as the A
alpha.
Note: Setting A channels to none is the

same as black or zero.
B channels Bchannels rgba Sets the channels to use from the B input, and which
channel to treat as the B alpha.
output output rgba Sets the channels to which the merge of the A and B
channels are written.
also merge also_merge none Sets the channels that are merged in addition to those
specified in the A channels and B channels controls.
channel to none.
REFERENCE GUIDE
537
MergeExpression |

Value
invert invert_mask disabled Inverts the use of the mask channel, so that the merge
the mask.

mask.
merge at 1.
MergeExpression
Allows you to merge two images using Tcl expressions. The syntax is the same as for the Color > Math >
Expression node, except that you need to precede channel references with either A or B to indicate the
input. For example, to reference the red, green, blue, and alpha channels of input A, you should use Ar, Ag,
Ab, and Aa. Similarly, to reference these channels in input B, use Br, Bg, Bb, and Ba.
To reference pixels in other layers, use INPUTlayer.channel, for example Bmatte.garbage. If you don’t
specify a layer, Nuke assumes the channel is in the current layer, for example Bgarbage.
For more information on the syntax for Tcl expressions, see Nuke's online help in the Nuke or select Help
> Documentation > Knob Math Expressions in Nuke.
See also Expression.
REFERENCE GUIDE
538
MergeExpression |
Inputs and Controls

Type Name
Input A The image sequence to merge with input B.

cannot see the mask input, ensure that the mask control isn’t disabled
or set to none.

Value
MergeExpression

REFERENCE GUIDE
539
MergeExpression |

Value




REFERENCE GUIDE
540
MergeExpression |

Value
channels channel0 red The channel(s) to which you want to apply the

You need to precede channel references with either A

or B to indicate the input. For example, to reference
the red, green, blue, and alpha channels of input A,
you should use Ar, Ag, Ab, and Aa. Similarly, to
reference these channels in input B, use Br, Bg, Bb,
and Ba.
To reference pixels in other layers, use

INPUTlayer.channel, for example Bmatte.garbage.
If you don’t specify a layer, Nuke assumes the channel
is in the current layer, for example Bgarbage.
channels channel1 green The channel(s) to which you want to apply the


and Ba.

REFERENCE GUIDE
541
MergeExpression |

Value
channels channel2 blue The channel(s) to which you want to apply the


and Ba.

channels channel3 alpha The channel(s) to which you want to apply the


and Ba.

REFERENCE GUIDE
542
MergeExpression |

Value
channel to none.

invert invert_mask disabled Inverts the use of the mask channel, so that the merge
the mask.

mask.
none.


here. This will simulate doing the merge before the
premultiplication was done.

effect at 1.
REFERENCE GUIDE
543
Min |
Min
Min is the same as the Merge node, only with operation set to min by default. It layers images together
using the min compositing algorithm: min(A,B). This algorithm outputs the minimum value for each
channel of each pixel.
See also Merge.
Multiply
Multiply is the same as the Merge node, only with operation set to multiply by default. It layers images
together using the multiply compositing algorithm: AB, A if A<0 and B<0. This algorithm multiplies the
pixel channel values but takes the value of A if both A and B are negative. It can be useful for compositing
darker values from A with the image of B - dark gray smoke shot against a white background, for example.
It can also be used to add a grain plate to an image regrained with F_ReGrain.
See also Merge.
Out
Out is the same as the Merge node, only with operation set to out by default. The Out node can be useful
for combining mattes. It layers images together using the out compositing algorithm: A(1-b), where b is
the alpha value of a pixel from image B. This algorithm only shows the areas of image A that overlap areas
of B which are rejected when using B in a matte.
See also Merge.
REFERENCE GUIDE
544
Plus |
Plus
Plus is the same as the Merge node, only with operation set to plus by default. It layers images together
using the plus compositing algorithm: A+B. This algorithm calculates the sum of image A and B. Note that
it may result in pixel values higher than 1.0.
Plus can be useful for compositing laser beams, but you’re better off not using this one for combining
mattes.
Premult
By default, Premult multiplies the input’s rgb channels by its alpha (in other words, premultiplies the input
image). You may need this node when:
• Merging unpremultiplied images. Because Merge nodes in Nuke expect premultiplied images, you
should use this node before any Merge operations if your input images are unpremultiplied. This avoids
unwanted artifacts, such as fringing around masked objects.
• Color correcting premultiplied images. When you color correct a premultiplied image, you should first
connect an Unpremult node to the image to turn the image into an unpremultiplied one. Then, perform
the color correction. Finally, add a Premult node to return the image to its original premultiplied state
for Merge operations.
Typically, most 3D rendered images are premultiplied. As a rule of thumb, if the background is black or
even just very dark, the image may be premultiplied.
See also Unpremult.
Inputs and Controls

Type Name
Input unnamed The unpremultiplied image sequence you want to premultiply.
REFERENCE GUIDE
545
Screen |

(Scripting) Value
Premult Tab
multiply channels rgb The channels to multiply (usually rgb).

by N/A enabled Enables the associated channel to the right. Disabling

alpha rgba.alpha Multiply the above channels by this channel (usually

alpha).
invert invert disabled Inverts the use of the alpha channel.
Screen
Screen is the same as the Merge node, only with operation set to screen by default. It layers images
together using the screen compositing algorithm: A+B-AB if A or B ≤1, otherwise max(A,B). In other
words, if A or B is less than or equal to 1, the screen algorithm is used, otherwise max is chosen. Screen
resembles plus. It can be useful for combining mattes and adding laser beams.
See also Merge.
Switch
Lets you switch between any number of inputs. This can be useful when using gizmos.
Inputs and Controls

Type Name
Input 0,1, 2, 3, 4... The inputs you intend to switch between.
REFERENCE GUIDE
546
TimeDissolve |

(Scripting) Value
Switch Tab
which which 0 The input to display. Each input is displayed at the value
corresponding to the number of the input. For example:
• Setting which to 4 displays the image from input 4.
• Setting which to 6 displays the image from input 6.
TimeDissolve
Dissolves between two inputs, starting the dissolve at the in frame and ending at the out frame. You can
specify the dissolve curve over time.
Inputs and Controls

Type Name
Input A The input you intend to dissolve from.
B The input you intend to dissolve to.

(Scripting) Value
TimeDissolve Tab
in in 1 Sets the frame to begin the dissolve over time.
out out 10 Sets the frame to end the dissolve.
Curves lut N/A Use the curve to define how the dissolve behaves in the
frame range:
• 0 - the frame before the in frame.
• 1 - the frame after the out frame.
Reset N/A N/A Click to reset the curve to the default shape.
REFERENCE GUIDE
547
Unpremult |
Unpremult
By default, this node divides the input’s rgb channels by its alpha. If your input images are premultiplied,
this can make color corrections more accurate.
When you color correct premultiplied images, you should first connect an Unpremult node to the image
to turn the image into an unpremultiplied one. Then, apply the color correction. Finally, add a Premult
node to return the image to its original premultiplied state - this last step is important if you want to
merge the image over another one, as Merge nodes in Nuke expect premultiplied images.
Typically, most 3D rendered images are premultiplied. As a rule of thumb, if the background is black or
even just very dark, the image may be premultiplied.
See also Premult.
Inputs and Controls

Type Name
Input unnamed The premultiplied image sequence to receive the unpremultiply.

(Scripting) Value
Unpremult Tab
divide channels rgb The channels to divide (usually rgb).

by N/A enabled Enables the associated alpha channel to the right.

channel to none.
alpha rgba.alpha Divide the above channels by this channel (usually alpha).
invert invert disabled Inverts the use of the alpha channel.
REFERENCE GUIDE
548
ZMerge |
ZMerge
Merges input images together at their appropriate 3D depths. It does this by setting each output pixel to
the input with the closest Z depth value. Z depth values are usually stored in a separate depth channel (for
example, depth.z) and determine the placing of objects within the field of view of the camera. By default,
ZMerge considers smaller Z depth values (darker areas) to be closer to the camera and larger values
(lighter areas) further away, but you can invert this behavior.
You can use ZMerge as an alternative to creating holdout geometry, but note that you may get poor
results or have problems with anti-aliasing if your input images include transparent areas or motion blur.
See also Deep Nodes.
Inputs and Controls

Type Name
Input [numbered Images to merge together based on their Z depth values.

inputs]
Each input should include a Z depth channel describing the depth of
the objects in that image. ZMerge merges the depth channels from
all inputs and uses the result to determine what input each output
pixel should come from.

(Scripting) Value
ZMerge Tab
Z channel N/A enabled Enables the associated Z channel to the right. Disabling
z_channel depth.Z The channel with the Z depth values. This should be the
same for each input. For example, if you are merging
two inputs, both data streams should have a Z depth
channel in the channel specified here.
REFERENCE GUIDE
549
ZMerge |

(Scripting) Value
smaller Z = backwards disabled When enabled, darker Z values are further away and
further away hidden by lighter Z values. This is correct for Nuke and
Pixar’s RenderMan renderer, which output 1/distance.
When disabled, lighter Z values are further away and

hidden by darker Z values.
Alpha channel N/A enabled Enables the associated alpha channel to the right.
channel to none.
alpha_channel rgba.alpha When enabled, ZMerge uses the inputs’ alpha channels
to blend between the foreground and background. You
may want to check this if your input elements have alpha
channels and you have a depth map that covers the
entire scene (rather than just the individual elements).
REFERENCE GUIDE
550
Transform Nodes
Transform nodes deal with translation, rotation, and scale as well as tracking, warping, and motion blur.
AdjustBBox
The AdjustBBox node lets you expand or crop the edges of the bounding box by a specified number of
pixels. The bounding box defines the area of the frame that Nuke sees as having valid image data. For
example, if you have an image with lots of black (0,0,0,0), you can adjust the bounding box to contain just
the useful area so that Nuke won’t waste time computing results where there is no change.
Inputs and Controls

Type Name
Input unnamed The image whose bounding box to resize.

(Scripting) Value
AdjBBox Tab
Add Pixels numpixels 25 Adjust the width and height of the input image’s
bounding box by adding or removing pixels. Click the 2
button to display separate fields for width (w) and
height (h).
If you crop the bounding box, the edge pixels of the

bounding box area get replicated towards the edges of
the image.
REFERENCE GUIDE
551
BlackOutside |
BlackOutside
The BlackOutside node fills everything outside the bounding box area with black. This removes stripes or
other anomalies caused by the edge pixels of the bounding box being replicated. It may be useful, for
example, after adjusting the bounding box with an AdjustBBox or CopyBBox node.
Inputs and Controls

Type Name
Input unnamed The image whose edges outside the bounding box you want to fill
with black.
CameraShake
Adds simulated camera shake to a sequence using random changes in amplitude, rotation, scale, and so
on. Also includes shutter controls to produce motion blur.
Inputs and Controls

Type Name
Input unnamed The sequence to which the camera shake is applied.

Value
CameraShake Tab
REFERENCE GUIDE
552
CameraShake |

Value
amplitude amplitude 10 Sets the size (pixels) of the shake.
rotation rotation 0 Sets the maximum rotation (degrees) around the cs_
center.
scaling scaling 0 Sets the maximum scale fluctuation.
frequency frequency 0.5 Sets the lowest frequency (cycles per frame) of the
shake.
octaves octaves 2 Adds higher frequency to the camera shake, in

addition to the basic shake, producing a more
random shake.
seed seed 0 Sets the seed used to create the random shake
pattern. You can change this number to produce a
slightly different pattern.
You can copy this seed to other CameraShake nodes

to copy the shake motion.
Randomize randomize_seed N/A Click to create a random seed for the shake pattern.
Seed
fixed scale scale 1 Scales the sequence to ensure the edges of the
image don’t enter the frame during extreme camera
shake.
center xy cs_center N/A The center of the camera shake scale and rotation.
filter filter Cubic Select the filtering algorithm to use when

remapping pixels from their original positions to
new positions. This allows you to avoid problems
with image quality, particularly in high contrast areas
of the frame (where highly aliased, or jaggy, edges
may appear if pixels are not filtered and retain their
original values).
values.
REFERENCE GUIDE
553
CameraShake |

Value

plus minor sharpening (as shown by the negative -
• Simon - remapped pixels receive some
smoothing, plus medium sharpening (as shown by
the negative -y portions of the curve).
• Rifman - remapped pixels receive some
smoothing, plus significant sharpening (as shown
by the negative -y portions of the curve).
Sinc4 the most.
clamp clamp disabled When using filters that employ sharpening, such as
Rifman and Lanczos, you may see a haloing effect.
If necessary, check clamp to correct this problem.
crop black_outside disabled When disabled, the outside area is filled with the
outermost pixels of the image sequence.
When enabled, this renders as black pixels outside

the image boundary, making it easier to layer the
element over another.
Note: This control also adds a solid alpha

covering the input image area if no alpha
is present.
REFERENCE GUIDE
554
CameraShake |

Value
motionblur motionblur 1 Sets the number of motion blur samples. A value of

1 should produce reasonable results for most
sequences.
Increase the value to produce more samples for

higher quality, or decrease it to shorten the
processing time. The higher the value, the smoother
the result.
shutter offset shutteroffset centered This value controls how the shutter behaves with
respect to the current frame value. It has four
options:
frame. For example, if you set the shutter value to
1 and your current frame is 30, the shutter stays
open from frame 29,5 to 30,5.
frame 30 to 31.
frame 29 to 30.
the current frame.
REFERENCE GUIDE
555
Card3D |

Value

Card3D
The Card3D node lets you apply the same geometric transformations possible with the Transform node,
but gives you an additional axis of operation, z. It transforms the image as though it was printed on a flat
card and placed in front of the camera. The Card3D node’s transformations are not truly 3D, but rather
what is sometimes called “2.5D” - meaning that you can move an element back on the z axis, but doing so
does not convey the sense that it is behind or in front of another element. 2.5D transformations are useful
for tasks like “cheating” the perspective of an element or “faking” a camera zoom.
Inputs and Controls

Type Name
Input axis An optional Axis node that works as null object to which the card can
be parented. If connected, you can use the Axis transformation
controls to control the Card3D node. Rotating the Axis node, for
example, rotates the Card3D node. This can be useful if you want to
control several nodes using the same Axis node.
cam An optional Camera node used for viewing the card.
unnamed The image to place on the card.
REFERENCE GUIDE
556
Card3D |

Value
Card3D Tab
file_menu N/A The File Menu lets you:

• Import chan file - Import a channel file and move
the card according to the transformation data in
that file. Channel files contain a set of Cartesian
coordinates for every frame of animation in a
given shot. You can create and export them using
Nuke or 3D tracking software, such as 3D-Equalizer,
Maya, or Boujou.
• Export chan file - Export the transformation data
you have applied to the card as a channel file. This
is a useful method of sharing setups between
artists.
snap_menu N/A The Snap Menu lets you:

• Match selection position - match the position of
the card to any vertices you have selected on a 3D
geometry object in the Viewer.
• Match selection position, orientation - match the
position and orientation of the card to any vertices
you have selected on a 3D geometry object in the
Viewer.
• Match selection position, orientation, size -
match the position, orientation, and size of the
card to any vertices you have selected on a 3D
geometry object in the Viewer.
To select vertices in the Viewer, you need to attach a

3D object to the Viewer and press Tab to activate
the Viewer’s 3D mode. Then, click to activate

the Vertex selection tool, and select vertices on the
object by dragging a marquee over them.
transform xform_order SRT Select the order by which Nuke executes scales,
order rotations, and translations (S signifies scale, R
REFERENCE GUIDE
557
Card3D |

Value
rotation, and T translation). The order can affect the

outcome.
rotation rot_order ZXY Select the order by which Nuke executes rotation on
order individual axes (X, Y, and Z).
translate xyz translate 0, 0, -1 Translates the card on the x,y, and z axes.
Alternatively, you can drag on any axis on the

transformation overlay in the Viewer.
rotate xyz rotate 0, 0, 0 Rotates the card on the x, y, and z axes. This is useful
for cheating the perspective.
Alternatively, you can press Ctrl/Cmd while

dragging on any axis on the transformation overlay
in the Viewer.
scale xyz scaling 1, 1, 1 Scales the card on the x, y, and z axes.
Setting one of these to a negative value reverses the

input image along that axis.
uniform scale uniform_scale 1 Scales the frame simultaneously on x, y, and z. This

effectively gangs the scale xyz values.
skew xyz skew 0, 0, 0 Skews the card on the x, y, and z axes.
pivot xyz pivot 0, 0, 0 When you make changes to the card’s position,
scaling and rotation, these occur from the location
of the card’s origin point or pivot. The pivot x, y,
and z controls allow you to offset the pivot point
and move it anywhere you like - you can even move
it outside of the card. Subsequent local
transformations will then occur relative to the new
pivot point location.
You can also position the pivot point by pressing

Ctrl/Cmd+Alt while dragging in the Viewer.
Local matrix
REFERENCE GUIDE
558
Card3D |

Value
specify useMatrix disabled Enable this control to specify matrix values for the
matrix object you’re transforming as an alternative to
setting transform, scale, skew and pivot values
above.
matrix matrix N/A The matrix displays values from the object’s
transform, rotate, scale, skew, and pivot controls.
Check specify matrix and copy or drag-and-drop

matrix values from another object to apply those
values, for example, if you wanted to align objects in
a scene.
geometry
lens-in focal lens_in_focal 1 The focal length of the camera that took the picture
on the card. The card is scaled to the correct size so
that at distance Z, the card fills this field of view.
lens-in lens_in_haperture 1 The horizontal aperture of the camera that took the
haperture picture on the card. The card is scaled to the correct
size so that at distance Z, the card fills this field of
view.
render params
output format root.format Select the format to which you want to output the
format sequence. If the format does not yet exist, you can
select new to create a new format from scratch. The
default setting, root.format, resizes the image to
the format indicated on the Project Settings dialog
box.

original values).
REFERENCE GUIDE
559
Card3D |

Value

values.
Sinc4 the most.
black outside black_outside enabled This renders as black pixels outside the image
boundary, making it easier to layer the element over
another. If you uncheck this control, the outside area
is filled with the outermost pixels of the image
sequence.
In most cases, you should keep black outside

checked. However, you may want to turn this off for
camera shake, or if you want to texture-map or
REFERENCE GUIDE
560
Card3D |

Value
intersect the output with a similar shape.
Note: Enabling black outside also adds a

solid alpha covering the input image area
if no alpha is present.

sequences.

the result.
shutter offset shutteroffset start This value controls how the shutter behaves with
options:
frame 30 to 31.
frame 29 to 30.
REFERENCE GUIDE
561
CornerPin2D |

Value

the current frame.

CornerPin2D
The CornerPin2D node is designed to map the four corners of an image sequence to or from positions
derived from tracking data. In practice, this node lets you replace any four-cornered feature with another
image sequence. You can use it to place an image in an on-screen television, for example.
Before using this node, you should use the Tracker node to generate four tracks, one per corner, on the
feature requiring replacement.
Note: In order to populate the to and from fields with the right values, you need to create the
CornerPin2D node with your input image selected in the Node Graph. If the CornerPin node is
created as an unconnected node, the to and from fields use the root format values.
REFERENCE GUIDE
562
CornerPin2D |
Inputs and Controls

Type Name
Input unnamed The image that will replace the feature tracked using the Tracker
node.
Note that the CornerPin2D node should NOT be connected to the

Tracker node or the Tracker node’s input image.

Value
CornerPin2D Tab
to1 xy to1 N/A Pin 1. This is the bottom left corner of the feature
requiring replacement. You should link this control
to the Tracker node’s tracking data for the bottom
left track. To do so, Ctrl/Cmd+drag the animation
button next to the track in the Tracker node on top
of the animation button here. You shouldn’t adjust
these values manually.
enable1 enable1 enabled Enable or disable pin 1.
to2 xy to2 N/A Pin 2. This is the bottom right corner of the feature
to the Tracker node’s tracking data for the bottom
right track. To do so, Ctrl/Cmd+drag the animation
to3 xy to3 N/A Pin 3. This is the top right corner of the feature
to the Tracker node’s tracking data for the top right
track. To do so, Ctrl/Cmd+drag the animation
REFERENCE GUIDE
563
CornerPin2D |

Value

to4 xy to4 N/A Pin 4. This is the top left corner of the feature
to the Tracker node’s tracking data for the top left
track. To do so, Ctrl/Cmd+drag the animation
Copy ’from’ copy_from_to N/A Click to copy and paste the from1-4 values to the
to1-4 values.
extra matrix transform_matrix N/A Copy tracking information from a Roto/RotoPaint

node to this matrix.
invert invert disabled When enabled, the current to values are inverted.

original values).
values.
REFERENCE GUIDE
564
CornerPin2D |

Value

Sinc4 the most.
sequence.



REFERENCE GUIDE
565
CornerPin2D |

Value
sequences.

the result.
options:
frame 30 to 31.
frame 29 to 30.
the current frame.

REFERENCE GUIDE
566
CornerPin2D |

Value

From Tab
from1 xy from1 N/A This value is identical to the original to1 value
(bottom left), and you can change this value
manually to adjust the corner pin result. You can, for
example, make your result bigger than the original
tracked polygon area by giving your corners new xy
values, or reposition the corners to get a mirrored
image.
from2 xy from2 N/A This value is identical to the original to2 value
(bottom right), and you can change this value
manually to adjust the corner pin result. You can, for
example, make your result bigger than the original
tracked polygon area by giving your corners new xy
values, or reposition the corners to get a mirrored
image.
from3 xy from3 N/A This value is identical to the original to3 value (top
right), and you can change this value manually to
adjust the corner pin result. You can, for example,
make your result bigger than the original tracked
polygon area by giving your corners new xy values,
or reposition the corners to get a mirrored image.
from4 xy from4 N/A This value is identical to the original to4 value (top
left), and you can change this value manually to
adjust the corner pin result. You can, for example,
make your result bigger than the original tracked
polygon area by giving your corners new xy values,
or reposition the corners to get a mirrored image.
Set to input set_to_input N/A Click to set the from values to the input format.
Copy 'to' copy_from_to N/A Click to copy and paste the to1-4 values to the
from1-4 values.
REFERENCE GUIDE
567
Crop |
Crop
The Crop node lets you cut out the unwanted portions of the image area. You can fill the cropped portion
with black or adjust the image output format to match the cropped image.
Inputs and Controls

Type Name
Input unnamed The image sequence to crop.

(Scripting) Value
Crop Tab
box x, y, r, t box N/A The area of the input image you want to keep. Anything
outside this box is cropped.
(or x, y, w, h)
• x - the distance (in pixels) between the left edge of the
original image and the left side of the crop box.
of the original image and the bottom edge of the crop
box.
• r - the distance (in pixels) between the left edge of the
original image and the right side of the crop box.
the original image and the top edge of the crop box.
• h - the height of the crop box. This is only available if
REFERENCE GUIDE
568
Crop |

(Scripting) Value

dragging its edges.
softness softness 0 Allows you to to vignette the edges of the cropped

portion. The larger the value, the more of the area
around the edges is faded to black.
A value of 0 produces no vignetting.
reformat reformat disabled When enabled, the image output format is changed to
match the cropped image.
When disabled, the original image output format is

used.
intersect intersect disabled When enabled, the output bounding box is an

intersection of the crop bounding box and the incoming
bounding box.
When disabled, the output bounding box matches the

crop bounding box and can extend outside the
incoming bounding box.
black outside crop enabled This renders as black pixels outside the image boundary,
making it easier to layer the element over another. If you
uncheck this control, the outside area is filled with the
In most cases, you should keep black outside checked.
REFERENCE GUIDE
569
GridWarp |

(Scripting) Value
However, you may want to turn this off for camera shake,
or if you want to texture-map or intersect the output
with a similar shape.

solid alpha covering the input image area if no
alpha is present.
GridWarp
The GridWarp node allows you to warp images by transferring image information from one Bezier grid
onto another. When using this node, you first position the source grid, which defines where to warp from.
Next, you position the destination grid, which defines where to warp the image to. This grid can be a
duplicate of the source grid, or you can define it separately. When you manipulate the destination grid,
the corresponding warp is applied to the source image.
See also SplineWarp.
Inputs and Controls

Type Name
Input bg The background image to receive the warp result when the
background control is set to bg.
dst The destination image to warp to.
src The source image to warp from.
mask An optional image to use as a mask. By default, the warp effect is

REFERENCE GUIDE
570
GridWarp |

Type Name
node, but when you drag it, it turns into an arrow labeled mask.

Value
GridWarp Tab
channels channels all The warp effect is only applied to these channels.

channel to none.
maskChannelInput none The channel to use as a mask. By default, the warp is

invert invert_mask disabled Inverts the use of the mask channel so that the warp is
the mask.

mask.
Source Grid
keys source_grid_keys 0, 0 Controls keyframes applied to the Source Grid:

• keys - displays the current and total number of
keyframes.
• - browse to the previous and next keyframe.
REFERENCE GUIDE
571
GridWarp |

Value
• - add or remove a keyframe on the current

frame.
• - copy a keyframe to the clipboard.
• - paste a keyframe from the clipboard.
Note: If Viewer tools auto is enabled, any

changes to the grid are automatically set as
keyframes at the current frame.
visible source_grid_ disabled When enabled, the Source Grid lines and handles are
visible visible.
locked source_grid_ disabled When enabled, the Source Grid lines and handles are
locked locked.
Resize to source_image_ N/A Click to resize the Source Grid to the same size as the
Image size src input image.
Destination Grid
keys destination_grid_ 0, 0 Controls keyframes applied to the Destination Grid:

keys • keys - displays the current and total number of
keyframes.

frame.
• - copy a keyframe to a temporary buffer.
• - paste a keyframe from the temporary buffer.
REFERENCE GUIDE
572
GridWarp |

Value

changes to the grid are automatically set as
visible destination_grid_ enabled When enabled, the Destination Grid lines and handles
visible are visible.
locked destination_grid_ disabled When enabled, the Destination Grid lines and handles
locked are locked.
Resize to destination_ N/A Click to resize the Destination Grid to the same size
Image image_size as the dst input image.
Settings
output output source Controls what output is displayed in the Viewer:

warped • source - the source image and source grid.
• sourcewarped - the source image and destination
grid.
• destination - the destination image and destination
grid.
• destinationwarped - the destination image and
source grid.
• morph - the morphed image, controlled by the
warp and mix parameters, and both grids.
warp warp 1 Controls grid distortion during morphing. The mix

value controls whether the source or destination grids
are affected.
REFERENCE GUIDE
573
GridWarp |

Value
Note: Warp is only enabled when the

output control is set to morph.
mix mix 0 Dissolves between the source image (at 0) and the
destination image (at 1).
Note: Mix is only enabled when the output

control is set to morph.
background background on src Sets what to use as the warp background:

• on black - render the warped image on top of a
constant black image.
• on src - render the warped image on top of the
image connected to the src input of the GridWarp
node.
• on dst - render the warped image on top of the
image connected to the dst input of the GridWarp
node.
• on bg - render the warped image on top of a
background image connected to the bg input of the
GridWarp node.
background background_mix 0 Blends between the output of the GridWarp node (at
mix 0) and whatever you have selected from the
background dropdown menu (at 1).
set bbox to boundary_box union Sets the boundary box properties.

• union - both input boundary boxes are combined.
• format - the input with the largest boundary box is
used.
Transform Tab
Source
translate xy source_grid_ 0, 0 Translates the selected grid(s) on the x and y axes.
REFERENCE GUIDE
574
GridWarp |

Value
transform_ Alternatively, you can drag on an axis on the

translate transformation overlay in the Viewer.
rotate source_grid_ 0 Rotates the selected grid(s) around the center xy

transform_rotate coordinates.
Alternatively, you drag the transformation overlay in

the Viewer.
scale source_grid_ 1 Scales the selected grid(s) width and height.

transform_scale
selected grid(s) along that axis.
skew X source_grid_ 0 Skews the selected grid(s) on the x axis.

transform_skewX
skew Y source_grid_ 0 Skews the selected grid(s) on the y axis.

transform_skewY
skew order source_grid_ XY Sets the order in which skew transforms are applied
transform_skew_ to the source grid:
order • XY
• YX
center xy source_grid_ Dependent Sets the center of scaling and rotation for the
transform_center on source selected grid(s).
input
extra matrix source_grid_ N/A Copy tracking information from either a Tracker or
transform_matrix Roto/RotoPaint node to this matrix.
Warps linked in this way follow the information from

the imported track automatically.
Destination
use source use_source_ enabled When enabled, the source and destination grids are
transform transform translated identically.
Disable this control if you want to translate the
REFERENCE GUIDE
575
GridWarp |

Value
destination image independently.
transform xy destination_grid_ 0, 0 Translates the destination grid on the x and y axes.

transform_
translate Alternatively, you can drag on an axis on the
rotate destination_grid_ 0 Rotates the destination grid around the center xy.
transform_rotate
Alternatively, you drag the transformation overlay in
the Viewer.
scale destination_grid_ 1 Scales the destination grid width and height.

transform_scale
input image along that axis.
skew X destination_grid_ 0 Skews the selected grid(s) on the x axis.

transform_skewX
skew Y destination_grid_ 0 Skews the selected grid(s) on the y axis.

transform_skewY
skew order destination_grid_ XY Sets the order in which skew transforms are applied
transform_skew_ to the destination grid:
order • XY
• YX
center xy destination_grid_ Dependent Sets the center of scaling and rotation for the
transform_center on destination grid.
destination
input
extra matrix destination_grid_ N/A Copy tracking information from either a Tracker or
transform_matrix Roto/RotoPaint node to this matrix.
Warps linked in this way follow the information from

the imported track automatically.
REFERENCE GUIDE
576
GridWarp |

Value
Render Tab
submesh render_res 10 Sets the number of subdivisions created between

resolution bezier curves in the grid during rendering.
filter filter cubic Select the filtering algorithm to use when remapping
highly aliased, or jaggy, edges may appear if pixels
are not filtered and retain their original values).
values.
Sinc4 the most.
Options Tab
REFERENCE GUIDE
577
GridWarpTracker |

Value
Colors
source color source_color N/A Sets the color of the source grid.
destination destination_color N/A Sets the color of the destination grid.

color
GridWarpTracker
GridWarpTracker is an alternate solution to manual tracking with PlanarTracker, which allows you to warp
and morph using custom grid shapes driven by tracking data, rather than being constrained to rigid
transformations. If you have a NukeX or Nuke Studio license, you can also use SmartVectors to drive the
grids.
The From and To grids allow you to add and copy tracking data between grids so that you can make
adjustments without losing your original data and without having to create a backup version of the node.
If you have a NukeX or Nuke Studio license, you can add keyframed adjustment grids to modify your From
and To grid shapes without altering the original grid data.
Inputs and Controls

Type Name
Input src The sequence you want to warp or morph.
dst An optional input to which the src image is morphed using the
Morph Amount and Mix controls.
SmartVector An optional input allowing you to drive the warp or morph using
NukeX's SmartVector node.
REFERENCE GUIDE
578
GridWarpTracker |

Value
GridWarpTracker Tab

channel to none.

effect is limited to the non-black areas of this
channel.
the mask.

mask.
Mode
Output output Warp Sets the output displayed in the Viewer:

• Warp - displays the warped src input from the
points in the From grid to the points in the To
grid.
• Morph - displays both grids and the morphed
image depending on the warp and mix
parameters.
Invert invert_warp disabled When enabled, swap the From and To grids so that
the warped destination is displayed, rather than the
REFERENCE GUIDE
579
GridWarpTracker |

Value
warped source.
Warp warp 1 When Output is set to Warp, controls the amount

Amount of warp between the From and To grids, where 1 is
the full warp and 0 is the source image.
Morph morph 1 When Output is set to Morph, controls the strength

Amount of morphing between the From and To grids, where
1 is the full morph and 0 is the source image.
Mix mix 0 When Output is set to Morph, adjusts the mix

between the source and destination images, where
0 is the source image and 1 is the destination image.
Set BBox To bbox Union Select how you want to output the bounding box.
The default is Union, which combines the two
bounding boxes. You can also select Format to set
the bounding box to the format size.
Background background Black If your grids are smaller than the format size, sets
whether the warp or morph is rendered against a
Black constant or the Source image.
Grids
Divisions XY divisions_x 5 Controls the number of columns in the grids.

Accepts values from 3 to 20
divisions_y 5 Controls the number of rows in the grids. Accepts

values from 3 to 20
Gang divisions_gang When enabled, changes to either of the divisions

sliders is applied to both axes.
Keys N/A 0 Displays the total number of user keyframes added

to the sequence.
N/A 0 Displays the current user keyframe.
N/A N/A Click to jump to the previous user keyframe.
REFERENCE GUIDE
580
GridWarpTracker |

Value
N/A N/A Click to jump to the next user keyframe.
N/A N/A Click to add a user keyframe at the current frame in

the sequence.
N/A N/A Click to delete the user keyframe at the current

frame in the sequence.
N/A N/A Click to copy the current keyframe to the clipboard.
N/A N/A Click to copy all keyframes to the clipboard.
N/A N/A Click to paste keyframes from the clipboard.
From N/A N/A Displays the state of the From grid where the warp
or morph operation begins.
• - click to link or unlink the From and To grids.
When linked, changes to either grid are applied to
the other.
• - click to make the selected grid active.
• - click to hide the selected grid.
• - click to lock the selected grid. Vertices on a

locked grid cannot be adjusted.
• - click to change the color of the selected grid.
To N/A N/A Displays the state of the To grid where the warp or
morph operation ends. Also displays the state of any
adjustment grids, if present.
• - click to link or unlink the From and To grids.
When linked, changes to either grid are applied to
the other. Adjustment grids cannot be linked to
other grids.
• - click to make the selected grid active.
• - click to hide the selected grid.
• - click to lock the selected grid. Vertices on a
REFERENCE GUIDE
581
GridWarpTracker |

Value
locked grid cannot be adjusted.

• - click to change the color of the selected grid.
Note: GridWarpTracker can process a

maximum of six adjustment grids.
Transform Tab
From
Translate xy from_base_ 0,0 Adjusts the translation, rotation, and scale values
transform_translate applied to the base grid.
Rotate from_base_ 0
transform_rotate
Scale from_base_ 1
transform_scale
Skew X from_base_ 0 Adjusts the skew applied to the base grid on the X
transform_skewX and Y axes.
Skew Y from_base_ 0
transform_skewY
Skew Order from_base_ XY Sets the order in which skew transforms are applied:
transform_skew_
order XY
YX
Center xy from_base_ dependent Sets the center of rotation and scaling. You can also
transform_center on input cmd + drag the Transform widget in the Viewer.
format or
Compositing This control defaults to the center of the input
environment format or the Project Settings > full size format
Project control if no input is connected.
Settings
Extra Matrix from_base_ N/A Adds an extra matrix which is calculated after the
REFERENCE GUIDE
582
GridWarpTracker |

Value
transform_matrix transforms defined by the other controls.
This can be useful, for example, if you are using a

3rd party application, such as a tracker, and your
tracking data cannot be fully represented using the
available transform controls.
To
Use From use-source_ enabled When enabled, the transforms applied to the From
Transform transform grid are mirrored on the To grid.
When disabled, you can apply transforms to the To

grid independent of the From grid.
Translate xy to_base_transform_ 0,0 Adjusts the translation, rotation, and scale values
translate applied to the base grid, providing Use From
Translation is disabled.
Rotate to_base_transform_ 0
rotate
Scale to_base_transform_ 1
scale
Skew X to_base_transform_ 0 Adjusts the skew applied to the base grid on the X
skewX and Y axes, providing Use From Translation is
disabled.
Skew Y to_base_transform_ 0
skewY
Skew Order to_base_transform_ XY Sets the order in which skew transforms are applied,
skew_order providing Use From Translation is disabled:
XY
YX
Center xy to_base_transform_ dependent Sets the center of rotation and scaling, providing
center on input Use From Translation is disabled. You can also cmd
format or + drag the Transform widget in the Viewer.
Compositing
environment This control defaults to the center of the input
REFERENCE GUIDE
583
GridWarpTracker |

Value
Project format or the ProjectSettings > full size format

Settings control if no input is connected.
Extra Matrix to_base_transform_ Adds an extra matrix which is calculated after the
matrix transforms defined by the other controls, providing
Use From Translation is disabled.
This can be useful, for example, if you are using a

3rd party application, such as a tracker, and your
tracking data cannot be fully represented using the
available transform controls.
Render Tab
Submesh render_res 10 Sets the number of subdivisions that are created

Resolution between bezier curves in the grid.
Filter filter Cubic Sets the image resampling filter to use when
original values).
values.
REFERENCE GUIDE
584
IDistort |

Value

Sinc4 the most.

aborts.
the render aborts.

IDistort
IDistort warps the input image based on the values in the image’s UV channels. The U and V values are
offsets for where a pixel will come from. For example, if pixel 51, 23 has a U and V value of -1, 5, the
pixel's value will come from 50, 28 of the input channels.
REFERENCE GUIDE
585
IDistort |
You can use the Copy node to merge the two distortion channels in with your image channels, and then
select the two channels from the UV channels dropdown menu.
Inputs and Controls

Type Name
Input unnamed This input needs to include both the channels to distort and the two
channels to use to calculate the distortion.
mask An optional image to use as a mask. By default, the distortion is

or set to none.

(Scripting) Value
IDistort Tab
channels channels all The distortion is only applied to these channels.

channels.
UV channels uv none The two channels that are used to calculate the
distortion for the input image. The U and V values are
offsets for where a pixel will come from. For example, if
pixel 51, 23 in the above input channels has a U and V
value of -1, 5, the pixel's new value will come from 50,
28 of the input channels.
UV offset uv_offset 0 This is subtracted from the UV channels to set a non-

zero center point for renderers that cannot output
negative numbers.
UV scale uv_scale 1 Multiply the UV channels by this value.
REFERENCE GUIDE
586
IDistort |

(Scripting) Value
blur channel N/A disabled Enable the associated blur channel to the right.
channel to none.
blur none Values in this channel are added to the size of the
sample area to add extra blur or diffusion to the
distortion.
blur scale blur_scale 1 Adjusts the amount of blur.
mask channel N/A disabled Enable the associated mask channel to the right.
channel to none.
maskChannel none The channel to use as a mask. By default, the distortion

invert invert_mask disabled Inverts the mask so that the distortion is limited to the
non-white areas of the mask.
premultiplied premultiplied disabled Check this if the UV and blurchannels have been
premultiplied by the alpha channel, such as when
output by a renderer.
filter filter Cubic Select the filtering algorithm to use when remapping
• Impulse - remapped pixels carry their original values.
• Keys - remapped pixels receive some smoothing, plus
minor sharpening (as shown by the negative -y
plus medium sharpening (as shown by the negative -y
REFERENCE GUIDE
587
LensDistortion |

(Scripting) Value

plus significant sharpening (as shown by the negative -
• Mitchell - remapped pixels receive some smoothing,
plus blurring to hide pixelation.
Sinc4 the most.
LensDistortion
The LensDistortion node (NukeX and Nuke Studio only) estimates the lens distortion in a given image,
either through Grid Detection or manual Line Detection. The warp can then be used to add or remove
distortion or produce an STMap in the motion channel for use elsewhere.
Note: You must perform the analysis in NukeX or Nuke Studio, but you can use the results in
Nuke.
Inputs and Controls

Type Name
Input Source The image sequence to warp.
REFERENCE GUIDE
588
LensDistortion |

Value
LensDistortion Tab
Local GPU gpuName N/A Displays the GPU used for

rendering when Use GPU if
available is enabled. Local
GPU displays Not available
when:
• Use CPU is selected as the
default blink device in the
Preferences.
• no suitable GPU was found
on your system.
• it was not possible to create
a context for processing on
the selected GPU, such as
when there is not enough
free memory available on
the GPU.
You can select a different

GPU, if available, by
navigating to the Preferences
and selecting an alternative
from the default blink device
dropdown.
Note: Selecting a
different GPU
requires you to
restart Nuke before
the change takes
effect.
Use GPU if available useGPUIfAvailable enabled When enabled, rendering

occurs on the Local GPU
specified, if available, rather
REFERENCE GUIDE
589
LensDistortion |

Value
than the CPU.
Note: Enabling this

option with no local
GPU allows the
script to run on the
GPU whenever the
script is opened on a
machine that does
have a GPU available.
You should also
select this if you
wish to render from
the command line
with the --gpu
option.
See Nuke's Online Help

Installation pages for more
information on the GPUs
supported by your operating
system.
Model Preset distortionModelPreset NukeX Sets the distortion model to

Classic use and adds or removes
Distortion Parameter
controls depending on the
preset selected:
• NukeX Classic - uses the
distortion model from
legacy versions of Nuke.
• CaraVR Radial, Degree 3 -
uses a distortion model
suitable for CaraVR.
• 3DEqualizer - provides
REFERENCE GUIDE
590
LensDistortion |

Value
several distortion models

suitable for 3DE.
• Custom - allows you to
customize the distortion
model from scratch.
Lens Type lens Spherical Sets the lens type used to

shoot the sequence:
• Spherical
• Anamorphic
Anamorphic Parameters - These controls are only displayed when the Lens Type control is set to
Anamorphic.
Squeeze anamorphicSqueeze 1 Sets the anamorphic squeeze

used to rescale the x
coefficients in the Distortion
Parameters.
Twist anamorphicTwist 0 Compensates for lens twist,

measured in degrees.
Scale x,y anamorphicScale 1,1 Compensates for lens

breathing, the slight changes
in focal length when changing
the focus.
Distortion Parameters - these controls are dependent on the distortion Model Preset applied.
Denominator s distortionDenominator0 0 Sets the denominator

coefficients for the symmetric
coefficients (s) of the
distortion model.
distortionDenominator1 0
These controls are linked to
the controls of the same name
on the Advanced tab.
Centre centre 0,0 Sets the position of the
REFERENCE GUIDE
591
LensDistortion |

Value
distortion center.
Output
Mode output Undistort Sets the type of output from

the node:
• STMap - renders both
undistortion and redistortion
STMaps in the motion
channels. Use the forward
channels for undistorting
and the backward channels
for redistorting. The other
input channels are copied to
the output directly. In this
mode, the overlay grids can
be displayed on top of the
source image.
• Undistort - undistorts the
input directly in the Viewer,
allowing to visualize the
undistorted overlay grid.
• Redistort - redistorts the
input directly in the Viewer,
allowing to visualize the
redistorted overlay grid.
Use Fisheye useFisheye enabled When enabled, lens

projection is considered as
part of the distortion when
warping fisheye lenses.
When disabled, Nuke renders

an 'ideal' fisheye without
distortion when undistorting,
and applies distortion to a
fisheye image when
REFERENCE GUIDE
592
LensDistortion |

Value
redistorting.
Note: This control

has no effect on
Rectilinear lenses.
Filter filter Cubic When Mode is set to

Undistort or Redistort, sets
the image resampling filter to
use when remapping pixels
from their original positions
to new positions. This allows
you to avoid problems with
image quality, particularly in
high contrast areas of the
frame (where highly aliased or
jaggy edges may appear if
pixels are not filtered and
retain their original values).
• Impulse - remapped pixels
carry their original values.
• Cubic - remapped pixels
receive some smoothing.
• Keys - remapped pixels
receive some smoothing,
plus minor sharpening (as
shown by the negative -y
• Simon - remapped pixels
plus medium sharpening (as
shown by the negative -y
• Rifman - remapped pixels
REFERENCE GUIDE
593
LensDistortion |

Value

plus significant sharpening
(as shown by the negative -y
• Mitchell - remapped pixels
plus blurring to hide
pixelation.
• Parzen - remapped pixels
receive the greatest
• Notch - remapped pixels
receive flat smoothing
(which tends to hide moire
patterns).
• Lanczos4, Lanczos6, and
Sinc4 - remapped pixels
receive sharpening which
can be useful for scaling
down. Lanczos4 provides
the least sharpening and
Sinc4 the most.
BBox bboxType Auto Sets how the bounding box is

calculated:
• Auto - the bounding box is
set automatically, which can
lead to a performance hit if
the bounds are much larger
than the image.
• Input - the bounding box is
set to the same dimensions
as the input image.
• Manual - allows you to set
the bounding manually
REFERENCE GUIDE
594
LensDistortion |

Value
using the outputBBox

controls.
outputBBox dependent When BBox is set to Manual,

on adjust the BBox by entering
Compositing values in the x, y, r, and t
environment controls.
Project
Settings You can also use the crop
widget in the Viewer to adjust
the BBox.
Output Format outputFormatType Input Sets the output format:

• Input - the output format is
the same as the input
format.
• Format - the output format
is controlled by the format
dropdown.
• BBox - the output format is
the same as the bounding
box.
format dependent Sets the output format to use

on when Output Format is set to
Compositing Format.
environment
Project
Settings
Distortion Scaling distortionScalingType Scale to Sets the rescaling mode:

Input Format • Scale to Input Format - use
the input format to scale the
distortion in the output.
• Format - use the selected
format to scale the output.
outputFormat dependent Sets the format to use to
REFERENCE GUIDE
595
LensDistortion |

Value
on reformat the image when

Compositing Distortion Scaling is set to
environment Format.
Project
Settings
Analysis Tab
Grid Detect
Detect detect N/A Click to run the selected

detection method on all key
frames for a calibration grid
shot.
Grid detection consists of

feature detection and feature
linking. It can be run in one
step, using Grids detection, or
broken down into two steps
to allow for manual
corrections.
See the Grid Detection

controls for more information
on changing the detection
parameters.
N/A detectionFrameType Current Sets the number of frames

Frame processed by the Detect
button:
• Current Frame - only detect
grids on the current frame.
• Every N Frames - detect
grids on the current frame
and every N frames after
that. N is specified by the
control to the right.
REFERENCE GUIDE
596
LensDistortion |

Value
The default value of N is 5

frames.
Preview preview disabled You can enable the Preview

control to view the features
that the LensDistortion node
is likely to find when you click
the Detect button.
You can adjust the controls in

the Settings dropdown to
improve the result and then
click Detect.
Settings
Number of Features numFeatures 5000 Sets the maximum number of

detected features.
You can try increasing the

Number of Features if the
grid is not fully covered by
feature points.
Patch Size patchSize 9 Sets the patch size for feature

relocalization.
Try increasing the Patch Size

if features are not consistently
located on saddle points.
Feature Separation featureSeparation 15 Sets the distribution of

features in relation to each
other.
This value should reflect the

scale of a square element of
the grid.
Detection Threshold detectionThreshold 100 Sets the detection threshold
REFERENCE GUIDE
597
LensDistortion |

Value
at which features are rejected

automatically.
You can increase the value to

reject weaker features, such as
points not on a grid.
Feature Linking
Angle Threshold angleThreshold 8 Sets the angle tolerance when

linking neighboring features.
If you find there are too many

missing links between
adjacent features, try
increasing this value.
High values can introduce

ambiguity in the linking
process, so adjust the Angle
Threshold to get the best
coverage of the grid.
Distance Threshold distanceThreshold 30 Sets the distance tolerance

allowed when merging
neighboring links.
You can increase the Distance

Threshold to recover missing
feature points after detection.
Peak Threshold peakThreshold 20 Sets the peak tolerance when

detecting linking directions.
You can decrease the Peak

Threshold to improve feature
linking if the grid image does
not contain a sufficient
contrast.
REFERENCE GUIDE
598
LensDistortion |

Value
Editing and Drawing
lineDrawingModeSelectAll N/A Click to enable the generic

selection tool.
lineDrawingModeSelectFeatures N/A Click to enable the feature

selection tool.
lineDrawingModeSelectLines N/A Click to enable the line

selection tool.
lineDrawingModeAddFeatures N/A Click to enable the add

feature tool.
lineDrawingModeRemoveFeatures N/A Click to enable the remove

feature tool.
lineDrawingModeAddLines N/A Click to enable the line

drawing tool.
Solve
Keys N/A 1 Displays the current keyframe

number.
N/A 1 Displays the total number of

keyframes.
N/A N/A Click to jump to the previous

keyframe.
N/A N/A Click to jump to the next key

frame.
N/A N/A Click to add a keyframe at the

current frame in the sequence.
N/A N/A Click to delete the keyframe

at the current frame in the
sequence.
N/A N/A Click to delete all keyframes

in the sequence.
REFERENCE GUIDE
599
LensDistortion |

Value
Solve solveDistortion N/A Click to estimate the

distortion model parameters
using the detected grids.
The solver attempts to warp

the distorted links into
straight lines.
Note: Solving the

distortion model
requires at least as
many feature links as
parameters in the
distortion model. If
the Detect method
doesn't find enough
links, try adding
more keyframes or
tuning the detection
parameters.
Reject deleteOutliers N/A Click to reject links for which

the residual error is outside
the error Threshold.
You can delete outliers drawn

red in the overlay and then
click Solve again to improve
the results by refining the
warp.
You can also delete links

manually by selecting them
and pressing the Delete or
Backspace keyboard
shortcuts. Holding Ctrl/Cmd
REFERENCE GUIDE
600
LensDistortion |

Value
while deleting links keeps

only the selected links.
Reset resetDistortion Resets the distortion

parameters and solve state.
Deleted features and links are
not restored.
Solve Error solveError 0 Displays the root mean square

(RMS) solve error in pixels.
This can be used to measure

the quality of the solve. It is
computed as the average
deviation to an ideal straight
line for each link. Move the
mouse over a link to display
its individual residual error.
Threshold errorThreshold 10 Sets the error threshold for

outliers rejection.
Lower values can produce a

closer fit, but with the risk of
overfitting. Too many outliers
can be an indication of an
issue with the selected
distortion model.
Overlay
Show overlayType All Selects what to display in the

overlay:
• None - disables the overlay.
• Features - only displays
detected feature points.
Only features are editable in
this mode. Deleting a
feature does not delete any
REFERENCE GUIDE
601
LensDistortion |

Value
associated links.
• Links - only display the links
between feature points. Only
links are editable in this
mode. Deleting a link does
not delete the features it
connects.
• All - displays both feature
points and links. When this
option is selected, both
types can be edited at the
same time.
Advanced Tab
Fisheye
Projection Model projection None Sets the projection type,

(Rectilinear) either None (Rectilnear) or
Fisheye. Fisheye offers
several sub-types:
• Stereographic
• Equidistant
• Equisolid
• Orthographic
Focal Length focal 9 Sets the camera focal length

(in mm).
Sensor Size x,y sensorSize 36,24 Sets the size of the camera
sensor (in mm).
Beam Splitter
Enable enableBeamSplitter disabled When enabled, the Direction

and Bending controls are
enabled to manage images
created using a beam splitter.
REFERENCE GUIDE
602
LensDistortion |

Value
Direction beamSplitterDirection 0 Controls the beam splitter's

cylindrical direction and
Bending beamSplitterBending 0 bending.
Distortion
Type distortionModelType Radial Sets the distortion model

Standard type, though selecting a
Preset alters the selection as
appropriate:
• Radial Standard
• Radial Asymmetric
• Radial-Tangential (Coupled
or Uncoupled)
Order distortionOrder 0,2 Sets the order of the radial

distortion's rational
polynomial function.
The first term controls the

numerator order and the
second controls the
denominator order.
Note: The default

values (0, 2)
correspond to the
legacy NukeX
model.
Exponent distortionExponent 2,2 Sets the base exponent of the

radial distortion polynomial.
The first term controls the

numerator exponents and the
second controls the
denominator exponents.
REFERENCE GUIDE
603
LensDistortion |

Value
Note: A base
exponent of 2 for
the denominator
corresponds to the
classic NukeX
model.
Direction distortionModelDirection Forward Set the distortion model

direction:
• Forward - estimated
directly but needs to be
inverted for image
rendering.
The NukeX Classic model is

an example of a Forward
model.
• Backward - estimated using
its inverse but can be used
directly for image rendering.
Model inversion is not an
exact process and may cause
loss of information due to
approximations.
The CaraVR Radial model is

an example of a Backward
model.
Normalisation normalisationType Maximum Sets how the focal length and

distortion parameters are
normalized:
• Width
• Height
• Diagonal
• Maximum
REFERENCE GUIDE
604
LensDistortion |

Value
Distort in Fisheye distortInFisheyeSpace enabled (if a When the Project Model is

Space fisheye set to Fisheye, this control
Projection determines whether to
Model is remove or apply the
selected) distortion in fisheye or
rectilinear space, that is,
before or after 'defishing' the
lens.
Equation x distortionModelDisplayX xu = xd / (1 Distortion model equation

+ k0 * rd^2 display.
+ k1 * rd^4)
Legend: (xd, yd) are the
distorted cartesian
coordinates, (rd, phid) are the
distorted polar coordinates,
(xu, yu) are the undistorted
cartesian coordinates, (ru,
phiu) are the undistorted
polar coordinates, and the k-
values are the distortion
coefficients. The coordinate
systems are relative to the
distortion centre.
Equation y distortionModelDisplayY yu = yd / (1 Distortion model equation

+ k0 * rd^2 display.
+ k1 * rd^4)
Legend: (xd, yd) are the
distorted cartesian
coordinates, (rd, phid) are the
distorted polar coordinates,
(xu, yu) are the undistorted
cartesian coordinates, (ru,
phiu) are the undistorted
polar coordinates, and the k-
values are the distortion
coefficients. The coordinate
REFERENCE GUIDE
605
LensDistortion |

Value
systems are relative to the

distortion centre.
The denominator and numerator controls are dependent on the distortion Model Preset applied on the
LenDistortion tab.
Denominator s distortionDenominator0Link 0 Sets the denominator

distortion model.
distortionDenominator1Link 0
on the LenDistortion tab.
Numerator s distortionNumerator0 0 Sets the numerator

distortion model.
distortionNumerator1 0
on the LensDistortion tab.
Centre centreLink 0,0 Sets the position of the

distortion center.
This control is linked to the

control of the same name on
the LenDistortion tab.
before render beforeRender none These functions run prior to

starting rendering in execute
(). If they throw an exception,
the render aborts.
before each frame beforeFrameRender none These functions run prior to
REFERENCE GUIDE
606
Mirror |

Value
starting rendering of each

individual frame. If they throw
an exception, the render
aborts.
after each frame afterFrameRender none These functions run after each
frame is finished rendering.
They are not called if the
render aborts. If they throw an
exception, the render aborts.
after render afterRender none These functions run after

rendering of all frames is
finished. If they throw an
error, the render aborts.
render progress renderProgress none These functions run during

rendering to determine
progress or failure.
Mirror
This node flips the input image around the center of the format area. A flip on the x axis mirrors the image
vertically. A flop on the on the y axis mirrors the image horizontally.
Inputs and Controls

Type Name
Input unnamed The image sequence to flip vertically or flop horizontally.
REFERENCE GUIDE
607
PointsTo3D |

(Scripting) Value
Mirror Tab
Horizontal Horizontal disabled When enabled, Nuke flops the image 180 on the y axis.
Vertical Vertical disabled When enabled, Nuke flips the image 180 degrees on the
x axis.
PointsTo3D
Use the PointsTo3D node to calculate the 3D location of a 2D point in an image sequence using the
parallax between two or three tracked points.
Attach an image and its corresponding tracked camera to the PointsTo3D node inputs. Select the 2D point
you want to calculate, place A on that point at a particular frame and click set frame. Now, change the
frame to a different camera angle and specify that same 2D feature with B and click set frame. Repeat for
point C, again on a different frame. Click calculate to generate the point in 3D space as well as that point
converted back to screen space.
Inputs and Controls

Type Name
Input cam A Camera node whose movement matches that of the camera used
to shoot the input sequence. This can be a camera you have created
using the CameraTracker node or a camera you have imported from a
third-party 3D application, for example. Note that the camera must
be animated.
img The image sequence that has a 2D point whose 3D position you want
to calculate.
REFERENCE GUIDE
608
PointsTo3D |

Value
PointsTo3D Tab
Camera type cameraMovType free move Select the type of tracked camera movement:
• free move - this turns a 2D point in the image into
a point in 3D space. This requires the camera to be
moving and for you to find the 2D point in three
frames. Select the 2D point you want to calculate,
place Point A on that point at a particular frame
and click set frame. Now, change the frame to a
different camera angle and specify that same 2D
feature with Point B and click set frame. Repeat
for Point C, again on a different frame. Click
calculate to generate the point in 3D space as well
as that point converted back to screen space. You
can also click generate axis to make a 3D Axis
object with this position.
• nodal - this ignores Point B and Point C. Instead, it
just puts Point A a fixed distance in front of the
camera, using the Camera's focal length to decide
the distance.
Point A
2D point xy pointA 50, 0 Choose a 2D point to track on the X and Y axes.
point ref ref_timeA 0 The frame number Point A is set to.

time
set frame N/A N/A Click to set Point A at the current xy location and
frame number.
Point B
2D point xy pointB 100, 0 Choose a 2D point to track on the X and Y axes.
point ref ref_timeB 0 The frame number Point B is set to.

time
set frame N/A N/A Click to set Point B at the current XY location and
frame number.
REFERENCE GUIDE
609
PointsTo3D |

Value
Point C
2D point xy pointC 150, 0 Choose a 2D point to track on the X and Y axes.
point ref ref_timeC 0 The frame number Point C is set to. You can leave
time the point ref time set to zero to omit Point C from
the computation. You might want to do this if you
have a free moving camera but you only want to use
Point A and Point B in the calculation.
set frame N/A N/A Click to set Point C at the current XY location and
frame number.
Output
Calculate N/A N/A Click to calculate the 3D point coordinates from the
2D point data.
3D point xyz point3D 0 The calculated X, Y, and Z axis location for the 3D
point.
2D point xy point2D 0 The current X and Y axis location of the point2d

marker.

aborts.
the render aborts.

REFERENCE GUIDE
610
Position |
Position
Position moves the input image by an integer number of pixels. It fills the area between the original
image and the moved image by repeating the edge pixels.
Inputs and Controls

Type Name
Input unnamed The image sequence to move.

(Scripting) Value
Position Tab
translate x, y translate 0, 0 How much to move the image on the x and y axes.
Positive values move the image up or to the right, while
negative values move it down or to the left.
Reconcile3D
Reconcile3D turns a point in 3D space into an x, y position in the camera. The 3D point is controlled by the
first fields in the properties panel, and by the optional Axis input. The point is projected through the
camera to the image and the position of the result is put in XY output. In addition, the absolute 3D point
is put in XYZ output (this is only useful if there is an Axis parent, otherwise it is the same as the input
point). This lets you lock a photographed plate to a 3D scene or track.
Note: The point must be in front of the camera to evaluate correctly.
REFERENCE GUIDE
611
Reconcile3D |
Inputs and Controls

Type Name
Input axis 3D point control coordinates are relative to this optional input.
cam The viewing camera. This might be created using CameraTracker or

imported from a 3rd party application.
img The image in which you want to position the 3D point.
Output resolution is determined from this image.

Value
Reconcile3D Tab
Input
3D point xyz point 0, 0, 0 The x, y, and z values of the 3D point that you want
to turn into an xy position in the camera.
Output
create N/A N/A Click to evaluate the XY output and XYZ output
keyframes keyframes for a specific frame range.
calculate calc_output disabled When enabled, the output values update

output live dynamically.
XY output output 0, 0 The x and y values of the 3D point projected by the

camera to the image. The size of the output image
is controlled by the format of the img input.
XYZ output outputxyz 0, 0, 0 The x, y, and z values of the point in 3D space,

transformed by the axis input, if applicable.
2D transform
track xy track 0, 0 A point of interest in the input image on the x and y

axes that you want to match with the Camera
REFERENCE GUIDE
612
Reconcile3D |

Value
movement.
offset xy offset 0, 0 Extra translation of the input image on the x and y

axes.
stabilize stabilize disabled Normally the transformation moves the track point
to the projected 3D point, thus moving a plate to
match a 3D scene. If you check this, the
transformation is reversed to remove 3D motion by
moving the projected 3D point to the track point.
filter filter Parzen Select the filtering algorithm to use when

original values).
values.
REFERENCE GUIDE
613
Reconcile3D |

Value

Sinc4 the most.
clamp clamp disabled Sets negative intermediate and final results to zero.
This will remove ringing around mattes when using
a filter that has negative lobes. When using filters
that employ sharpening, such as Rifman and
Lanczos, you may see a haloing effect. If necessary,
check clamp to correct this problem.
sequence.

REFERENCE GUIDE
614
Reconcile3D |

Value


sequences.

the result.
options:
frame 30 to 31.
frame 29 to 30.
REFERENCE GUIDE
615
Reconcile3D |

Value

the current frame.


aborts.
the render aborts.

REFERENCE GUIDE
616
Reformat |
Reformat
Reformat lets you resize and reposition your image sequences to a different format (width and height).
This also allows you to use plates of varying image resolution on a single script without running into issues
when combining them. All scripts should include Reformat nodes after each Read node to specify, at the
very least, the output resolution of the images in the script.
Inputs and Controls

Type Name
Input unnamed The image to resize.

(Scripting) Value
Reformat Tab
type type to format Select how to resize the image:

• to format - Set the output width and height to the
selected format. Choose the format in the output
format dropdown menu. If the format does not yet
exist, you can select new to create a new format from
scratch. The default setting, root.format, resizes the
image to the format indicated on the Project Settings
dialog.
• to box - Set the output width and height to
dimensions you define (in pixels). Enter values in the
width/height and pixel aspect fields to specify the
dimensions. This option is useful for creating a
thumbnail of the input image.
• scale - Set the output width and height to a multiple
of the input size. Use the scale slider to define the
REFERENCE GUIDE
617
Reformat |

(Scripting) Value
factor. The scale factor is rounded slightly, so that the

output image is an integer number of pixels in the
direction chosen under resize type.
output format format root.format The format to which you want to output the image
sequence. If the format does not yet exist, you can select
new to create a new format from scratch. The default
setting, root.format, resizes the image to the format
indicated in the Project Settings.
This control is only available if you have set type to to

format.
width/height box_width 200 The output width for your image. The units are in pixels.

box.
box_height 200 The output height for your image. The units are in pixels.

box.
force this box_fixed disabled When enabled, the output image matches the
shape width/height fields exactly, even if the original image is
a different shape. To achieve this, one direction gets
either clipped or padded.
When disabled, the output image is approximately the

same shape as the original, rounded to the nearest
integer number of pixels.
REFERENCE GUIDE
618
Reformat |

(Scripting) Value

box.
pixel aspect box_pixel_ 1 Sets the pixel aspect ratio for the output image.
aspect
box.
scale scale 1 The scale factor for the width and the height. To scale
each direction separately using different scale factors,
click the 2 button.
This control is only available if you have set type to

scale.
resize type resize width Choose the method by which you preserve or override
the original pixel aspect ratio. Select:
• none - to not resize the original.
• width - to scale the original until its width matches the
output width. Height is then scaled in such a manner as
to preserve the original aspect ratio.
• height - to scale the original so that it fills the output
height. Width is then scaled in such a manner as to
preserve the original aspect ratio.
• fit - to scale the original so that its smallest side fills
the output width or height. The longest side is then
scaled in such a manner as to preserve the original
aspect ratio.
• fill - to scale the original so that its longest side fills
the output width or height. The smallest side is then
aspect ratio.
• distort - to scale the original so that both sides fill the
output dimensions. This option does not preserve the
original aspect ratio, so distortions may occur.
REFERENCE GUIDE
619
Reformat |

(Scripting) Value
center center enabled When enabled, Reformat translates the image to center
it in the output.
When disabled, Reformat translates the image so that

the lower left corners line up.
flip flip disabled When enabled, Reformat flips the image upside down.
flop flop disabled When enabled, Reformat flops the image left and right.
turn turn disabled When enabled, Reformat rotates the image 90 degrees
counter-clockwise.
REFERENCE GUIDE
620
SphericalTransform |

(Scripting) Value

Sinc4 the most.
Rifman and Lanczos, you may see a haloing effect. If
necessary, check clamp to correct this problem.
black outside black_outside enabled This renders as black pixels outside the image boundary,
making it easier to layer the element over another. If you
uncheck this control, the outside area is filled with the
In most cases, you should keep black outside checked.

However, you may want to turn this off for camera shake,
or if you want to texture-map or intersect the output
with a similar shape.

solid alpha covering the input image area if no
alpha is present.
preserve pbb disabled When enabled, pixels outside the output format are
bounding box preserved.
When disabled, pixels outside the output format are

clipped off.
SphericalTransform
SphericalTransform converts images between different projections, including 360 rigs using the CaraVR
toolset in NukeX. These view projections can be divided into two broad categories:
• full frame, such as Latlong, encompassing the entire 360 world around a single point, and
• partial frame, such as the Rectilinear view, that Nuke was designed to work in.
REFERENCE GUIDE
621
The Output rotation is also controllable using an in-viewer control system. Hold down Ctrl/Cmd+Alt and
left-click and drag to move the image around, setting the pan and tilt setting. Add Shift to lock into a
single dimension for the movement. In a partial frame projection, use the right mouse button to set the
focal length, in essence zooming in and out.
Inputs and Controls

Type Name
Input -Z An environment map you want to convert to another type. Input

names vary according to the Project and Format selected.
SphericalTransform Tab
Local GPU gpuName N/A Displays the GPU used for rendering when
Use GPU if available is enabled. Local
GPU displays Not available when:
• no suitable GPU was found on your
system.
• it was not possible to create a context
for processing on the selected GPU, such
as when there is not enough free
memory available on the GPU.
You can select a different GPU, if available,

by navigating to the Preferences and
selecting an alternative from the default
blink device dropdown.

requires you to restart Nuke
before the change takes effect.
REFERENCE GUIDE
622
Use GPU if useGPUIfAvailable enabled When enabled, rendering occurs on the

available Local GPU specified, if available, rather
than the CPU.
Note: Enabling this option with

no local GPU allows the script to
run on the GPU whenever the
script is opened on a machine
wish to render from the
command line with the --gpu
option.
See Nuke's Online Help Installation pages

for more information on the GPUs
supported by your operating system.
Input
Projection projTypeInput LatLong Sets the input projection type from which
to convert. The type you select is
dependent on what directional
information your textures contain. For
example, Cube has +X,+Y,+Z -X,-Y,-Z
• LatLong - the most common full 360
frame projection. Many VR pipelines use
latlongs for both ingest and export due
to their simplicity and wide use.
• Cubemap - another full 360 projection.
Each of the six faces is essentially
rectilinear, so the data can be more
familiar to work in.
• Rectilinear - a partial frame, standard
projection you're most likely familiar
with.
REFERENCE GUIDE
623
When you’re going to or from rectilinear,

you’re only able to cover part of the
frame (up to virtually 180 degrees).
• Fisheye - covers a number of
projections, all of which emulate
common optical models used in fisheye
lenses. These are all partial frame
projections, and so enable specific
camera parameters similar to the
Rectilinear projection type.
• MirrorBall - A reflective ball, scaled up
to fill the available area.
Format packingTypeInput Image Sets how faces in the input cubemap are
displayed:
• Image
• Views
• Faces
Note: The Format control is

only displayed when Projection
is set to Cubemap.
Packing cubemapPackingInput LL-Cross Sets how faces are packed in a single

image:
• LL-Cross
• 6x1
• 3x2
Note: The Packing control is

is set to Cubemap and Format is
set to Image.
REFERENCE GUIDE
624
Note: The Focal to Shift controls are only available when Projection is set to Rectilinear or
Fisheye.
Focal focalInput 16 Sets the focal length used for the input
cameras.
Sensor sensorInput 36, 24 Sets the sensor size used for the input
cameras.
Film Back filmBackPresetInput Custom Sets the camera Sensor and Focal
Preset automatically for the preset selected.
Position positionInput 0, 0, 0 Sets the position of the input cameras with

respect to the center of the sphere.
K KInput 0, 0, 0 Sets the input camera's distortion

coefficient on the xyz axes.
Shift shiftInput 0, 0 Sets the input lens's center shift on the xy

axes.
Rotation modeInput Pan-Tilt-Roll Sets the input rotation mode:

• Look - adds an directionInput widget to
the Viewer allowing you to control the
input's look direction.
• From-To - adds fromInput and toInput
widgets to the Viewer allowing you to
transform the input by moving the
widgets around.
• Pan-Tilt-Roll - allows you to transform
the input using the Pan Tilt Roll
controls.
• Rotation Angles - allows you to rotate
the input using the Rotation Order and
Rotation Angles controls.
Pan Tilt Roll panTiltRollInput 0, 0, 0 When Rotation is set to Pan-Tilt-Roll,

allows you to transform the input
manually.
REFERENCE GUIDE
625
Output
Projection projTypeOutput LatLong Sets the output projection type to which

the input is converted. The type you select
is dependent on what directional
information your textures contain. For
example, Cube has +X,+Y,+Z -X,-Y,-Z
• LatLong - the most common full 360
frame projection. Many VR pipelines use
latlongs for both ingest and export due
to their simplicity and wide use.
• Cubemap - another full 360 projection.
Each of the six faces is essentially
rectilinear, so the data can be more
familiar to work in.
• Rectilinear - a partial frame, standard
projection you're most likely familiar
with.
When you’re going to or from rectilinear,
you’re only able to cover part of the
frame (up to virtually 180 degrees).
• Fisheye - covers a number of
projections, all of which emulate
common optical models used in fisheye
lenses. These are all partial frame
projections, and so enable specific
camera parameters similar to the
Rectilinear projection type.
• MirrorBall - A reflective ball, scaled up
to fill the available area.
Format packingTypeOutput Image Sets how faces in the output cubemap are
displayed:
• Image
• Views
• Faces
REFERENCE GUIDE
626
Note: The Format control is

is set to Cubemap.
Packing cubemapPackingOutput LL-Cross Sets how faces are packed in a single

image:
• LL-Cross
• 6x1
• 3x2
Note: The Packing control is

is set to Cubemap and Format is
set to Image.
Note: The Focal to Shift controls are only available when Projection is set to Rectilinear or
Fisheye.
Focal focalOutput 16 Sets the focal length used for the output
cameras.
Sensor sensorOutput 36, 24 Sets the sensor size used for the output
cameras.
Film Back filmBackPresetOutput Custom Sets the camera Sensor and Focal
Preset automatically for the preset selected.
Position positionOutput 0, 0, 0 Sets the position of the output cameras

with respect to the center of the sphere.
K KOutput 0, 0, 0 Sets the output camera's distortion

coefficient on the xyz axes.
Shift shiftOutput 0, 0 Sets the output lens's center shift on the xy

axes.
Rotation modeOutput Pan-Tilt-Roll Sets the output rotation mode:
REFERENCE GUIDE
627
• Look - adds an directionInput widget to

the Viewer allowing you to control the
look direction.
• From-To - adds fromOutput and
toOutput widgets to the Viewer
allowing you to transform the rig by
moving the widgets around.
• Pan-Tilt-Roll - allows you to transform
the output using the Pan Tilt Roll
controls.
• Rotation Angles - allows you to rotate
the output using the Rotation Order and
Rotation Angles controls.
Pan Tilt Roll panTiltRollOutput 0, 0, 0 When Rotation is set to Pan-Tilt-Roll,

allows you to transform the output
manually.
Filter filter Cubic Select the filtering algorithm to use when

remapping pixels from their original
positions to new positions. This allows you
to avoid problems with image quality,
particularly in high contrast areas of the
frame (where highly aliased, or jaggy,
edges may appear if pixels are not filtered
and retain their original values).
• Impulse - remapped pixels carry their
original values.
• Cubic - remapped pixels receive some
smoothing.
• Keys - remapped pixels receive some
smoothing, plus minor sharpening (as
shown by the negative -y portions of the
curve).
smoothing, plus medium sharpening (as
REFERENCE GUIDE
628
curve).
smoothing, plus significant sharpening
(as shown by the negative -y portions of
the curve).
• Mitchell - remapped pixels receive
some smoothing, plus blurring to hide
pixelation.
• Parzen - remapped pixels receive the
greatest smoothing of all filters.
• Notch - remapped pixels receive flat
smoothing (which tends to hide moire
patterns).
• Lanczos4, Lanczos6, and Sinc4 -
remapped pixels receive sharpening
which can be useful for scaling down.
Lanczos4 provides the least sharpening
and Sinc4 the most.
Format formatMode To Scale Sets which format to use or how to

compute it starting from the input:
• To Format - force the specified input
format as the output format. When using
camera type projections, such as
Rectilinear, the sensor size y parameter
is ignored.
• To Width - the output format has the
specified input width, but the height is
computed according to the output
Projection format ratio.
• To Scale - the output format width is
computed from the input format width,
and the height is computed according to
the output Projection format ratio.
Width scale 1 Sets the scaling factor to convert input

width to output width.
REFERENCE GUIDE
629
Output output root.format When Format is set to To Format,

Format specifies the output format used.
AdjustBBox adjustBBox 0 Increases the size of the output image

bounding box.
You might want to expand the bounding

box if you need access to pixels at the
edge of the image for nodes such as Blur
or LensDistortion.
Swap swap N/A Click to swap the Input and Output

Input/Output control settings.
CaraVR Options
Metadata metaDataOperation Ignore Sets how incoming metadata is handled:

• Ignore - any metadata present in the
stream is passed downstream unaltered.
• Apply - rotates the image according to
the metadata in the stream and the total
transform described by the controls in
the C_SphericalTransform node.
Note: This option assumes that

the incoming image is in LatLong
space.
• Transform - sets the output metadata to

the sum of the input metadata and the
total transform described by the controls
in the C_SphericalTransform node, but
does not modify the output image.
Export exportMenu C_ Sets the export nodes to add to the Node

SphericalTransform Graph:
(Inverted Linked) • C_SphericalTransform (Inverted
Linked) - creates an expression linked C_
REFERENCE GUIDE
630
SplineWarp |
SphericalTransform set to invert the

current transforms.
This can be useful when switching

between rectilinear and latlong to
perform VFX tasks, such as roto.
• C_SphericalTransform (Concatenated) -
sets the current C_SphericalTransform
node to Input+Output mode and
creates a second C_SphericalTransform
node in Input mode.
This can be used to create incremental

latlong rotations without introducing
unnecessary filtering hits by passing only
metadata downstream until the final C_
SphericalTansform in the node tree
performs the concatenated rotation.
Create exportButton N/A Click to create the export specified in the

export dropdown.
SplineWarp
The SplineWarp node warps an image based on multiple shapes or pins that you create. Unlike the
GridWarp node, you can draw these shapes anywhere on either image, rather than only add points on the
existing grid lines, and corresponding shapes can have differing numbers of points. You can also copy and
paste shapes from other SplineWarp and Roto or RotoPaint nodes.
See also GridWarp.
REFERENCE GUIDE
631
SplineWarp |
Inputs and Controls

Type Name
Input A The image to warp from, though B to A warps are equally valid.
B The image to warp to, though B to A warps are equally valid.
mask An optional image to use as a mask. By default, the warp effect is

limited to the non-black areas of the mask. All channels for pixels
outside the mask region are copied from input A.
node, but when you drag it, it turns into an arrow labeled mask.

(Scripting) Value
SplineWarp Tab

channels.
Note: There is no need to include the output

mask channel in this layer.
premultiply premultiply disabled When enabled, premultiply the input channels by a mask
input from the combined cookie cutter shapes prior to
warping. This control uses the same mask as output
mask but may be used without assigning an output
mask channel.
REFERENCE GUIDE
632
SplineWarp |

(Scripting) Value
Note: This control can only be enabled when

the curves list contains shapes designated as
cookiecutters.
output mask outputMask mask_ Sets the channel to be replaced by a mask from the
splinwarp.a combined cookie cutter shapes.
Note: This control can only be enabled when

the curves list contains shapes designated as
cookiecutters.
crop to crop_to_ enabled When disabled, the input image is not cropped to the
format format Project Settings > full format size. As a result, warping
can introduce pixels from the image outside the format
size, rather than black.
Note: Disabling crop to format can affect

performance as Nuke has to calculate the warp
for the entire image, not just the area inside
the format size.
bbox boundary_ enabled When enabled, a curve is added at the format

boundary bbox boundaries, effectively pinning the corners of the image.
curve
This boundary can be tightened by increasing boundary
curve resolution on the Render tab.
output output source Controls what output is displayed in the Viewer:

warped • A - the A input image.
• A warped - the warped A input image.
• B - the B input image.
• B warped - the warped B input image.
• AB morph - the morphed image, controlled by the
warp and mix controls.
REFERENCE GUIDE
633
SplineWarp |

(Scripting) Value
mix mix 0 Dissolves between the source image (at 0) and the
destination image (at 1).
Note: This control is only enabled when

output is set to AB morph.
root warp root_warp 1 Sets the global warp multiplier for all shape pairs and
layers in the curves list.
For AB pairs, root warp determines the warp where the

A image is warped towards B by the warp amount. B is
warped towards A by 1-warp.
layer warp layer_warp 1 Sets the warp multiplier for all selected layers in the
curves list.
Note: This control is disabled when anything

other than layers are selected in the curves list.
pair warp pair_warp 1 Sets the warp multiplier for the selected pairs of shapes
in the curves list.
For AB pairs it determines the warp where the A image is

warped towards B by the warp amount. B is warped
towards A by 1-warp.
Note: This control is disabled when anything

other than paired shapes are selected in the
curves list.
spline key key 0, 0 Controls spline keyframes:

• keys - displays the current and total number of
keyframes.
REFERENCE GUIDE
634
SplineWarp |

(Scripting) Value

frame.

changes to the splines are automatically set as
curves list curves N/A Shows the hierarchy of shapes and groups, allowing you
to adjust them and how they’re displayed in the Viewer:
• Name - Double-click to edit the shape or group Name.
• Pairing - Displays the pairing relationship between
shapes.
• Type - Displays which input shapes exist in A, B, or AB.
• - Designates the shape or layer as transform

linked to another shape or layer. Entries linked this way
are affected by transforms applied to transform source.
• - Designates the shape as a cookie cutter for the

premultiply and output mask controls.
• - Whether the shape or layer is visible and

rendered or not.
• - Lock or unlock the shape or layer.
• - Set the color in which you want the outline of

your shape to appear in the Viewer.
• - Toggles whether or not the current shape is a

hard boundary (removes warping outside the shape).
• - Indicates whether or not the current shape is a
REFERENCE GUIDE
635
SplineWarp |

(Scripting) Value
boundary (limits warping).
Use the shape list to:

layers.
• Right-click for a context menu containing
Addnewlayer and shape editing functions such as
Copy, Paste, and Duplicate.
Transform Tab
translate xy translate 0, 0 Translates the selected curves on the x and y axes.
Alternatively, you can drag on an axis on the

rotate rotate 0 Rotates the selected curves around the center xy

coordinates.
Alternatively, you drag the transformation overlay in the

Viewer.
scale scale 1 Scales the selected curves’ width and height.
REFERENCE GUIDE
636
SplineWarp |

(Scripting) Value

curve along that axis.
skew X skewX 0 Skews the selected curve(s) on the x axis.
skew Y skewY 0 Skews the selected curve(s) on the y axis.
skew order skew_order XY Sets the order in which skew transforms are applied to
the selected curve(s):
• XY
• YX
center xy center Dependent Sets the center of scaling and rotation for the selected
on source curves.
input
extra matrix transform_ N/A Copy tracking information from either a Tracker or
matrix Roto/RotoPaint node to this matrix.
Warps linked in this way follow the information from the

imported track automatically.
Render Tab
curve curverez 3 Adjusts the accuracy of the warp/spline match. Higher

resolution values increase accuracy, but sacrifice speed and vice-
versa.
Note: Correspondence points may be used to

improve the warping accuracy in a specific part
of the curve if turning this value up too high
causes performance problems.
boundary boundary_ 3 Adjusts the number of interpolated points on boundary

curve curverez and hard boundary curves. Higher values stop the warp
resolution from filtering through the boundary curves, but sacrifice
speed and vice versa.
preview previewrez 100 Improves the accuracy of the preview at higher values
REFERENCE GUIDE
637
SplineWarp |

(Scripting) Value
resolution and the rendering speed at lower values.
Classic rbf disabled Sets the type of warp function to use:

warping • disabled - uses an updated quadratic warp function
that copes well with overlapping control points, but
has a more local warp effect.
• enabled - employs a warping function from previous
versions of Nuke that has a more global warp effect,
but doesn’t cope well with overlapping control points.
Additionally, Classic warping produces a more linear
warp between points, which may produce better
results when using a cookie cutter mask to cut out the
warped image.
filter filter cubic Select the filtering algorithm to use when remapping
REFERENCE GUIDE
638
Stablilize |

(Scripting) Value

Sinc4 the most.
Stablilize
The Stabilize2D node is designed to remove unwanted camera movement, rotation, and/or scaling from
an image sequence. The node requires data from only a single track if you only need to stabilize
movement; it requires data from two tracks if you need to stabilize for rotation and/or scaling.
The basic procedure for using Stabilize2D is to first use the Tracker node to generate the required tracks,
then follow the Tracker node with a Stabilize2D node. To this node, you apply the tracking data in inverse
form, thus negating the unwanted transformations.
STMap
The STMap node allows you to move pixels around in an image. STMap uses two channels to figure out
where each pixel in the resulting image should come from in the input channels. You can use the Copy
node to merge the two distortion channels in with your image channels and then select the two channels
in the U and V selection boxes. The U and V values are the absolute position of the source pixel. The
REFERENCE GUIDE
639
STMap |
values are normalized to be between 0 and 1, where 0 is the bottom left corner of the input image, and 1
is the top right corner.
You can also calculate the lens distortion on one image and apply that distortion to another image using
the STMap node. See LensDistortion for more details.
Inputs and Controls

Type Name
Input stmap The input you want to use to alter the source image.
src The source image to which you want to apply the effect.
mask An optional image to use as a mask. By default, the distortion is

or set to none.

(Scripting) Value
STMap Tab
channels channels all The distortion is only applied to these channels.

channels.
UV channels uv none The two channels that are used to calculate the
distortion for the input image. The U and V values are
the absolute position of the source pixel. The values are
normalized to be between 0 and 1, where (0,0) is the
bottom left corner of the input image, and (1,1) is the
top right corner.
blur channel N/A disabled Enables the associated blur channel to the right.
REFERENCE GUIDE
640
STMap |

(Scripting) Value

channel to none.
blur none Values in this channel are added to the size of the area
to sample, to add extra blur or diffusion to the
distortion.
blur scale blur_scale 1 Adjust the blur amount by multiplying the blur values
by this.
mask channel N/A disabled Enables the associated mask channel to the right.
channel to none.
maskChannel none The channel to use as a mask. The distortion is limited

to the non-black areas of this channel.
invert invert_mask disabled Inverts the mask so the distortion is limited to the non-
white areas of the mask.
premultiplied premultiplied disabled Check this if the UV and blurchannels have been
premultiplied by the alpha channel, such as when
output by a renderer.
REFERENCE GUIDE
641
Tile |

(Scripting) Value

Sinc4 the most.
Tile
Tile produces an output image that contains scaled-down, tiled copies of the input image. The output
image is the same format as the input.
Inputs and Controls

Type Name
Input unnamed The image to reproduce by rows or columns.

(Scripting) Value
Tile Tab
REFERENCE GUIDE
642
Tile |

(Scripting) Value
rows rows 1 The number of times the image is replicated vertically.

Note that the value can be fractional.
mirror mirrorRows disabled Check this to flip adjacent tiles vertically to form mirror
images.
columns columns 1 The number of times the image is replicated

horizontally. Note that the value can be fractional.
mirror mirrorColumns disabled Check this to flip adjacent tiles horizontally to form
mirror images.
REFERENCE GUIDE
643
Tracker |

(Scripting) Value

Sinc4 the most.
Tracker
This is a 2D tracker that allows you to extract animation data from the position, rotation, and size of an
image. Using expressions, you can apply the data directly to transform and match-move another element.
Or you can invert the values of the data and apply them to the original element - again through
expressions - to stabilize the image.
This is the general process for tracking an image:

1. Connect a Tracker node to the image you want to track.
2. Use auto-tracking for simple tracks or place tracking anchors on features at keyframes in the image.
3. Calculate the tracking data.
4. Choose the tracking operation you want to perform: stabilize, match-move, etc.
See also the Tracking tab in Roto and RotoPaint.
Inputs and Controls

Type Name
Input unnamed The sequence you want to track.

Value
Tracker Tab
Tracks
REFERENCE GUIDE
644
Tracker |

Value
e N/A enabled When enabled, the associated track anchor in the

Viewer is used to track a feature from the input.
name N/A track 1 Sets the track name.
track_x N/A Dependent Adjusts the x coordinates of the associated track

on source anchor’s center.
input
track_y N/A Dependent Adjusts the y coordinates of the associated track

on source anchor’s center.
input
offset_x N/A 0 Sets the offset amount, in pixels, between the

tracking anchor and the feature position on the x
and y axes.
offset_y N/A 0 This is especially useful when the feature you want
to track is obscured or otherwise unavailable,
enabling you to track another pattern to produce
the required tracking data.
T N/A enabled When enabled, translation is calculated during

tracking.
R N/A disabled When enabled, rotation is calculated during

tracking.
Note: Tracking rotation requires at least

two tracks.
S N/A disabled When enabled, scale is calculated during tracking.
Note: Tracking scale requires at least two

tracks.
error N/A 0 Displays the overall track error rating.
REFERENCE GUIDE
645
Tracker |

Value
add track add_track N/A Click to add a new track and anchor to the Viewer.
delete track del_tracks N/A Click to delete all selected tracks.
select all select_all N/A Click to select all tracks in the Tracks list.
average average_tracks N/A Click to average all selected tracks together into a
tracks single new track. This can be especially useful for
stabilization tracking.
Export CornerPin2D
CornerPin cornerPinOptions CornerPin2D Sets the node to output when you click create. The
Options (use current baked options do not use expression links between
frame) Tracker and the exported node:
• CornerPin2D (use current frame) - creates an
expression linked CornerPin2D node that warps
the image according to the relative transform,
using the current frame as a reference.
• CornerPin2D (use transform ref frame) - creates
an expression linked CornerPin2D node that warps
the image according to the relative transform,
using the frame specified in the Transform tab as
a reference.
• Transform (stabilize) - creates an expression
linked Transform node with control presets to
stabilize the clip.
• Transform (match-move) - creates an expression
linked Transform node with control presets to
match-move the clip.
create createCornerPin N/A Click to create the CornerPin2D or Transform node

specified in Export dropdown.
Note: There must be 4 tracks selected in

the Tracks list to create a CornerPin2D
node.
REFERENCE GUIDE
646
Tracker |

Value
Settings Tab
General
track channels rgb Tracking only occurs in these channels.

channels
pre-track pretrack_filter median Sets the filter applied before image patches are
filter compared:
• none - no filter is applied.
• adjust contrast - the default filter, stretches the
image contrast to better suit the tracking
algorithm. This is the recommended setting and
shouldn’t need changing in most circumstances.
• median - attempts to remove image noise.
adjust for adjust_for_ disabled When enabled, Tracker does some extra pre-
luminance luminance_changes filtering to compensate for changes in brightness.
changes
This option slows the tracking process and can
reduce the accuracy of tracks, so only enable this
control if there are known changes in brightness.
Note: Enabling adjust for luminance

changes can occasionally produce better
tracks on shots with no differences in
luminance, particularly on jittery shots
where sub-pixel accuracy is vitally
important.
max max_iter 100 Sets the maximum number of iterations before the
iterations tracking algorithm stops searching for features.
REFERENCE GUIDE
647
Tracker |

Value
Note: This setting does not normally

need adjusting.
epsilon / epsilon 0.01 Sets the error level at which Tracker is assumed to
resolution have found the feature - no further search for a
better match is performed. Higher values may result
in a faster but less accurate track.
Note: This setting does not normally

need adjusting.
max_error max_error 0.2 Sets the error level at which Tracker stops searching
for features.
clamp super- clamp_footage enabled When enabled, tracked patches are clamped to
white, sub- values between 0-1.
zero footage
Note: If you want to track using the full

dynamic range available in your footage,
disable this control and adjust the max-
error value accordingly. For example,
images with pixel intensity 40 may require
a max-error value of 40.
show error show_error_on_ disabled When enabled, keyframes on the track are colored
on track track_links according to their relative error:
paths • green - a good match for the grabbed pattern.
• amber - a reasonable match for the grabbed
pattern.
• red - a poor match for the grabbed pattern.
High track error values aren’t necessarily bad

keyframes. Rather, they indicate that the pattern has
changed considerably since the previous pattern
REFERENCE GUIDE
648
Tracker |

Value
grab.
hide progress hide_progress_bar disabled When enabled, the track progress dialog doesn’t
bar display during tracking.
snap to snap_to_markers disabled When enabled, adding tracks places a guide on

markers suitable marker patterns, such as dots or blobs, in
the Viewer. Move the tracking anchor to a guide
and release the mouse to snap the anchor to the
guide’s position.
show zoom zoom_window_ always Sets when the zoom window is visible in the
window behavior Viewer:
• always - the zoom window is always visible.
• on track change - only display the zoom window
when a track changes.
• when tracking - only display the zoom window
during tracking.
• when tracking or track change - only display the
zoom window during tracking or when a track
changes.
• never - the zoom window is never displayed.
zoom zoom_window_size 200px Sets the size of the zoom window.

window size
/ zoom_ x1 Sets the magnification in the zoom window.
magnification magnification_size
zoom zoom_window_ on playback Sets when filtering is applied to the zoom window:
window filter filter_behaviour • always
• on playback
• never
REFERENCE GUIDE
649
Tracker |

Value
Note: The filter applied is the same as

that selected on the Transform tab, and
can produce a more visually stable track. It
can make track positioning more difficult,
however.
Auto-Tracking
predict track predict_track disabled When enabled, use the current tracker animation
path to determine where to look in the next frame.
Note: If Tracker fails to find the feature in

the next frame, click the clearfwd button
above the Viewer before continuing, or
the same error is repeated.
warp type warp Translate Selects the transformations that are tried on the
pattern to match it to the image. Translate is
fastest, but may lose the track if the pattern rotates,
scales, or shears over the course of the track:
• Translate - only expect pattern translation.
• Translate/Rotate - expect pattern translation and
rotation.
• Translate/Scale - expect pattern translation and
scaling.
• Translate/Rotate/Scale - expect pattern
translation, rotation, and scaling.
• Affine - expect straight lines and the distance
between the points on them to remain equal.
REFERENCE GUIDE
650
Tracker |

Value
Note: This control does not relate to how

the image is transformed as an output, but
what transforms the pattern matching
algorithm utilizes to find the best pattern
match.
pattern grab grab_behavior if error Sets when the tracker attempts to grab a new
behavior above pattern:
• on first frame - grab a pattern on the first frame
only.
• every frame - grab a new pattern on every frame.
• every n frames - use the every n frames control
to specify the frame interval.
• if error above - use the when error > control to
specify the grab behavior.
• if error below - use the when error < control to
specify the grab behavior.
• custom - use the every n frames and when error
<> controls to specify grab behavior.
every n grab_interval 0 When pattern grab behavior is set to update

frames every n frames or custom, sets the interval at which
Tracker automatically re-grabs a pattern.
when error > grab_error_above 0.05 When pattern grab behavior is set to update if
above tolerance or custom, sets the error level
above which Tracker automatically re-grabs a
pattern.
when error < grab_error_below 0 When pattern grab behavior is set to update if
below tolerance or custom, sets the error level
below which Tracker automatically re-grabs a
REFERENCE GUIDE
651
Tracker |

Value
pattern.
when auto_regrab_pattern disabled When enabled, Tracker re-grabs a pattern at the

tracking is current position each time tracking is stopped.
stopped
when tracker regrab_when_offset enabled When enabled, Tracker re-grabs a pattern when a
is moved track is manually adjusted.
Keyframe Tracking
retrack when retrack_on_move enabled When enabled, re-track a pattern when a track is
keyframe is manually adjusted or a new keyframe is created.
moved /
created
create new create_key_on_ enabled When enabled, create a new keyframe when the
key when move track is manually adjusted.
track is
moved
auto-tracks auto-tracks_delete_ enabled When enabled, auto-tracking deletes manual

delete keyframes keyframes.
keyframes
When disabled, auto-tracking adjusts the position
of manual keyframes.
keyframe keyframe_display scroll, single Sets how keyframe snapshots are displayed in the
display line only Viewer:
• all - show all keyframe snapshots in the available
Viewer space.
• nearest, single line only - show the nearest
keyframe snapshot to the playhead, and the
surrounding whole snapshots, on a single line
only.
• scroll, single line only - show all keyframe
snapshots in a scrolling, single line only.
• none - don’t show any keyframe snapshots.
REFERENCE GUIDE
652
Tracker |

Value
keyframe size keyframe_size 100px Sets the size of keyframe snapshots in the Viewer.
on_screen max_number_of_ 3 Sets the maximum number of tracks with keyframes

key-track keyframe_tracks_to_ that can be displayed in the Viewer.
limit display
Set this control to 0 to always display selected
tracks with keyframes.
Transform Tab
transform transform none Sets the type of transform to apply.
Note: The transform control must be set

to none when calculating tracks.
• none - no transformation is applied to the input

image.
• stabilize - transforms the image so that the
tracked points do not move.
• stabilize 1pt - similar to stabilize, but designed for
use with a single track.
• match-move - transforms another image so that it
moves to match the tracked points.
• match-move 1pt - similar to match-move, but
designed for use with a single track.
• remove jitter - transforms the image so that the
tracked points move smoothly, with high-
frequencies removed.
• add jitter - transform the input image by the
high-frequency components of the track to
increase the shake or add it to another image.
reference reference_frame 1 When transform is set to stabilize or match-move,

frame sets the frame in the input sequence to use as the
reference or identity frame.
set to current N/A N/A Click to set the current frame as the reference or
REFERENCE GUIDE
653
Tracker |

Value
frame identity frame.
jitter period jitter_period 10 When transform is set to add/removejitter, sets

the number of frames to average together to get a
stable position.
smooth smoothT 0 Smooths the translation by averaging this many

frames together.
smoothR 0 Smooths the rotation by averaging this many frames

together.
smoothS 0 Smooths the scale by averaging this many frames

together.
live-link livelink_transform disabled When enabled, changes to tracks instantly update

transform the transform. Dynamic updating may be useful if a
track position is controlled by an expression.
translate xy translate 0, 0 Displays the translation applied to the Transform

widget for the current track. Once tracking is
complete, you can drag-and-drop the translate
information to link other nodes, such as the
Stabilize node.
rotate rotate 0 Displays the rotation applied to the Transform

widget for the current track. Once tracking is
complete, you can drag-and-drop the rotate
Stabilize node.
scale scale 1 Displays the scale applied to the Transform widget

for the current track. Once tracking is complete, you
can drag-and-drop the scale information to link
other nodes, such as the Stabilize node.
skew X skewX 0 Displays the skew applied to the Transform widget

on the x axis for the current track. Once tracking is
complete, you can drag-and-drop the skew
REFERENCE GUIDE
654
Tracker |

Value
Stabilize node.
skew Y skewY 0 Displays the skew applied to the Transform widget

on the y axis for the current track. Once tracking is
complete, you can drag-and-drop the skew
Stabilize node.
skew order skew_order Sets the order in which skew transforms are applied:
• XY
• YX
center xy center Dependant Sets the center of rotation and scaling. You can also
on input cmd + drag the Transform widget in the Viewer.

with image quality, particularly in high contrast
areas of the frame (where highly aliased, or jaggy,
edges may appear if pixels are not filtered and
values.
REFERENCE GUIDE
655
Tracker |

Value

Sinc4 the most.
sequence.



REFERENCE GUIDE
656
Tracker |

Value
sequences.

the result.
shutter offset shutteroffset start Controls how the shutter behaves with respect to
the current frame value:
frame 30 to 31.
frame 29 to 30.
of - 0.5 would open the shutter half a frame

REFERENCE GUIDE
657
Transform |
Transform
Transform lets you not only translate elements, but also rotate, scale, and skew them from a single control
panel.
You can also create this node by pressing T on the Node Graph.
See also TransformMasked.
Inputs and Controls

Type Name
Input unnamed The image to translate, rotate, scale, or skew.

Value
Transform Tab
translate x, y translate 0, 0 Translates the image along the x and y axes. You can
also adjust translate values by dragging the
rotate rotate 0 Rotates the image around the center x y

coordinates. You can also adjust rotate values by
dragging the transform handle in the Viewer.
scale scale 1 Scales the image width and height around the
center x y coordinates.
skew X skewX 0 Skews the image on the x axis.
skew Y skewY 0 Skews the image on the y axis.
skew order skew_order XY Sets the order in which skew transforms are applied
to the image:
REFERENCE GUIDE
658
Transform |

Value
• XY
• YX
center x, y center N/A Sets the center of rotation and scale on the x and y
axes.
invert invert_matrix disabled When enabled, any transform you applied using the
translate xy, rotate, scale, skew, or center xy
controls is inverted.

original values).
values.
REFERENCE GUIDE
659
Transform |

Value

Sinc4 the most.
sequence.



REFERENCE GUIDE
660
Transform |

Value
sequences.

the result.
shutter offset shutteroffset start Controls how the shutter behaves with respect to
the current frame value:
frame 30 to 31.
frame 29 to 30.
the current frame.

REFERENCE GUIDE
661
TransformMasked |
TransformMasked
This node lets you translate, rotate, scale, or skew an image in 2D in the same way as the Transform node,
but it also offers controls for assigning a mask to protect certain areas of the frame from transforms.
Inputs and Controls

Type Name
Input unnamed The input you want to translate, rotate, scale, or skew.
mask An optional image to use as a mask. By default, the transform is

or set to none.

Value
Transform Tab
channels channels all The transform is only applied to these channels.

translate xy translate 0, 0 Translates the input on the x and y axes.
rotate rotate 0 Rotates the input by degrees. Negative values

scale scale 1 Scales the input. Scale width and height are ganged
by default.
REFERENCE GUIDE
662
TransformMasked |

Value
skew X skewX 0 Skews the input on the x axis.
skew Y skewY 0 Skews the input on the y axis.
skew order skew_order XY Sets the order in which skew transforms are applied
to the input:
• XY
• YX
center xy center N/A The center of rotation and scaling. You can also
move it using Ctrl/Cmd+drag.
invert invert_matrix disabled When enabled, any transform you applied using the
translate xy, rotate, scale, skew, or center xy
controls is inverted.

original values).
values.
REFERENCE GUIDE
663
TransformMasked |

Value

Sinc4 the most.
sequence.


motionblur motionblur 0 Sets the number of motion blur samples. A

motionblur value of 1 should produce reasonable
REFERENCE GUIDE
664
TransformMasked |

Value
results for most sequences.
Increase the motionblur value to produce more

samples for higher quality, decrease it to shorten the
processing time.
shutter shutter 0.5 How long the shutter remains open (measured in
frames) to produce motion blurring. The default
value sets the shutter open for half a frame.
options:
frame 30 to 31.
frame 29 to 30.
the current frame.

REFERENCE GUIDE
665
TVIScale |

Value
channel to none.
maskChannelInput none The channel to use as a matte. By default, he

transform is limited to the non-black areas of this
channel.
mask.
the mask.

mask.
mix mix 1 Dissolves between the original image at 0 and the

full effect at 1.
TVIScale
This node scales an image by a factor of two. It uses a Total Variational Inpainting (TVI) technique that
minimizes noise while still preserving the edges. The Lambda control is the main control here. The larger
the value, the less smoothing is allowed and the more noise present in the original image makes it to the
larger image.
REFERENCE GUIDE
666
TVIScale |
Inputs and Controls

Type Name
Input unnamed The input you want to scale by a factor of two.
mask An optional image to use as a mask. By default, the transform is

or set to none.

Value
TVIScale Tab
channels channels all The scale is only applied to these channels.

Iterations Iterations 5 Sets the number of iterations to apply to achieve the

best solution.
This is an iterative filter - the longer it runs, the closer

it gets to the ’best’ solution. In this case, the noisy
smooth areas converge on a non-noisy smooth area.
Lambda Lambda 5 Controls how far the filtered values are allowed to
stray from the original input.
The larger the value, the less smoothing is allowed

and the more noise present in the original image
makes it to the larger image.
A A 0.0001 This is a regularization control that should always be
REFERENCE GUIDE
667
VectorCornerPin |

Value
above zero to prevent Divides-By-Zero in the

algorithm.
channel to none.
maskChannelInput none The channel to use as a matte. By default, he transform

mask.
the mask.

mask.
effect at 1.
VectorCornerPin
The VectorCornerPin node takes paint or an image from a reference frame and propagates it through the
rest of the sequence using the motion vectors generated by the SmartVector node. This node, similar to
CornerPin2D, allows you to set keyframes on the image, but uses vector information from SmartVector
rather than tracking information to drive the warp.
See SmartVector and CornerPin2D.
REFERENCE GUIDE
668
VectorCornerPin |
Inputs and Controls

Type Name
Input Source The sequence containing the paint corrections or image and the
source of the motion vectors.
SmartVector The motion vectors generated by the SmartVector node. You can
connect the SmartVector directly or read in the .exr files created by
the node.

Value
VectorDistort Tab
Local GPU gpuName N/A Displays the GPU used for rendering when Use GPU
if available is enabled. Local GPU displays Not
available when:
the Preferences.
there is not enough free memory available on the
GPU.


takes effect.
REFERENCE GUIDE
669
VectorCornerPin |

Value

gpu option.

operating system.
User Keys N/A 0 Displays the total number of user keyframes added
to the sequence.
N/A 0 Displays the current user keyframe.
N/A N/A Click to jump to the previous user keyframe.
N/A N/A Click to jump to the next user keyframe.
N/A N/A Click to add a user keyframe at the current frame in

the sequence.
N/A N/A Click to delete the user keyframe at the current

N/A N/A Click to delete all keyframes in the sequence, both

user and baked.
User1 xy user1 N/A Controls the position of the pins at the current
keyframe.
User2 xy user2 N/A
User3 xy user3 N/A
User4 xy user4 N/A
To Corners
REFERENCE GUIDE
670
VectorCornerPin |

Value
To1 xy to1 N/A Displays the position of the pins at the current
frame, once you've set a user keyframe.
To2 xy to2 N/A
To3 xy to3 N/A
To4 xy to4 N/A
Copy 'From' N/A N/A Click to copy and paste the From1-4 values to the
To1-4 values.
Bake Corners bakeCorners N/A Click to calculate the position of the pins for each
frame specified in the Render dialog. The baked
position is driven by the vectors, independent of the
user keyframes, so the pins need less drastic
correction than between keyframes.
Filter imageFilter Cubic Select the filtering algorithm to use when

original values).
values.
REFERENCE GUIDE
671
VectorCornerPin |

Value

Sinc4 the most.
From Tab
From1 xy to1 N/A Pin 1 controls the bottom left corner of the warp.
From2 xy to2 N/A Pin 2 controls the bottom right corner of the warp.
From3 xy to3 N/A Pin 3 controls the top right corner of the warp.
From4 xy to4 N/A Pin 4 controls the top left corner of the warp.
Set To Input N/A N/A Click to set the From controls to the size of the input
format.
Copy 'To' N/A N/A Click to copy and paste the To1-4 values to the
From1-4 values.
SmartVector Tab
Frame frameDistance 1 frame Sets the trade-off between warping things well for
Distance nearby frames against frames further away. Each shot
has its own optimum frame distance value so try
several values to see what produces the best results.
For example, warping up to 20-30 frames away from

the reference frame, or sequences with rapid
motion, typically require values closer to 1 frame.
Warping over 100 frames away, or sequences with
slower motion, typically require values closer to 64
frames.
Output outputMode warped src Sets the type of output produced by

VectorCornerPin:
REFERENCE GUIDE
672
VectorCornerPin |

Value
• warped src - output the warped Source input

using the reference frame as a template for
warping the rest of the sequence.
• st-map - output the st-map for the warp from the
reference frame to the current frame. This is useful
if you want to bake the results of the node for
later.
• st-map inverse - output the inverse st-map for the
warp, that is, the warp from the current frame to
the reference frame.
Blur Size blurSize 0 Controls the amount of blur applied to the internally
calculated STMap. Increasing the blur size can
remove local distortions in the warped result,
particularly in longer sequences.

aborts.
the render aborts.

REFERENCE GUIDE
673
VectorDistort |
VectorDistort
The VectorDistort node takes the paint from a reference frame and propagates it through the rest of the
sequence using the motion vectors generated by the SmartVector node.
See also SmartVector.
Inputs and Controls

Type Name
Input Source The sequence containing the paint corrections and the source of the
motion vectors.
SmartVector The motion vectors generated by SmartVector node. You can

connect the SmartVector directly or read in the .exr files created by
the node.

Value
VectorDistort Tab
available when:
the Preferences.

REFERENCE GUIDE
674
VectorDistort |

Value

effect.

gpu option.

system.
Reference referenceFrame 1 Sets the frame used to calculate the warp at every
Frame other frame in the sequence.
set to current N/A N/A Click to set the reference frame to the frame currently
frame under the playhead.
Hold Frame holdFrame enabled When enabled, the reference frame is warped instead
of the current frame. This is equivalent to inserting a
FrameHold node into the node tree after the source.
When disabled, the current frame is warped instead of

the reference frame.
Frame frameDistance 1 frame Sets the trade-off between warping things well for
Distance nearby frames against frames further away. Each shot
REFERENCE GUIDE
675
VectorDistort |

Value
has its own optimum Frame Distance value so try

several values to see what produces the best results.
For example, warping up to 20-30 frames away from

the reference frame, or sequences with rapid motion,
typically require values closer to 1. Warping over 100
frames away, or sequences with slower motion,
typically require values closer to 64.
Note: The maximum value for this control

is passed down the node tree from a
SmartVector node's frame distance control.
Output
Output outputMode warped src Sets the type of output produced by VectorDistort:
• warped src - output the warped Src input using the
reference frame as a template for warping the rest
of the sequence.
• st-map - output the st-map for the warp from the
reference frame to the current frame. This is useful if
you want to bake the results of the node for later.
• st-map inverse - output the inverse st-map for the
warp, that is, the warp from the current frame to the
reference frame.
Blur Size blurSize 0 Controls the amount of blur applied to the internally
calculated STMap. Increasing the blur size can remove
local distortions in the warped result, particularly in
longer sequences.
Channels channels rgba Sets the channels to which the warp is applied.
Filter imageFilter Cubic Select the filtering algorithm to apply the warp to the
image, similar to that of the filters used in the IDistort
and STMap nodes.
REFERENCE GUIDE
676
VectorDistort |

Value
values.
Sinc4 the most.
REFERENCE GUIDE
677
3D Nodes
3D nodes deal with Nuke's 3D workspace, which allows you to set up a 3D composite for camera moves,
set replacement, and other applications where you need to simulate a "real" dimensional environment.
AmbientOcclusion
This shader node calculates ambient occlusion for 3D scenes rendered into 2D using the RayRender node.
AmbientOcclusion is not supported by ScanlineRender or PrmanRender.
See also RayRender.
Inputs and Controls

Connection Type Connection Name Function
Input unnamed Either:

• The 2D image you’re using for the surface texture, or
• Another shader node, such as Diffuse, Specular, or
Emission. Adding several shader nodes one after the
other allows you to produce more complex effects.
samples samples 10 Controls the number of rays used to detect if

an object is occluded. Increasing the number
of samples reduces noise and produces a
better quality result, but increases render
times.
spread spread 1 Controls the angular spread around the normal

in the range 0-1, where 1 is equal to 90° (the
REFERENCE GUIDE
678
ApplyMaterial |
full hemisphere).
falloff falloff 0 Sets the exponential falloff rate for the

occlusion along the ray distance.
near clip near 0 Controls the minimum (near) and maximum

(far) occlusion distances sampled when
far clip far 10 calculating ambient occlusion.
ApplyMaterial
Apply a material from the mat input to your 3D object(s). For example, you can use this node to apply a
global material to several merged objects (note that this overrides any individual materials applied to the
geometry before it was merged).
Inputs and Controls

Type Name
Input unnamed The 3D geometry object to which you want to apply a material. If you
want to apply the same material to several objects, you can also
connect a MergeGeo node here.
mat The material you want to apply to your 3D object, for example, a
BasicMaterial node.

(Scripting) Value
ApplyMaterial Tab
display display unchanged Adjust the display characteristics of the 3D object. These
settings don’t affect the render output of the scene;
REFERENCE GUIDE
679
ApplyMaterial |

(Scripting) Value
these are for display purposes only in the 3D Viewer.

• off - hides the 3D object.
• wireframe - displays only the outlines of the object’s
geometry.
• solid - displays all geometry with a solid color.
• solid+wireframe - displays the geometry as solid
color with the object’s geometry outlines.
• textured - displays only the surface texture.
• textured+wireframe - displays the wireframe plus the
surface texture.
• unchanged - doesn't change the object's display
mode. The Viewer can override this setting.
selectable selectable enabled When enabled, you can make selections as normal in the
Viewer. When disabled, points cannot be selected or
changed.
render render_mode unchanged Sets how the object will render. This control is
independent from the display selection, but has the
same settings.
filter filter_type all Select the geometry objects to which the material is
applied:
• all - apply the material to all incoming geometry
objects.
• name - only apply the material to incoming geometry
objects that match the filter name settings.
Note: For name to work, the geometry has to

have a name attribute. The only nodes that add
this attribute are ModelBuilder and ReadGeo
(when reading in an Alembic and USD files),
but third-party plug-ins can add it too.
filter_name_ contains Sets how to filter the incoming geometry objects to
REFERENCE GUIDE
680
Axis |

(Scripting) Value
match determine which objects the material is applied to:

• equals - set the material on any objects whose name
matches the string in the filter name field exactly.
• doesn't equal - set the material on any objects whose
name does not match the string in the filter name field
exactly.
• contains - set the material for any objects whose name
contains the string in the filter name field.
This can be useful when you have some structure to

your object names. For example, if you have objects
like /Root/Chair/Seat, /Root/Chair/Back, and
/Root/Table, you can select contains and set the filter
name field to chair to apply the material to all parts of
the chair while leaving the table alone.
• doesn't contain - set the material for any objects
whose name does not contain the string in the filter
name field.
This control is only available when filter is set to name.
filter name N/A The filter name to use to determine which objects the
material is applied to. You can type the filter name
directly into this field or use the choose button to select
a filter name from a list of incoming geometry objects.
choose filter_refresh N/A Open the Object Name Chooser dialog. This allows you
to select a filter name from a list of incoming geometry
objects.
Axis
REFERENCE GUIDE
681
Axis |
Axis nodes act as null objects by adding a new transformational axis to which other objects may be
parented. Even when objects already have their own internal axes, it’s sometimes useful to parent in a
separate axis. For example, when an Axis node is parented to other objects in a scene, the node globally
controls the scene - rotating the axis rotates all objects in the scene.
Inputs and Controls

Type Name
Input axis An optional Axis node input. This links the position, rotation, scale, and
skew of the transformed 3D object(s) to the Axis node, so that the
transformation controls on the Axis node override the corresponding
controls on the TransformGeo node.
If you’ve worked with other 3D applications, you may know the Axis
node as a “null” or “locator” object.
look An optional input where you can connect a Camera, Light, or another
Axis that the Axis is automatically rotated to point toward. The Axis
automatically rotated to point towards the connected input whenever
the look input is moved.

(Scripting) Value
Axis Tab
read from file read_from_ disabled When disabled, the axis is configured using the controls
file on the Axis tab. Enable read from file if you want to
read in axis information from an .fbx file using the File
tab.
Note: This control is linked to the read from

file control on the File tab.
REFERENCE GUIDE
682
Axis |

(Scripting) Value
display display wireframe Adjust the display characteristics of the 3D object. These
• off - hides the 3D geometry object.
geometry.
• solid + lines - displays the geometry as solid color
with the object’s geometry outlines.
• textured + lines - displays the wireframe plus the
surface texture.
Note: The Viewer node 3D > display setting

overrides this control.
changed.
file_menu N/A Select to import or export a channel file:

• Import chan file - import a channel file and transform
the objects attached to the Axis node according to the
transformation data in the channel file. Channel files
contain a set of Cartesian coordinates for every frame
of animation in a given shot. You can create and export
them using Nuke or 3D tracking software, such as 3D-
Equalizer, Maya, or Boujou.
• Export chan file - export the translation parameters
that you’ve applied to the Axis node as a channel file.
This is a useful method of sharing setups between
artists.
REFERENCE GUIDE
683
Axis |

(Scripting) Value
snap_menu N/A • Match selection position - the objects attached to the

Axis are snapped to a new position depending on the
points selected.
• Match selection position, orientation - the objects
attached to the Axis are snapped to a new position and
orientation depending on the points selected.
• Match selection position, orientation, size - the
objects attached to the Axis are snapped to a new
position, orientation, and size depending on the points
selected.
transform xform_order SRT Sets the operation order for scale (S), rotation (R), and
order translation (T). The possible operation combinations are
SRT, STR, RST, RTS, TSR, TRS.
rotation order rot_order ZXY Sets the order of rotation. The possible axial
combinations are ZXY, XYZ, XZY, YXZ, YZX, ZXY, ZYX.
translate translate 0, 0, 0 Lets you translate the 3D objects attached to the Axis
node along the x, y, and z axes. You can also adjust
translate values by clicking and dragging the axis in the
3D Viewer.
rotate rotate 0, 0, 0 Lets you rotate the 3D objects attached to the Axis node
around the x, y, and z axes. You can adjust rotate values
by holding down Ctrl/Cmd and dragging in the 3D
Viewer.
scale scaling 1, 1, 1 Lets you scale the 3D objects attached to the Axis node
on the x, y, and z axes.
uniform scale uniform_scale 1 Lets you scale the 3D objects attached to the Axis node
simultaneously on the x, y, and z axes.
skew skew 0, 0, 0 Lets you skew the 3D objects attached to the Axis node
on the x, y, and z axes.
pivot pivot 0, 0, 0 When you make changes to a 3D object’s position,

scaling, skewing, and rotation, these occur from the
REFERENCE GUIDE
684
Axis |

(Scripting) Value
location of the object’s origin point or pivot. The pivot

x, y, and z controls allow you to offset the pivot point
and move it anywhere you like - you can even move it
outside of the object. Subsequent transformations
applied will then occur relative to the new pivot point
location.
You can also hold down Ctrl/Cmd+Alt and drag the

pivot point to a new location in the 3D Viewer.
Local Matrix
specify matrix useMatrix N/A Enable this control to specify matrix values for the
object you’re transforming as an alternative to setting
transform, scale, skew and pivot values above.
matrix matrix N/A The matrix displays values from the object’s transform,
rotate, scale, skew, and pivot controls.
Check specify matrix and copy or drag-and-drop matrix

values from another object to apply those values, for
example, if you wanted to align objects in a scene.
World Matrix
World matrix world_matrix N/A Displays the world or absolute xyz transform of the node
in world coordinates.
Note: Unlike the Local matrix, you can’t adjust

the World matrix manually.
File Tab
read from file read_from_ disabled When disabled, the axis is configured using the controls
file on the Axis tab. Enable read from file if you want to
read in axis information from an .fbx file.
REFERENCE GUIDE
685
Axis |

(Scripting) Value

file control on the Axis tab.
file file N/A Sets the file path for the .fbx file from which you intend
to import axis information.
reload reload N/A Click to reload the axis information from the specified
file.
animation fbx_take_ N/A When the file control is pointing to a valid .fbx file,
stack name name select the required take name from the dropdown
menu.
node name fbx_node_ N/A When the file control is pointing to a valid .fbx file,
name select the required node name from the dropdown
menu.
frame rate frame_rate 24 When use frame rate is enabled, enter the required
frame rate to use instead of the rate specified in the
input file.
use frame rate use_frame_ disabled When enabled, the frame rate from the input file is
rate ignored and the specified frame rate is used instead.
Look Tab
look axis look_axis -Z The axis around which the Axis is rotated to face the
look input.
rotate X look_rotate_x enabled Determines whether the rotation occurs around the X
axis. Note that for the rotation to truly "look at" the look
input, all three of these options must be activated.
rotate Y look_rotate_y enabled Determines whether the rotation occurs around the Y
input, you have to set all three of these options.
rotate Z look_rotate_z enabled Determines whether the rotation occurs around the Z
REFERENCE GUIDE
686
BasicMaterial |

(Scripting) Value
look strength look_strength 1 Determines to what extent the Axis rotates to the new
orientation. The smaller the value, the less the object is
rotated. Setting the value to 0 produces no rotation.
use look_use_ disabled Uses an alternate scheme to calculate the look rotation.
quaternions quaternions This option may be useful to smooth out erratic
rotations along the look axis.
BasicMaterial
This node lets you control what material your objects seem to be made of. It combines the Diffuse,
Specular, and Emission nodes, allowing you to control all three aspects of the material with a single node.
Note: To see the effect of your changes to an object’s material properties, you need to have at
least one light node in your Scene. You also need to enable transparency and lighting in the
Viewer settings. Press S over the 3D Viewer, and check transparency and headlamp on the 3D
tab.
Inputs and Controls

Type Name
Input mapSh An optional mask for varying the shininess value. Where the mask is
black, the shininess is set to min shininess. Where the mask is white,
the shininess is set to max shininess. Values in between (where the
matte is gray) are attenuated accordingly.
mapS An optional mask for limiting the effect of the specular component.
Any changes you make to specular are limited to the non-black areas
of the mask.
REFERENCE GUIDE
687
BasicMaterial |

Type Name
mapE An optional mask for limiting the effect of the emissive component.
Any changes you make to emission are limited to the non-black areas
of the mask.
mapD An optional mask for limiting the effect of the diffuse component.
Any changes you make to diffuse are limited to the non-black areas
of the mask.
unnamed Either:
• Another shader node, such as Diffuse, Specular, or Emission. Adding
several shader nodes one after the other allows you to produce
more complex effects.

(Scripting) Value
BasicMaterial Tab
channels channels rgb The effect is only applied to these channels.

channels.
emission emission 0 Sets the color of the light the material emits. Note that
when you have an image connected to the unnamed
input of the BasicMaterial node and adjust this value, you
need to look at the rendered 2D image to see the effect
of your changes. Changing the emission value does not
have any effect in the 3D Viewer.
diffuse diffuse 0.18 Sets the color of the material when illuminated. Note
that when you have an image connected to the unnamed
input of the BasicMaterial node and adjust this value, you
need to look at the rendered 2D image to see the effect
of your changes. Changing the diffuse value does not
have any effect in the 3D Viewer.
REFERENCE GUIDE
688
BasicMaterial |

(Scripting) Value
specular specular 0.8 Sets how bright the highlights on the material seem.
min shininess min_shininess 10 Sets the shininess value controlling the width of the
highlights. The higher the value, the wider the highlights.
If you have connected a mask to the mapSh input of the

node, pixel values in the mask are used to vary the
shininess value. Where the matte is black, the shininess is
set to min shininess. Where the matte is white, the
shininess is set to max shininess. Values in between
(where the matte is gray) are attenuated accordingly.
If you’re not using the mapSh input, the average of min

shininess and the maxshininess value is used as the
shininess value for the material.
max shininess max_shininess 10 Sets the shininess value controlling the width of the
highlights. The higher the value, the wider the highlights.


shininess shininess_ luminance Selects which channel from the mapSh input is used to
channel channel map the black and white values to the minshininess and
maxshininess controls. Choose:
• red - use the red channel for the mapping.
• green - use the green channel.
• blue - use the blue channel.
• alpha - use the alpha channel.
• luminance - use the luminance.
REFERENCE GUIDE
689
BlendMat |

(Scripting) Value
• average rgb - use the average of the red, green, and

blue channels.
BlendMat
The BlendMat node sets how the pixels colored by the material it is applied to combine with the pixels
from objects behind. It is like the MergeMat node, but instead of blending with another material, it blends
with whatever is rendered behind in the 3D scene.
Note: Multiple BlendMat operations on the same layer may produce unexpected results.
Inputs and Controls

Type Name
Input unnamed The 2D image you’re using for the material to merge with the
background pixels in the 3D scene.

(Scripting) Value
BlendMat Tab
REFERENCE GUIDE
690
BlendMat |

(Scripting) Value

operation operation over Select how you want to composite the BlendMat node’s
input material and the background pixels together:
• none - set the material to black.
• replace - show the material where the material and the
background overlap.
• over - composite the material over the background
pixels according to the material’s alpha.
• stencil - show the background pixels where the
material’s alpha is black. Where the material’s alpha is
white, the material is set to black. For this to work, the
BlendMat node needs to process the alpha channel, so
set channels to rgba. This operation is the opposite of
mask.
• mask - show the background pixels where the
material’s alpha is white. Where the material’s alpha is
black, the material is also set to black. For this to work,
the BlendMat node needs to process the alpha channel,
so set channels to rgba. This operation is the opposite
of stencil.
• plus - add the background pixels to the material.
• max - use the material if its pixel values are greater
than the background pixels or else use the background
pixels.
• min - use the material if its pixel values are less than
the background pixels or else use the background
pixels.
surface blend surfaceblend input Determines how the input texture blends with the vertex
fragment color of the geometry to which the shader is applied:
• input texture - only use the input texture.
• vertex color - only use the vertex color.
• modulate - texture * vertex color.
REFERENCE GUIDE
691
Camera |

(Scripting) Value
• plus - texture + vertex color.

• over - (texture + ((1 - a) * vertex color)).
• subtract - texture - vertex color.
Camera
Cameras may be connected to either the Scene node or the ScanlineRender node. Camera connected to a
ScanlineRender node define the projection for use by the 3D renderer. You can connect additional
cameras to the Scene node and switch between them by choosing the viewing camera from the
dropdown menu at the top of the Viewer. You can also read in .fbx and .abc scene files containing
standard cameras, though only one camera per Camera node can be read.
Cameras can also be used to project 2D textures onto 3D objects in a scene using the controls on the
Projection tab and a Project3D node.
Inputs and Controls

Type Name
Input axis An optional Axis node input. This links the position, rotation, scale,
and skew of the transformed 3D object(s) to the Axis node, so that
the transformation controls on the Axis node override the
corresponding controls on the TransformGeo node.
look An optional input where you can connect a Camera, Light, or Axis that
REFERENCE GUIDE
692
Camera |

Type Name
the Light is automatically rotated to point toward. The Light is

automatically rotated to point towards the connected input whenever
the look input is moved.

(Scripting) Value
Camera Tab
read from file read_from_ disabled When disabled, the camera is configured using the
file controls on the Camera tab. Enable read from file if you
want to read in camera information from an .fbx file
using the File tab.

file control on the File tab.
display display wireframe Adjust the display characteristics of the camera. These
• off - doesn't display the camera geometry.
• wireframe - displays only the outlines of the camera
geometry.
• solid - displays the camera geometry with a solid
color.
• solid + lines - displays the camera geometry as solid
color with the camera’s geometry outlines.
• textured + lines - displays the wireframe plus the
surface texture.
changed.
REFERENCE GUIDE
693
Camera |

(Scripting) Value

the camera according to the transformation data in the
channel file. Channel files contain a set of Cartesian
coordinates for every frame of animation in a given
shot. You can create and export them using Nuke or 3D
tracking software, such as 3D-Equalizer, Maya, or
Boujou.
that you’ve applied to the camera as a channel file. This
is a useful method of sharing setups between artists.
snap_menu N/A • Match selection position - the camera is snapped to a

new position depending on the points selected.
• Match selection position, orientation - the camera is
snapped to a new position and orientation depending
on the points selected.
camera is snapped to a new position, orientation, and
size depending on the points selected.
translate translate 0, 0, 0 Lets you translate the camera along the x, y, and z axes.
You can also adjust translate values by clicking and
dragging the camera in the 3D Viewer.
rotate rotate 0, 0, 0 Lets you rotate the camera around the x, y, and z axes.
You can adjust rotate values by holding down Ctrl/Cmd
and dragging in the 3D Viewer.
scale scaling 1, 1, 1 Lets you scale the camera on the x, y, and z axes.
uniform scale uniform_scale 1 Lets you scale the camera simultaneously on the x, y, and
REFERENCE GUIDE
694
Camera |

(Scripting) Value
z axes.
skew skew 0, 0, 0 Lets you skew the camera on the x, y, and z axes.
pivot pivot 0, 0, 0 When you make changes to a camera’s position, scaling,

skewing, and rotation, these occur from the location of
the camera’s origin point or pivot. The pivot x, y, and z
controls allow you to offset the pivot point and move it
anywhere you like - you can even move it outside of the
object. Subsequent transformations applied will then
occur relative to the new pivot point location.

Local Matrix
camera you’re transforming as an alternative to setting
matrix matrix N/A The matrix displays values from the camera’s transform,

World Matrix
World matrix world_matrix N/A Displays the world or absolute xyz transform of the node

Projection Tab
projection project_mode perspective Sets the camera projection type when projecting 2D
textures:
REFERENCE GUIDE
695
Camera |

(Scripting) Value
• perspective - objects in front of the camera have the

illusion of depth defined by the camera's focal-length
and aperture.
• orthographic - objects are viewed using parallel
projection.
• uv - every object renders its UV space into the output
format. Use this to cook out texture maps.
• spherical - the entire 360° world is rendered as a
spherical map.
focal length focal 50 Adjusts the camera’s level of magnification.
horiz aperture haperture 24.576 Sets the horizontal aperture of the camera.
vert aperture vaperture 18.672 Sets the vertical aperture of the camera.
near near 0.1 Adjusts the position of the camera’s forward clipping
plane. Objects closer to the camera than this plane are
not rendered.
Note: The value for the near clipping plane

must always be positive to produce a sensible
result.
far far 10000 Adjusts the position of the camera’s rearward clipping
plane. Objects farther from the camera than this plane
are not rendered.
window win_translate 0, 0 Translates the camera’s output along the uv axes.

translate uv
window scale win_scale 1, 1 Scales the camera’s output along the uv axes.
uv
window roll winroll 0 Rotates the camera’s output around the z axis.
focal distance focal_point 2 Controls the distance from the lens where the camera
focuses.
REFERENCE GUIDE
696
Camera |

(Scripting) Value
fstop fstop 16 Sets the f-stop value (relative aperture) of the camera.
File Tab
read from file read_from_ disabled When disabled, the camera is configured using the
file controls on the Camera tab. Enable read from file if you
want to read in camera information from an .fbx or .abc
file using the File tab.

file control on the Camera tab.
file file N/A Sets the file path for the .fbx or .abc file from which you
intend to import camera information.
reload reload N/A Click to reload the camera information from the
specified file.
animation fbx_take_ N/A When the file control is pointing to a valid .fbx or .abc
stack name name file, select the required take name from the dropdown
menu.
node name fbx_node_ N/A When the file control is pointing to a valid .fbx or .abc
name file, select the required node name from the dropdown
menu.
input file.
compute compute_ disabled When enabled, compute the camera rotation values
rotation rotation using the look up vector and look at position. The
rotation values are always computed when there is a
look at target.
Look Tab
REFERENCE GUIDE
697
CameraTracker |

(Scripting) Value
look axis look_axis -Z The axis around which the Camera is rotated to face the
look input.
look strength look_strength 1 Determines to what extent the Camera rotates to the
new orientation. The smaller the value, the less the
object is rotated. Setting the value to 0 produces no
rotation.
CameraTracker
CameraTracker (NukeX and Nuke Studio only) is designed to provide an integrated camera tracking or
matchmoving tool, allowing you to create a virtual camera whose movement matches that of your original
camera. Tracking camera movement in a 2D footage enables you to add virtual 3D objects to your 2D
footage.
REFERENCE GUIDE
698
CameraTracker |
Inputs and Controls

Type Name
Input Source The sequence whose camera movement you want to track.
Mask An optional input to use as a mask. By default, feature tracking is

The Mask input is hidden until the Source input is connected.

Value
CameraTracker Tab
Input
Source sourceType Sequence Sets the type of source footage you intend to
track:
• Sequence - designed to track a contiguous
frame sequence with minimal intervention
from you.
• Stills - creates a camera track from
reference frame stills taken on site. These
represent the minimum number of frames
required to cover your subject with
sufficient overlap.
Mask ignore None Set the mask type to exclude areas of the
sequence when tracking:
• None - none of the footage is ignored.
• Source Alpha - use the alpha channel of the
source clip to define which areas to ignore.
• Source Inverted Alpha - use the inverted
alpha channel of the source clip to define
which areas to ignore.
REFERENCE GUIDE
699
CameraTracker |

Value
• Mask Luminance - use the luminance of the

Mask input to define which areas to ignore.
• Mask Inverted Luminance - use the
inverted luminance of the Mask input to
define which areas to ignore.
• Mask Alpha - use the Mask input alpha
channel to define which areas to ignore.
• Mask Inverted Alpha - use the inverted
Mask input alpha channel to define which
areas to ignore.
Range trackRange Input Sets the range of frames processed by

CameraTracker:
• Input - the default value, sets the frame
range to the length of the sequence
attached to the Source input.
• Global - sets the tracking frame range to
the range set in the Project Settings >
frame range controls. If no frame range is
defined, the frame range of the first image
you read in is used as the Global frame
range.
• Custom - sets a tracking range of frames
described by the from and to fields.
• Reference Frames - allows you to manually
define keyframes in the sequence using the
+/- buttons. You can also add keyframes to
all frames or a specific frame range using
the Add dropdown. This option should only
be used if you have set Source to Stills.
from trackStart Dependent on When Range is set to Custom, this control

the Project sets the first frame to track in the footage.
Settings
to trackStop Dependent on When Range is set to Custom, this control
REFERENCE GUIDE
700
CameraTracker |

Value
the Project sets the last frame to track in the footage.

Settings
Frames referenceFrames 0 Displays the reference frames to track and

solve.

when Range is set to Reference
Frames.
addReferenceFrame N/A Adds the current frame to the set of frames

to track and solve.

Frames.
deleteReferenceFrame N/A Removes the current frame from the set of

frames to track and solve.

Frames.
Add addReference- Add Allows you to quickly add reference frames:

FramesMenu • Add All - Adds all frames based on the
input's start and end range. Any missing
frames on the input clip are skipped.
• Add Range - Opens a dialog that lets you
select a specific frame range to add.
REFERENCE GUIDE
701
CameraTracker |

Value

Frames.
Delete deleteReference- Delete Allows you to quickly remove reference

FramesMenu frames:
• Delete All - Removes all reference frames.
• Delete Range - Opens a dialog that lets
you select a specific frame range to remove.

Frames.
Camera
Principal View principalView left Sets the principal view where features are
tracked and the Mask input is applied. The
secondary camera is calculated relative to this
Principal View camera.
This control also appears on the Settings tab.
Note: Principal View is only

displayed if you have set up more
than one view in your Project
Settings (typically, if you’re working
on a stereoscopic project).
Camera cameraMotionLink Free Camera Sets the type of camera motion to

Motion compensate for:
• Rotation Only - select this if the camera is
stationary, but rotating.
• Free Camera - select this if the camera is
REFERENCE GUIDE
702
CameraTracker |

Value
moving freely, rotating, and translating.

• Linear Motion - select this if the camera
motion is a straight, linear path.
• Planar Motion - select this if the camera
has a flat path, moving in a two-dimensional
plane only.
Lens lensDistotionTypeLink No Lens Specifies the type of lens distortion

Distortion Distortion expected:
• No Lens Distortion - treats the footage as
having no distortion. Use this option if the
Source footage has already been corrected
to remove lens distortion.
• Unknown Lens - calculates the lens
distortion automatically from the sequence
(in the same way as the Image Analysis tab
in the LensDistortion node) and then refines
the distortion in the camera solve.
Undistort undistort disabled When enabled, the Source is undistorted

Input based on the calculated lens.
Focal Length focalLengthType Unknown Sets the focal length for the camera:
Constant • Known - select this option if the focal
length is available and enter a value in the
Length control.
• Approximate Varying - select this option if
an approximate focal length is available and
enter keyframed focal length values in the
Length control.
• Approximate Constant - select this option
if an approximate focal length is available
and there is no zoom, and enter a focal
length value in the Length control.
REFERENCE GUIDE
703
CameraTracker |

Value
Note: CameraTracker attempts to

refine the focal length during the
solve if you select an Approximate
option.
• Unknown Varying - select this option if

the focal length is unknown and changing.
• Unknown Constant - this is the default
option. Use this option if the focal length is
unknown and there is no zoom.
Length knownFocalLength 50 Sets the focal length for approximate and

known solves. You can animate this control to
define a varying focal length. The units are
the same as those used for the Film Back Size
(mm or inches).
Film Back filmBackSizePresets Custom Sets the Film Back Size controls
Preset automatically, depending on which preset
you choose.
The default setting allows you to enter your

own settings manually.
Tip: You can add your own defaults

by editing the
../NukeScripts/camerapresets.py
file in the Nuke installation package.
Film Back Size filmBackSize 36,24 Sets the size of the camera imaging sensor.
xy Specify the units you want to use by selecting
either millimeters or inches in the Units
dropdown. The units should match the units
used for the Length.
REFERENCE GUIDE
704
CameraTracker |

Value
Units filmbackUnits mm Sets the unit of measure for the Film Back
Size, either millimeters or inches.
Analysis
Track trackFeatures N/A Click to begin auto-tracking using the

controls specified in the Settings tab. Track
defines a set of 2D feature tracks that
correspond to fixed points, or features, in the
scene.
Update Track updateTrack N/A Click to display a frame range dialog,

allowing you to retrack part of the sequence,
such as when you have extended the frame
range or used some of the Settings tab
Tracking refinement controls.
Clear Tracks clearTrack N/A Click to clear all auto-tracking data. A

confirmation message needs acknowledging
before the action is completed.
Solve solveCamera N/A Click to begin calculating the solve using the
controls specified in the Settings tab. Solve
calculates the camera path and projection
creating a 3D point for each 2D feature track
within a minimum projection error, including
auto-tracks, User Tracks, and tracks created
from imported Trackers.
Update Solve updateSolve N/A Click to update the solve calculation, such as
when you have extended the frame range or
used some of the Settings tab Solving
refinement controls.
Recalculating the solve is handy when you

want to see how good the solve is after
you’ve refined it, without permanently
deleting tracks. Recalculating the solve also
recalculates stereo geometry.
REFERENCE GUIDE
705
CameraTracker |

Value
Clear Solve resetSolve N/A Click to clear all solve data. A confirmation
message needs acknowledging before the
action is completed.
Error solveRMSE N/A Displays the RMS (root mean square) error of
your solve, in pixels. As a general rule of
thumb, if your solve reports an RMS
projection rate higher than 1.0 pixel, you may
want to consider fine-tuning your solve or
tracking data.
per frame solveError N/A Displays the RMS (root mean square) error of
your solve at each frame, in pixels.
Export
[export menu] exportMenu Camera Determines how the track and solve data is
used when you click Create:
• Camera - creates a single animated camera.
• Camera rig - creates an animated camera
for each view in the script and a JoinViews
to node to pass the animation downstream.
• Scene - creates a single animated camera, a
point cloud from the solved 3D points, and
a Scene node.
• Scene+ - creates a scene containing all the
components included in the other create
modes.
• Point cloud - creates a point cloud from
the solved 3D points.
• Distortion - creates a LensDistortion node
with Undistort disabled.
• Undistortion - creates a LensDistortion
node with Undistort enabled.
• Cards - creates a 3D card for each solved
REFERENCE GUIDE
706
CameraTracker |

Value
Note: Creating Cards from

sequences can take long time,
depending on the number of
frames present.
Create exportButton N/A Click to create the option specified in the

export dropdown.
Link output linkOutput enabled When enabled, cameras and points are
updated when the solve is updated.
If you disable linking, the expression link

between the CameraTracker and Camera
nodes is disconnected.
UserTracks Tab
User Track userTrackViews Dependent on You can only set up User Tracks in a
Views the Project maximum of two views. This control allows
Settings you to select which views correspond with
the left (lx, ly) and right (rx, ry) 2D User Track
location columns.
This control is only displayed if you have set

up more than one view in your Project
Settings (typically, if you’re working on a
stereoscopic project).
[user tracks userTracks N/A Displays positional information for any User
table] Tracks in the script.
Add Track addUserTrack N/A Click to add a User Track in the center of the
Viewer.
Delete Tracks deleteUserTracks N/A Click to delete the current selections in the
tracks list.
Autotrack autotrackUserTracks N/A Click to begin auto-tracking the current

selections in the tracks list.
REFERENCE GUIDE
707
CameraTracker |

Value
Update XYZ updateUserTrackXYZ N/A Click to recalculate the XYZ coordinates of

the current selections in the tracks list.
Import Tracks importUserTracks N/A Click to import previously exported User

Tracks.
Export Tracks exportUserTracks N/A Click to export the current selections in the
tracks list.
Import importTracker N/A Click to import track information from a

Tracker Tracker node within the script.
Export Tracker exportTracker N/A Click to export the current selections in the
tracks list to a Tracker node within the script.
AutoTracks Tab
Solve Error SolveRMSE N/A Displays the RMS (root mean square) error of
your solve, in pixels. As a general rule of
thumb, if your solve reports an RMS
projection rate higher than 1.0 pixel, you may
want to consider fine-tuning your solve or
tracking data.
per frame solveError N/A Displays the RMS (root mean square) error of
your solve at each frame, in pixels.
Curves View curvesView Dependent on Sets the view for the track curves.
the Project
Settings This control is only displayed if you have set
[track curves] trackCurves N/A Displays the following track and solve
information in the form of curves:
• num tracks - the number of tracked
features at each frame.
• track len - min - the minimum length of
REFERENCE GUIDE
708
CameraTracker |

Value
the tracks at each frame (in frames).

• track len - avg - the average length of the
tracks at each frame (in frames).
• track len - max - the maximum length of
the tracks at each frame (in frames).
• Min Length - the threshold for minimum
track length. You can adjust the minimum
using the Min Length control.
• Solve Error - displays the constant Solve
Error parameter.
• error min - the minimum reprojection error
at each frame (in pixels).
• error rms - the root mean reprojection
error at each frame (in pixels).
• error track - the maximum root mean
reprojection error calculated over the track
lifetime at each frame (in pixels).
• error max - the maximum reprojection
error at each frame (in pixels).
• Max Track Error - displays the constant
Max RMS Error parameter. You can adjust
the maximum using the Max Track Error
control.
• Max Error - displays the Max Error
threshold parameter. You can adjust the
maximum using the Max Error control.
REFERENCE GUIDE
709
CameraTracker |

Value
Min Length minLengthThreshold 0 Redefines your thresholds if you find your

solve or camera output needs adjusting:
• Min Length - increase the minimum length
threshold to reject short tracks. You might
find a lot of short tracks cropping up in
long sequences with a slow camera
movement.
Max Track maxRMSEThreshold 0
• Max Track Error - reduce this threshold to
Error
reject tracks based on RMS reprojection
error.
• Max Error - reduce this threshold to reject
tracks with a large reprojection error in
isolated frames.
Max Error maxErrorThreshold 0

Note: Feature tracks that fall within
the threshold limits are called
inliers, and are used when you
recalculate a solve instead of
starting from scratch.
Refinement
Refine Solve refineSolve N/A Click to attempt to refine the solve

calculation using the methods outlined in the
Focal Length, Position, and Rotation
controls.
Focal Length refineOptions disabled When enabled, clicking Refine Solve causes
CameraTracker to finess the solve calculation
using an updated Focal Length.
Position refineOptions disabled When enabled, clicking Refine Solve causes

using an updated camera position.
Rotation refineOptions disabled When enabled, clicking Refine Solve causes
REFERENCE GUIDE
710
CameraTracker |

Value

using an updated camera orientation.
Delete deleteInvalidTracks N/A Click to permanently delete tracks for which

Unsolved 3D points could not be calculated in the
solve.
Delete deleteRejectedTracks N/A Click to delete any tracks rejected by the

Rejected threshold controls.
Delete Auto- deleteAutoTracks N/A Click to delete all auto-tracks from the script,
tracks while retaining User Track and solve data. You
can use this button to reduce the size of the
script once you're happy with the solve.
Settings Tab
Features
Number of numberFeatures 150 Sets the number of features to track in each

Features frame - ideally you should use more than 100
tracks per frame.
Detection featureThreshold 0.1 Sets the distribution of features over the

Threshold input image. Low values track features evenly
on all parts of the image.
Feature featureSpearation 12 Sets the distribution of features in relation to

Separation each other. High values spread features at
even distances over the image.
Refine Feature refineFeatures disabled When enabled, lock detected features to

Locations local corners. CameraTracker finds the closest
corner point in your footage and locks
feature points to them.
Preview previewFeatures disabled When enabled, preview the potential features

Features available for tracking. This allows you to
check that the features cover the image. If
not, tweak the tracking controls before
REFERENCE GUIDE
711
CameraTracker |

Value
tracking.
Tracking
Minimum minTrackLength 3 Sets a threshold value for the minimum track

Length length to reject short tracks. You might find a
lot of short tracks cropping up in long
sequences with a slow camera movement.
Track matchThreshold 0.69999999 Controls how similar features look over a

Threshold number of frames. You can adjust this value
to test whether a track is reliable.
Track smoothnessThreshold 0.1 Sets the threshold for smooth track

Smoothness generation. Adjusting this value can be useful
in preventing poor tracks in complex
sequences. Increase the smoothness value to
remove tracks that glitch over time.
Track consistencyThreshold 0 Sets the threshold for consistent track

Consistency generation. Increase this value to ensure track
motion is locally consistent. Adjust
consistency to prevent poor tracks in
complex sequences.
Solving
Principal View principalView left Sets the principal view where features are
tracked and the Mask input is applied. The
secondary camera is calculated relative to this
Principal View camera.
This control also appears on the

CameraTracker tab.
REFERENCE GUIDE
712
CameraTracker |

Value
Note: Principal View is only

displayed if you have set up more
than one view in your Project
Settings (typically, if you’re working
on a stereoscopic project).
Camera cameraMotion Free Camera Sets the type of camera motion to

Motion compensate for:
• Rotation Only - select this if the camera is
stationary, but rotating.
• Free Camera - select this if the camera is
moving freely, rotating, and translating.
• Linear Motion - select this if the camera
motion is a straight, linear path.
• Planar Motion - select this if the camera
has a flat path, moving in a two-dimensional
plane only.
Keyframe keyFrameSeparation 0.30000001 Controls the separation between keyframes.

Spacing Use a high separation to spread keyframes
out in long sequences with small camera
movements. Use a low separation to generate
more keyframes for fast camera moves.
Smoothness motionSmoothness 0 Adjusts the smoothness of your camera path.

Increase this value to add weighting to the
camera path and create a smoother result.
Reference rootReferenceFrame 0 When Set reference frame is enabled,

Frame specifies the first frame to use as a keyframe
in the solve. This should be a frame where
there is a large number of tracks distributed
over the image with a good variation in
depth.
Set reference setRootFrame disabled When enabled, use the Reference Frame field
REFERENCE GUIDE
713
CameraTracker |

Value
frame to manually define how keyframes are

specified in the sequence. This can be useful
if you have a difficult sequence to solve.
Stereo (These controls are only displayed if you have set up more than one view in your Project
Settings, typically when working on a stereoscopic projects.)
Aligned alignSecondary enabled When enabled, the solve aligns secondary

Stereo Cameras cameras to the principal camera position.
Cameras
Constant constantInterAxial enabled When enabled, a constant interaxial distance

Interaxial Distance is maintained between views.
Distance
Interaxial interAxialDistance 1 Sets a known interaxial distance for stereo

Distance views to define the scene scale.
This control is only displayed if you have set

Constant constantInterAxial enabled When enabled, a constant interaxial

Interaxial Convergence convergence is maintained between views.
Covergence
Display
Show tracks displayTracks enabled When enabled, the individual tracks

generated during analysis are displayed in the
Viewer.
Show displayReprojected enabled When enabled, circular highlights are

projected 3D displayed around the reprojected points in
point the 2D Viewer. Hover over points to display
track length and reprojection error
information.
Show key keyframePointsOnly disabled When enabled, only the longest tracks used
REFERENCE GUIDE
714
CameraTracker |

Value
tracks only to calculate the solve are displayed in the

Viewer.
Show 3D displayMarker disabled When enabled, cone shaped 3D markers are

marker displayed on your points in the 3D Viewer.
Note: When Show 3D marker is

enabled, set the size of the markers
using the scale controls.
Scene Tab
[scale scaleConstraints N/A Displays known distances for scale constraints

constraints against a measured distance between two
table] solved points in the scene.
Delete deleteScaleConstraints N/A Click to delete the selected scale constraints.

Constraints
Scene Transform

• Import chan file - import a channel file and
transform the input object according to the
transformation data in the channel file.
Channel files contain a set of cartesian
coordinates for every frame of animation in
a given shot. You can create and export
them using Nuke or 3D tracking software,
such as 3D-Equalizer, Maya, or Boujou.
• Export chan file - export the translation
parameters that you’ve applied to the input
object as a channel file. This is a useful
method of sharing setups between artists.
REFERENCE GUIDE
715
CameraTracker |

Value
snap_menu N/A • Match selection position - the scene is

snapped to a new position depending on
the points selected.
• Match selection position, orientation -
the scene is snapped to a new position and
orientation depending on the points
selected.
• Match selection position, orientation, size
- the scene is snapped to a new position,
orientation, and size depending on the
points selected.
combinations are ZXY, XYZ, XZY, YXZ, YZX,
ZXY, ZYX.
translate xyz translate 0,0,0 Translates the scene along the x, y, and z axes.
You can also adjust translate values by
clicking and dragging the handles in the 3D
Viewer.
rotate xyz rotate 0,0,0 Rotates the scene around the x, y, and z axes.
You can adjust rotate values by holding down
Ctrl/Cmd and dragging in the 3D Viewer.
scale xyz scale 0,0,0 Scales the scene around the x, y, and z axes.
You can adjust scale values by holding down
Ctrl/Cmd+Shift and dragging in the 3D
Viewer.
Note: This control is only displayed

if you've scaled the scene on
individual axes.
uniform scale uniform_scale 1 Scales the scene simultaneously on the x, y,

and z axes.
REFERENCE GUIDE
716
CameraTracker |

Value
Local matrix
specify matrix useMatrix disabled When enabled, specify matrix values for the
object you’re transforming as an alternative
to setting transform and scale values above.
[matrix] local_matrix N/A The matrix displays values from the object’s
transform, rotate, scale, skew, and pivot
controls.
Check specify matrix and copy or drag-and-

drop matrix values from another object to
apply those values, for example, if you
wanted to align objects in a scene.
World matrix
[matrix] world_matrix N/A Displays the world or absolute xyz transform

of the node in world coordinates.
Note: You can’t adjust the World

matrix manually.
Reset Scene resetSceneTransform N/A Click to reset any transforms applied to the
Transform scene to the default value.
Output Tab
Camera
Translate xyz camTranslate 0,0,0 Sets the translate values for the virtual
camera’s position.
Rotate xyz camRotate 0,0,0 Sets the rotation values for the virtual
camera’s position.
Focal Length focalLength 0 Sets the focal length for the virtual camera.
Aperture xy aperture 0,0 Sets the aperture angle for the virtual camera.
REFERENCE GUIDE
717
CameraTracker |

Value
Window windowTranslate 0,0 Sets the center point offset for the camera
Translate xy projection.
Window Scale windowScale 0,0 Sets the relative pixel scaling value for the
xy camera projection.
Lens
Lens lensDistortionType No Lens Specifies the type of lens distortion

Distortion Distortion expected:
• No Lens Distortion - treats the footage as
having no distortion. Use this option if the
Source footage has already been corrected
to remove lens distortion.
• Unknown Lens - calculates the lens
distortion automatically from the sequence
(in the same way as the Image Analysis tab
in the LensDistortion node) and then refines
the distortion in the camera solve.
Lens Type lensType Spherical Sets the lens type that CameraTracker should
expect: Spherical or Anamorphic.
Radial distortion1 0 Sets the first radial distortion term. This is

Distortion 1 proportional to r^2, where r is the distance
from the distortion center.
Radial distortion2 0 Sets the second radial distortion term,

Distortion 2 proportional to r^4.
Distortion distortionCenter 0,0 Sets the values for the center of the radial
Center xy distortion.
Anamorphic anamorphicSqueeze 1 When Lens Type is set to Anamorphic,

Squeeze defines anamorphic squeeze - the distortion
in x is scaled by this amount.
Asymmetry asymmetricDistortion 0,0 When Lens Type is set to Anamorphic,

Distortion xy defines asymmetric distortion to correct for
REFERENCE GUIDE
718
CameraTracker |

Value
slight misalignments between multiple

elements in the lens.
Undistort undistort disabled When enabled, the Source is undistorted

Input based on the calculated lens.
filter filter Cubic When Undistort Input is enabled, sets the

filtering algorithm to used. This allows you to
avoid problems with image quality,
particularly in high contrast areas of the frame
(where highly aliased, or jaggy, edges may
appear if pixels are not filtered and retain
their original values).
original values.
smoothing.
curve).
curve).
smoothing, plus significant sharpening (as
curve).
smoothing (which tends to hide moiré
REFERENCE GUIDE
719
CameraTracker |

Value
patterns).
Card Parameters
scale xyz cardScale 1,1,1 Equivalent value for the scale parameter on
the Card node, used to produce horizontal
scale in conjunction with the Radial
Distortion controls.
a (r^3) a 0 Equivalent value for the a (r^3) control on

LensDistortion tab of the Card node. Defines
a cubic term that affects the radial distortion
of the texture applied to the card, and the
edges of the card.
This allows you to distort an image projected

onto a card in 3D, for example, to simulate
camera lens distortion effects, such as barrel
distortion.
Positive values make the card rounder,

negative make the corners sharper.
b (r^2) b 0 Equivalent value for the b (r^2) control on

a square term that affects the radial distortion
edges of the card.

distortion.

c (r^1) c 0 Equivalent value for the c (r^1) control on

REFERENCE GUIDE
720
Card |

Value
a linear term that affects the radial distortion

edges of the card.

distortion

Card
The Card node creates the simplest type of object you can add to a 3D scene - a plane onto which you can
map a texture. A card object may be distorted using the Lens Distortion tab or deformed as a bilinear or
bicubic object using the Deform tab.
Inputs and Controls

Type Name
Input img The texture you want to project on to the card.

(Scripting) Value
Card Tab
display display textured Adjusts the display characteristics of the 3D object.

These settings don’t affect the render output of the
REFERENCE GUIDE
721
Card |

(Scripting) Value
scene; these are for display purposes only in the 3D

Viewer.
geometry.
• textured+wireframe - displays the wireframe plus
the surface texture.
selectable selectable enabled When enabled, you can make selections as normal in
the Viewer. When disabled, points cannot be selected
or changed.
render render_mode textured Sets how the object will render. This control is
same settings.
cast shadow cast_shadow enabled When enabled, the card object can cast shadows.
receive receive_ enabled When enabled, and a material shader has been added,
shadow shadow the card object can have shadows cast onto it by other
objects.
frame range range_first 1 Sets the start frame for the frame range displayed in the
Viewer when the Timeline range is set to Input. This can
be used to display the offset used in the TimeOffset
node.
range_last 1 Sets the end frame for the frame range displayed in the
node.
orientation orientation XY Set the orientation of the card on the XY, YZ, or ZX axes.
image aspect image_aspect enabled When enabled, the card is the same shape as the input
REFERENCE GUIDE
722
Card |

(Scripting) Value
image. Disable imageaspect to fit the input image on a

square card.
rows/columns rows 8 The number of rows in the selection grid. The maximum
value is 4096, but high values may cause instability.
columns 8 The number of columns in the selection grid. The

maximum value is 4096, but high values may cause
instability.
z z 0 The card is placed this far from the origin and scaled in
line with the lens-in focal distance and lens-in
haperture value.
lens-in focal lens_in_focal 1 This is the focal length of the camera that took the
picture on the card. The card is scaled to the correct
size so that at distance z it fills this field of view.
lens-in lens_in_ 1 This is the horizontal aperture of the camera that took
haperture haperture the picture on the card. The card is scaled to the correct
size so that at distance z it fills this field of view.

the card according to the transformation data in the
shot. You can create and export them using Nuke or
3D tracking software, such as 3D-Equalizer, Maya, or
Boujou.
that you’ve applied to the card as a channel file. This is
a useful method of sharing setups between artists.
snap_menu N/A • Match selection position - the card is snapped to a

• Match selection position, orientation - the card is
REFERENCE GUIDE
723
Card |

(Scripting) Value
• Match selection position, orientation, size - the card

is snapped to a new position, orientation, and size
depending on the points selected.
translate translate 0, 0, 0 Lets you translate the card along the x, y, and z axes. You
can also adjust translate values by clicking and dragging
the card in the 3D Viewer.
rotate rotate 0, 0, 0 Lets you rotate the card around the x, y, and z axes. you
can adjust rotate values by holding down Ctrl/Cmd and
dragging in the 3D Viewer.
scale scaling 1, 1, 1 Lets you scale the card on the x, y, and z axes.
uniform scale uniform_scale 1 Lets you scale the card simultaneously on the x, y, and z
axes.
skew skew 0, 0, 0 Lets you skew the card on the x, y, and z axes.

location.

REFERENCE GUIDE
724
Card |

(Scripting) Value
Local Matrix

matrix values from another object to apply those values,
for example, if you wanted to align objects in a scene.
Lens Distortion Tab
lock to vertical use_vertical_ disabled When enabled, the distortion is limited to the width -
lock vertical distortion is locked.
a (r^3) lens_in_ 0 Defines a cubic term that affects the radial distortion of
distort_a the texture applied to the card, and the edges of the
card.
Positive values make the card rounder, negative make

the corners sharper. The lens distort values can be used
to simulate camera lens distortion effects, such as barrel
distortion.
b (r^2) lens_in_ 0 Defines a square term that affects the radial distortion
distort_b of the texture applied to the card, and the edges of the
card.

distortion.
c (r^1) lens_in_ 0 Defines a linear term that affects the radial distortion of
distort_c the texture applied to the card, and the edges of the
card.
REFERENCE GUIDE
725
Card |

(Scripting) Value

distortion.
distortion lens_in_ 0 Applies a lens distortion effect using an older Nuke

distortion algorithm. Smaller values make the edges of the card
rounder, large values make the corners sharper.
power lens_in_power 4 Controls the strength of the distortion parameter on

the card.
Deform Tab
type type none Select the deform type from the dropdown menu:
• none - the card is not deformed.
• bilinear - the card is deformed using handles and the
bilinear controls.
• bicubic - the card is deformed using handles,
tangents, and the bicubic controls.
reset handles reset_handles N/A Click to reset the handles and tangents to their default
position. A warning displays to confirm the reset.
reset tangents reset_ N/A When break tangents is disabled, click to reset the
tangents tangents to a third of the distance between their handle
points to smooth out the deform curves. A warning
displays to confirm the reset.
Note: This control has no effect when type is

set to bilinear.
copy copy_shape N/A Click to copy the current handle and tangent positions
into a temporary buffer.
paste paste_shape N/A Click to paste the handle and tangent positions from
the temporary buffer onto the card.
set key set_key_shape N/A Click to set a key for this frame on the handles and
REFERENCE GUIDE
726
Card |

(Scripting) Value
tangents.
delete key del_key_shape N/A Click to delete a key from this frame on the handles and
tangents.
delete anim del_anim_ N/A Click to delete the entire animation of all handles and
shape tangents.
Bilinear
x/y points x_points 3 Sets the number of x/y control points in the grid. The
grid is not modified until you click new shape.
y_points 3
new shape new_shape N/A Click to apply the current x/y points values to the grid.
A warning displays to confirm the new shape.
x subdivide x_subdivide N/A Subdivides the control points on the x axis creating new
points equal to x points -1.
y subdivide y_subdivide N/A Subdivides the control points on the y axis creating new
points equal to y points -1.
uv position uv_position 0.3525 Sets the position of the uv subdivide handle.
uv subdivide uv_subdivide N/A Click to subdivide the card at the uv position. If the uv
subdivide handle is on the card edge then only the u or
v is subdivided.
u subdivide u_subdivide N/A Subdivide the card at the u position. If the uv subdivide
handle is on the card edge then there is no subdivision.
v subdivide v_subdivide N/A Subdivide the card at the v position. If the uv subdivide
cp control_ N/A This collection of controls is used to navigate around

points the control points on the card. The index starts at 0 and
each control point is described by column/row
locations and three xyz coordinates.
• The left and right arrows change the column index of
the control point.
• The up and down arrows change the row index of the
REFERENCE GUIDE
727
Card |

(Scripting) Value
control point.
Bicubic
y_points 3
new shape new_shape N/A Click to apply the current x/y points values to the grid.
A warning displays to confirm the new shape.
x subdivide x_subdivide N/A Subdivides the control points on the x axis creating new
points equal to x points -1.
y subdivide y_subdivide N/A Subdivides the control points on the y axis creating new
points equal to y points -1.
uv position uv_position 0.3525 Sets the position of the uv subdivide handle.
uv subdivide uv_subdivide N/A Click to subdivide the card at the uv position. If the uv
subdivide handle is on the card edge then only the u or
v is subdivided.
u subdivide u_subdivide N/A Subdivide the card at the u position. If the uv subdivide
v subdivide v_subdivide N/A Subdivide the card at the v position. If the uv subdivide
uniform uniform_ enabled When enabled, the tangents are shortened to create a
subdivide subdivide more uniform subdivide.
break tangent break_tangent disabled This control is present for the purposes of backward
compatibility and should not be used in new scripts.
mirror tangent mirror_ disabled This control is present for the purposes of backward
tangent compatibility and should not be used in new scripts.
REFERENCE GUIDE
728
CrosstalkGeo |

(Scripting) Value
cp control_ N/A This collection of controls is used to navigate around

points the control points on the card. The index starts at 0 and
each control point is described by two grid locations
and xyz values corresponding to the position, right
tangent, left tangent, up tangent, and down tangent -
tangent values are relative to the position value.
• The left and right arrows change the column index of
the control point.
• The up and down arrows change the row index of the
control point.
Note: Control points along the edge of the

card only display tangent values if they are
valid. For example, the lower left corner
control point only displays the up and right
tangents.
CrosstalkGeo
With the CrosstalkGeo node, you can move the XYZ points on your geometry using the crosstalk curves for
each axis.
Inputs and Controls

Type Name
Input unnamed The 3D geometry object whose vertices you want to adjust.
REFERENCE GUIDE
729
CrosstalkGeo |

(Scripting) Value
CrosstalkGeo Tab
display display unchanged Adjusts the display characteristics of the 3D object.

Viewer.
• off - hide the 3D geometry object.
geometry.
surface texture.
changed.
same settings.
[crossover crossover N/A From the curve list on the left, select the curve you want
curves] to modify.
For example, you’d select z to only modify the vertex z

values.
You can also select y->x, for example, to use the vertex
y value to evaluate the curve and add the result to the
vertex x value. Adjust the curve as necessary. To insert
Reset N/A N/A Resets the selected curves to their initial positions.
REFERENCE GUIDE
730
Cube |
Cube
The Cube node creates an adjustable cube in a 3D scene. You can map a texture on to the cube by
attaching an image to the img input.
Inputs and Controls

Type Name
Input img The texture you want to project on to the object’s faces.

(Scripting) Value
Cube Tab

Viewer.
geometry.
or changed.
REFERENCE GUIDE
731
Cube |

(Scripting) Value
same settings.
cast shadows cast_shadow enabled When enabled, the cube object can cast shadows.
shadows shadow the cube object can have shadows cast onto it by other
objects.
frame range frame_first 1 Sets the start frame for the frame range displayed in the
node.
frame_last 1 Sets the end frame for the frame range displayed in the
node.
rows/columns rows 10 Sets the number of rows on each face of the object. The
maximum value is 512, but high values may cause
instability.
columns 10 Sets the number of columns on each face of the object.

The maximum value is 512, but high values may cause
instability.
separate faces separate_ enabled When enabled, the vertices that describe the separate
faces cube faces do not share UV coordinate information.
Disable this control when you want to guarantee that

the edges of connected faces share vertices.
cube cube -0.5, 0.5 Specifies the dimensions of the cube (from a positive z
axis viewpoint):
• x - controls the left face on the x axis.
• y - controls the bottom face on the y axis.
• n - controls the back face on the z axis.
REFERENCE GUIDE
732
Cube |

(Scripting) Value
• r - controls the right face on the x axis.

• t - controls the top face on the y axis.
• f - controls the front face on the z axis.

the object according to the transformation data in the
Boujou.
that you’ve applied to the object as a channel file. This
snap_menu N/A • Match selection position - the object is snapped to a

• Match selection position, orientation - the object is
object is snapped to a new position, orientation, and
translate translate 0, 0, 0 Lets you translate the object along the x, y, and z axes.
REFERENCE GUIDE
733
Cube |

(Scripting) Value
dragging the object in the 3D Viewer.
rotate rotate 0, 0, 0 Lets you rotate the object around the x, y, and z axes.
scale scaling 1, 1, 1 Lets you scale the object on the x, y, and z axes.
uniform scale uniform_scale 1 Lets you scale the object simultaneously on the x, y, and
z axes.
skew skew 0, 0, 0 Lets you skew the object on the x, y, and z axes.

location.

Local Matrix

REFERENCE GUIDE
734
Cylinder |
Cylinder
The Cylinder node creates an adjustable cylinder in a 3D scene. You can map a texture on to the cylinder
by attaching an image to the img input.
Inputs and Controls

Type Name
Input img The texture you want to project on to the object’s face.

(Scripting) Value
Cylinder Tab

Viewer.
geometry.
or changed.
REFERENCE GUIDE
735
Cylinder |

(Scripting) Value

same settings.
cast shadows cast_shadow enabled When enabled, the cylinder object can cast shadows.
shadows shadow the cylinder object can have shadows cast onto it by
other objects.
node.
node.
rows/columns rows 30 Sets the number of rows on the outer face of the
object. The maximum value is 512, but high values may
cause instability.
columns 30 Sets the number of columns on the outer face of the

cause instability.
radius radius 1 Sets the radius of the cylinder.
height height 2 Sets the height of the cylinder.
u extent u_extent 360 Sets the extent to which the input image wraps around
the cylinder without extrapolating from the image
edges.
For example, a value of 180 wraps the input image

halfway around the cylinder and closes any remainder
by extrapolating from the edges of the input image.
close top close_top disabled When enabled, the top of the cylinder is closed off. The
input image is extrapolated from the edge to the center
REFERENCE GUIDE
736
Cylinder |

(Scripting) Value
of the cylinder.
close bottom close_bottom disabled When enabled, the bottom of the cylinder is closed off.
The input image is extrapolated from the edge to the
center of the cylinder.

Boujou.

REFERENCE GUIDE
737
DepthGenerator |

(Scripting) Value

z axes.

location.

Local Matrix

DepthGenerator
The DepthGenerator node (NukeX and Nuke Studio only) allows you to generate a depth map from your
footage. The node uses information from a tracked camera to create a channel that displays variations in
REFERENCE GUIDE
738
DepthGenerator |
depth.
A depth map is an image that uses the brightness of each pixel to specify the distance between the 3D
scene point and the virtual camera used to capture the scene.
DepthGenerator also allows you to output depth as normals and position passes, and create a Card3D
node positioned in 3D space and displaced according to the depth channel.
Inputs and Controls

Type Name
Input Mask The mask used to ignore areas of the source image. DepthGenerator
expects alpha values of either 0 (for regions to use) or 1 (for regions
to ignore).
Depth can only be calculated where the real world 3D position of

objects doesn’t change. If there are moving objects in your Source
footage, DepthGenerator is likely to struggle to create an accurate
depth map for those regions. To prevent this, you can exclude
moving foreground regions from the depth calculation by
connecting a matte to this input and setting Ignore Mask in the
DepthGenerator properties to the channel that contains the matte.
The Mask input only appears once you’ve connected the other two
inputs.
Camera A tracked Camera that matches the Source footage. If necessary, you
can create one using the CameraTracker node.
Source The source footage from which the Camera track was derived.

Value
DepthGenerator Tab
Ignore Mask ignoreMask None Sets the mask type to exclude areas of the Source
footage during depth calculation. DepthGenerator
expects values of either 0 (for regions to use) or 1 (for
REFERENCE GUIDE
739
DepthGenerator |

Value
regions to ignore).
Depth can only be calculated where the real world 3D

position does not change. If your Source footage
includes moving foreground objects, it’s a good idea
to supply a mask to prevent distortion in moving
regions:
• Source Alpha - use the alpha channel of the Source
clip to define which areas to ignore.
channel of the Source clip to define which areas to
ignore.
• Mask Luminance - use the luminance of the Mask
input to define which areas to ignore.
luminance of the Mask input to define which areas
to ignore.
• Mask Alpha - use the Mask input alpha channel to
• Mask Inverted Alpha - use the inverted Mask input
alpha channel to define which areas to ignore.
Alternatively, you can try using a smaller Frame

Separation to capture moving objects.
Output
Depth Output outputType Depth (1/Z) Sets the type of output produced:
• Depth 1/Z - output 1/Z where Z is the distance
along the camera Z axis. This matches the depth
output of the ScanlineRender node.
• Distance - output the distance along the ray from
the camera center to the 3D surface point.
Create Card createCard N/A Click to create a group containing a displaced Card to
REFERENCE GUIDE
740
DepthGenerator |

Value
visualize depth as a surface in 3D space.
Note: The DepthOutput control must be

set to Depth (1/Z).
Surface Point P_channel none Sets the channel(s) to output depth as a position pass.
The position pass includes the X, Y, and Z coordinates

for each pixel in the image. You can use it with the
Relight node, or with the PositionToPoints node to
visualize depth as a 3D point cloud.
Surface N_channel none Sets the channel(s) to output depth as a normals pass.
Normal Use the NormalDetail control to affect the
smoothness of the normals pass.
The normals pass contains three vectors of

information for each pixel in the image: X direction, Y
direction, and Z direction. In other words, it stores the
direction in which each point in the image is facing.
You can use a normals pass with the Relight node.
Frame Separation
Analyze analyzeSequence N/A Click to automatically calculate the FrameSeparation

Sequence to use in the sequence.
For this to work, the camera in the Camera input must

be defined for all frames within the frame range.
Note: If you analyze the sequence and only

then decide to add an ignore mask, you
should analyze the sequence again after
adding the mask.
Analyze analyzeFrame N/A Click to calculate the Frame Separation to use at the
Frame playhead frame.
REFERENCE GUIDE
741
DepthGenerator |

Value
This gives you more control than Analyze Sequence,

as you can work through the timeline, analyze
particular frames, and if necessary tweak the Frame
Separation value manually.
The camera in the Camera input must be defined for

all frames within the frame range.
Frame frameSeparation 1 Adjusts the offset between the current frame and the
Separation frame against which to calculate depth for your input.
Use a small separation value for fast camera moves

and a larger value for slow camera moves.
Increasing the separation gives a larger baseline and

more accurate depth. However, a high separation
reduces the quality of image matching and gives
incorrect 3D points.
Animate the FrameSeparation to compensate for

variable camera speeds.
The camera in the Camera input must be defined for

all frames within the frame range.
Note: The Analyse Sequence and Analyse

Frame buttons attempt to calculate the best
separation.
Calculated accuracy N/A Displays the depth accuracy calculated when analyzing
Accuracy frame separation. Values closer to 1 are considered
accurate, and values closer to 0 inaccurate.
Use frames with a high accuracy when placing

elements in 3D (for example, by clicking Create Card
or using a PositionToPoints node).
If you’re not getting accurate depth values, try
REFERENCE GUIDE
742
DepthGenerator |

Value
adjusting Frame Separation or using an Ignore Mask.
Depth Generation
Depth Detail vectorDetail 0.5 Adjust this to vary the resolution of the images used
to calculate the depth map. The default value of 0.5
equals half the image resolution. Lower values speed
up processing and deliver a smoother result. Higher
values pick up finer details, but also increase
processing time.
Normal normalDetail 0.25 Adjusts the resolution of the surface normals. Reduce
Detail the detail to smooth out the normals pass.
The default value of 0.25 causes the normals pass to

be calculated at a quarter resolution.
Noise noiseLevel 0.01 Sets the amount of noise to ignore in the input
footage when calculating the depth map.
Try increasing this to smooth depth in flat image

regions.
Strength strength 1 Sets the strength in matching pixels between frames.

Increase this to force the matches, for example where
fine details are missed, or reduce to smooth depth.
In most cases, you can leave this at the default value.
Sharpness sharpness 0.5 Defines how distinct object boundaries should be.
Increase sharpness to separate objects and reduce it to
smooth depth.
Smoothness smoothness 0.5 Adjusts the smoothness weighting of the disparity

calculation. Rather than affecting the depth
calculation, this applies an intelligent blur on the
result.
REFERENCE GUIDE
743
DepthToPoints |

Value
A high smoothness can miss lots of local detail, but is

less likely to produce noisy depth values. A low
smoothness will concentrate on detail matching, even
if the resulting depth map is jagged.
Depth Limits
Near Clip near 0.1 Sets the minimum z depth. All depth values that are
Plane lower than this number are set to this value.
Far Clip Plane far 10000 Sets the maximum z depth. All depth values that are
higher than this number are set to this value.
Mark Bad markRegions disabled When enabled, the regions where the depth
Regions calculation is ambiguous are marked in the depth map
as very large values.
DepthToPoints
DepthToPoints is a gizmo containing the DepthToPosition and PositionToPoints nodes. It can be used to
generate a 3D point cloud from a depth pass and 3D camera.
See also DepthToPosition, PositionToPoints, DepthGenerator, and Camera.
REFERENCE GUIDE
744
DepthToPoints |
Inputs and Controls

Type Name
Input camera The camera required by the DepthToPosition node to create the
position pass from the depth pass.
image The image from which the point cloud is generated, and where all the
depth data resides. This is also where the color information for the
point cloud is taken from.
norm The image containing normals data. Note that this input takes the
place of the channels specified in the surface normal control.

(Scripting) Value
User Tab
display display solid Adjusts the display characteristics of 3D objects. These

geometry.
surface texture.
changed.
render render_mode solid Sets how the points will render. This control is
REFERENCE GUIDE
745
DepthToPosition |

(Scripting) Value
same settings.
cast shadow cast-shadow enabled When enabled, and a material shader is added to the
object, the object can cast shadows.
receive receive_ enabled When enabled, and a material is added to the object, the
shadow shadow object can receive shadows.
frame range range_first 1 Sets the start frame for the frame range displayed in the
node.
range_last 1 Sets the end frame for the frame range displayed in the
node.
depth depth depth.Z Sets the channel that you want to use for the depth.
invert depth invert disabled Invert the depth before processing the depth pass
image. This is best used when the depth channel is z, as
opposed to 1/z.
surface N_channel none The point normals are taken from these channels in the
normal first input. If the norm input is used, these channels are
ignored.
point detail detail 0.5 Controls the number of available points to display in the
Viewer. A value of 0 means no points are displayed. A
value of 1 displays all available points.
point size pointSize 5 Controls the display size of the points in the Viewer.
DepthToPosition
Generates a position pass from a depth pass and a 3D camera.
See also DepthToPoints, PositionToPoints, DepthGenerator, and Camera.
REFERENCE GUIDE
746
Diffuse |
Inputs and Controls

Type Name
Input camera The 3D tracked camera that corresponds to the image input.
image The image from which the position pass is generated, and where all
the depth data resides.

(Scripting) Value
DepthToPosition Tab
output output rgb Specifies the channels the output position pass is put
into, and includes the option to create a new layer for
this purpose.
depth depth depth.Z Sets the channel that you want to use for the depth.
invert depth invert disabled Invert the depth before processing the depth pass
image. This is best used when the depth channel is z, as
opposed to 1/z.
far far 10000 Specifies a value at which depth beyond this distance
from the camera are ignored when calculating the
position pass.
Diffuse
The Diffuse node lets you adjust the color of an illuminated material in a scene. The material appears
darker as the surface points away from the light.
REFERENCE GUIDE
747
Direct |
Inputs and Controls

Type Name
Input map The image used to modulate the diffuse properties of the material.
unnamed Either:

(Scripting) Value
Diffuse Tab

white white 0.18 The color of the material when illuminated. Note that
input of the node and adjust this value, you need to look
at the rendered 2D image to see the effect of your
changes. Changing the white value does not have any
effect in the 3D Viewer.
Direct
Direct lights emit parallel light in one direction and appear to illuminate all objects with equal intensity, as
if it was coming from a far away source - you can use direct light to simulate sunlight and moonlight, for
example. Being at an infinite distance from the objects, direct light has orientation, but no position. To
generate accurate shadows from a Direct light, view the scene through the light (using the Viewer's camera
REFERENCE GUIDE
748
Direct |
dropdown menu, just like a camera) and adjust the Direct light's scale control so that the part of the scene
that should cast shadows fits within the view. This ensures that none of the shadow-casting geometry is
missed by the depth map.
See also Light, Spot, Point, Environment, and Relight.
Inputs and Controls

Type Name
Input unnamed An Axis, a Camera, or another Light node.

(Scripting) Value
DirectLight Tab
color color 1 Sets the color of the light.
intensity intensity 1 Sets the brightness of the light.
display display wireframe Adjusts the display characteristics of the 3D object.

Viewer.
geometry.
• solid+lines - displays the geometry as solid color with
the object’s geometry outlines.
• textured+lines - displays the wireframe plus the
surface texture.
changed.
REFERENCE GUIDE
749
Direct |

(Scripting) Value

channel file. Channel files contain a set of cartesian
Boujou.

REFERENCE GUIDE
750
Direct |

(Scripting) Value
z axes.

location.

Local Matrix

World Matrix
matrix matrix N/A Displays the world or absolute xyz transform of the node

Shadows Tab
cast shadows cast_shadows disabled When enabled, the light casts shadows as defined by the
REFERENCE GUIDE
751
Direct |

(Scripting) Value
Shadow controls.
shadow mode shadow_ solid Presents three shadow casting modes that affect
mode shadows cast by objects, based on the objects’ opacity:
• solid - objects that cast shadows are considered to be
completely solid. This option can be used with both
ScanlineRender and PrmanRender.
• clipped alpha - objects that cast shadows are
considered to be transparent if the object’s alpha is
below the light’s clipping threshold control in the
Shadows tab. This option is only relevant if you are
using ScanlineRender to render your shadows.
• full alpha - shadows are calculated based on how light
is reduced when it passes through non-opaque
occluders. This option is only relevant if you are using
ScanlineRender to render your shadows.
filter filter cubic Determines the type of filter that the shadow mode
uses when it’s set to clipped alpha or full alpha.
scene epsilon scene_epsilon 0.001 An offset that moves the sampling point away from the
geometry surface, towards the light that is casting the
shadow. Increasing this value can reduce self-shadowing
artifacts.
This control is only relevant if your shadows are

generated using depth mapping and shadow mode is
set to full alpha.
samples samples 1 Sets the number of samples for the light when
generating soft shadows. If soft shadows in your scene
appear dotty or noisy, try increasing this value. The
higher the value, the smoother the soft the shadows
become.
This control is only relevant when shadow mode is set

to solid or clipped alpha.
REFERENCE GUIDE
752
Direct |

(Scripting) Value
sample width sample_width 1 Sets the size of the light for soft shadows. This value
determines the width of the soft area around the egde
of a shadow. The higher the value, the larger the soft
area.

generated using raytracing.
bias depthmap_ 0.01 Sets the bias for the raytracing or shadow map. Increase
bias this value if self shadowing artifacts appear in the image.
This moves the surface sample point away from surface.
Note, however, that if you increase the value too much,
some shadows may start moving away from the base of
the objects that cast them.

slope bias depthmap_ 0.01 Bias for the shadow map. This is like bias, but the offset
slope_ is proportional to the slope of the depth map. This
bias allows you to give a different offset to each value in the
depth map, depending on the surface’s slope relative to
the light.
If increasing bias reduced the existing self-shadowing

artifacts but introduced more artifacts in other areas of
the image, you may want to bring bias down a little and
increase slope bias instead.

set to solid or clipped alpha.
clipping clipping_ 0.5 Objects that are set to cast shadows are considered
threshold threshold transparent if their alpha is below the value set here.

set to clipped alpha.
REFERENCE GUIDE
753
DisplaceGeo |

(Scripting) Value
jitter scale shadow_jitter_ 3 Sets the amount of jitter used when doing percentage-
scale closer filtering (PCF) for soft shadows. A larger jitter
scale value results in softer, more perceptually accurate
shadows.
PCF works by sampling the depth map at many different

positions around the same spot. The final shadow value
for that spot is an average of how many of the samples
were occluded or visible from point of view of the light.

depthmap depthmap_ 1024 Sets the resolution of the depth map. Larger values will
resolution width result in a less crunchy edge, but will require more time
to process.
Note that you can also fix crunchy edges by increasing

the number of samples instead of increasing the depth
map resolution.

output mask shadow_mask disabled Enables the associated channel to the right. Disabling
none Sets the channel you want to output the shadow map
into. This can be enabled even if the cast shadows box
is disabled.
DisplaceGeo
REFERENCE GUIDE
754
DisplaceGeo |
With the DisplaceGeo node, you can modify the shape of your 3D geometry object based on an image.
When using the node, each vertex is displaced along its normal with a value corresponding to the image
pixel the vertex’s uv attribute points to. The higher the pixel value, the greater the displacement.
See also Displacement.
Inputs and Controls

Type Name
Input unnamed The 3D geometry object you want to modify. This can be a Card
node, for example. If you want to modify several objects together,
you can also connect a MergeGeo node here.
displace The image you want to use to modify the 3D geometry object.

(Scripting) Value
DisplaceGeo Tab
geometry.
surface texture.
REFERENCE GUIDE
755
DisplaceGeo |

(Scripting) Value
changed.
same settings.
channels channels rgb The displace input’s channels to use for the
displacement value.

channels.
source source luminance The source channel you want DisplaceGeo to get a
displacement map from. For example, if you selected
rgb or rgba from the channels dropdown menu, you can
use the red, green, or blue channel, or the pixel
luminance as the source.
• luminance - create a displacement map from
luminance values.
• red - create a displacement map from red color values.
• green - create a displacement map from green color
values.
• blue - create a displacement map from blue color
values.
• alpha - create a displacement map from the alpha
channel values.
• rgb absolute - move your geometry object’s vertices
to the values in rgb.
• rgb relative - move your geometry object’s vertices
on the x, y and z axes by the amounts in rgb.
scale scale 0.1 Increase or decrease the scale of the displacement. The
higher the value, the bigger the displacement.
weight weight 1, 1, 1 To give x, y, and z different weightings, enter new

weights the weight fields. By default, each weighting is
REFERENCE GUIDE
756
DisplaceGeo |

(Scripting) Value
set to 1. If you don’t want to make changes to a value,

set its weight to 0.
offset offset 0, 0, 0 To offset x, y, and z values from the center of the

geometry object, enter the value by which you want to
offset them in the offset fields. For example, if you
enter 0.5 in the offset y field, 0.5 is added to the y value.
filter size filter_size 5 Change the size of the filtering applied to the image
before the displacement.
filter filter Parzen Select the filtering algorithm to use when remapping
REFERENCE GUIDE
757
Displacement |

(Scripting) Value

Sinc4 the most.
attrib name uv_attrib_ uv Change the name of the attribute that’s used as the
name vertex’s UV coordinates to find the image pixel. Normals
are stored as a named attribute of a piece of geometry.
• Cf - surface color
• N - normal
• PW - world-space point
• uv - texture coordinates
• vel - point velocity.
recalculate recalc_ enabled After the displacement, check this to recalculate the
normals normals normals, which usually aren't correct after the vertices
have been moved.
Displacement
Like the DisplaceGeo node, the Displacement shader node also performs displacement mapping, and at
first glance the nodes seem very similar. However, the approach they have on performing displacement
mapping is different.
REFERENCE GUIDE
758
Displacement |
Displacement mapping is a technique for adding geometric detail to object surfaces as you render them.
Unlike the DisplaceGeo node, the Displacement node does this on the fly, only displacing those parts of
the geometry that are visible at any given moment. Displacement considers the point of view of camera to
determine which parts of the displacement need rendering, thus saving render time. It’s also possible to
optimize the level of tessellation to be the level that you need for an object at a certain distance.
Inputs and Controls

Type Name
Input normal An optional map for calculating the normals (an RGB image where
each pixel corresponds to the X, Y, and Z coordinates of a surface
normal).
displacement The image to create the displacement from.
unnamed Either:

(Scripting) Value
Displacement Tab
displacement displacement_ luminance Choose the channel from the displacement input that
channel channel you want to use as the displacement map:
• red - use the red channel.
• green - use the green channel.
• blue - use the blue channel.
• alpha - use the alpha channel.
• luminance - use the luminance.
• average - use the average of the red, green, and blue
channels.
normal normal_ none If you’re using the normals input, uncheck the build
REFERENCE GUIDE
759
Displacement |

(Scripting) Value
expansion expansion normals box and choose how you want the normals to
be calculated:
• none - to use the normals as they are (XYZ).
• XY - to multiply the normals in the x and y dimensions
(2*XY-1).
• XYZ - to multiply the normals in the x, y, and z
dimensions
(2*XYZ-1).
scale scale 0.1 An overall multiplier that scales the displacement at

each vertex.
filter size filter_size 5 The size of the filter to use when sampling the input
image.
REFERENCE GUIDE
760
Displacement |

(Scripting) Value

Sinc4 the most.
build normals build_normals enabled Check this to automatically calculate the normals after
the displacement.
Uncheck this to have the normals calculated from the

normals input according to the normals expansion
control.
Tessellation Tab
max displacement_ 4 Sets the number of times that the polygon subdivision
subdivision max_ is applied.
tessellation
mode displacement_ screen Select the polygon subdivision mode:

mode • uniform - uniform polygon tessellation.
• screen - evaluates dependent tessellation and ensures
that no new triangles are generated once a certain
screen size is reached.
• adaptive - tessellate according to the displaced
geometry complexity.
pixel edge displacement_ 20 Sets the length, in pixels, of the subdivision polygon
REFERENCE GUIDE
761
Displacement |

(Scripting) Value
length _edge_length edges.
No polygons are generated beyond this threshold.
edge displacement_ 0.01 Sets the edge threshold controlling subdivision. Edges
threshold edge_ larger than this get subdivided automatically, edges
threshold smaller than this get subdivided based on the normal
threshold and the displace threshold. The smaller the
threshold, the higher the amount of tessellation.
Note: The displacement mode must be set to

adaptive to enable this control.
normal displacement_ 0.9 Sets the normal threshold when detecting changes in
threshold normal_ the normal orientation of the displaced samples. A
threshold value of 0.0 indicates normals with the same orientation.
A value of 90.0 indicates orthogonal normals. The
smaller threshold, the higher the amount of tessellation.

displace displacement_ 0.01 Sets the displace threshold when detecting changes in
threshold displace_ the height of the displaced samples. The smaller the
threshold threshold, the higher the amount of tessellation.

REFERENCE GUIDE
762
EditGeo |
EditGeo
The EditGeo node lets you modify geometry, including point cloud meshes, by selecting vertices and
manipulating them in the 3D Viewer. EditGeo also includes a Transform tab to translate, rotate, and scale
3D objects.
Inputs and Controls

Type Name
Input unnamed The 3D geometry object you want to edit. If you want to edit several
objects together, you can also connect a MergeGeo node here.

(Scripting) Value
EditGeo Tab
geometry.
• solid +wireframe - displays the geometry as solid
• textured +wireframe - displays the wireframe plus
REFERENCE GUIDE
763
EditGeo |

(Scripting) Value
changed.
same settings.
axis axis_align object Sets how the current selection's axis is aligned in the
alignment Viewer:
• object - the position of the xyz axis is determined by
the average position of all vertices in the selection. The
orientation of the axis is the same as the object's
orientation.
• average normal - the position of the xyz axis is
determined by the average position of all vertices in
the selection. The orientation of the axis is aligned to
the average of the current selection's normals.
set_key N/A Click to set a keyframe for the position of the currently
selected vertices at the current frame.
remove_key N/A Click to remove a keyframe for the position of the

currently selected vertices at the current frame.
reset reset N/A Click to reset the vertices of the selected geometry to
geometry their original positions.
Transform Tab

the input object according to the transformation data
in the channel file. Channel files contain a set of
cartesian coordinates for every frame of animation in a
given shot. You can create and export them using Nuke
or 3D tracking software, such as 3D-Equalizer, Maya, or
Boujou.
that you’ve applied to the input object as a channel
REFERENCE GUIDE
764
EditGeo |

(Scripting) Value
file. This is a useful method of sharing setups between

artists.

translate translate 0,0,0 Lets you translate the 3D object along the x, y, and z
axes. you can also adjust translate values by clicking and
dragging the axis in the 3D Viewer.
rotate rotate 0,0,0 Lets you rotate the 3D object around the x, y, and z axes.
scale scale 1,1,1 Lets you scale the 3D object on the x, y, and z axes.
uniform scale uniform_scale 1 Lets you scale the 3D object simultaneously on the x, y,
and z axes.
skew skew 0,0,0 Lets you skew the object on the x, y, and z axes.
pivot pivot 0,0,0 When you make changes to the 3D object’s position,
REFERENCE GUIDE
765
Emission |

(Scripting) Value

outside of the object. Subsequent transformations will
then occur relative to the new pivot point location.

Local Matrix
specify matrix useMatrix disabled Enable this control to specify matrix values for the

Emission
The Emission node simulates lamps or other sources that emit light.
REFERENCE GUIDE
766
Environment |
Inputs and Controls

Type Name
Input map The image used to modulate the emission properties of the material.
unnamed Either:

(Scripting) Value
Emission Tab

emission emission 0 The color of the light the material emits. Note that when
you have an image connected to the unnamed input of
the node and adjust this value, you need to look at the
rendered 2D image to see the effect of your changes.
Changing the emission value does not have any effect in
the 3D Viewer.
Environment
Environment lights illuminates objects using an image of light from a real-world environment. This image-
based lighting is generated using High Dynamic Range (HDR) images.
REFERENCE GUIDE
767
Environment |
To use environment light, you first need to shoot a real life environment as an HDR image. Using the
SphericalTransform node, you then convert this image into a spherical mapped image. The sphere is used
to surround the 3D objects, so that the mapped image color illuminates them.
Environment light only works with shiny object materials that can reflect the mapped image. It results in a
very realistic lighting that makes it easier to integrate the objects into the environment.
See also Light, Spot, Point, Direct, and Relight.
Inputs and Controls

Type Name
Input map The transformed HDR image to create the environment light.
unnamed An Axis, a Camera, or another Light node.

(Scripting) Value
Environment Tab
REFERENCE GUIDE
768
Environment |

(Scripting) Value

Sinc4 the most.
mirror image mirror enabled Mirror the HDR image attached to the map input.
blur size blur_size 1 Sets the blur size of the HDR image attached to the map
input.

Viewer.
geometry.
surface texture.
REFERENCE GUIDE
769
Environment |

(Scripting) Value
changed.

Boujou.

REFERENCE GUIDE
770
Environment |

(Scripting) Value
z axes.

location.

Local Matrix

World Matrix

REFERENCE GUIDE
771
FillMat |
FillMat
The FillMat node lets you replace selected material channels with a constant color. Typically, you would
use this node to make one object hold out the others, similar to using a black Constant node as the input
texture. However, the advantage of using the FillMat node is that you can easily apply it to the alpha
channel in addition to the rgb channels and the it doesn’t break the shading sequence, so you can insert it
after other material nodes in your node tree.
Inputs and Controls

Type Name
Input unnamed The 2D image you’re using for the surface texture.

(Scripting) Value
FillMat Tab

color color 0 Adjusts the color used as the constant to replace the
material.
GeoSelect
The GeoSelect node allows you to select individual vertices on a 3D geometry object and save and restore
your selections.
REFERENCE GUIDE
772
GeoSelect |
Inputs and Controls

Type Name
Input unnamed The 3D geometry object you want to select vertices on, for example,
a Sphere or a Cube node. If you want to manipulate several objects
together, you can also connect a MergeGeo node here.

(Scripting) Value
GeoSelect Tab

Viewer.
geometry.
changed.
same settings.
REFERENCE GUIDE
773
Light |

(Scripting) Value
save selection save_selection N/A Saves your selection into the GeoSelect node.
restore restore_ N/A Restores last saved selection.

selection selection
Light
You can use the Light node to add a direct, point, or spot light into your script rather than using specific
predefined light nodes. The light type dropdown defines which light is created and enables or disables
the relevant controls in the properties panel. The node can also import lights from .fbx files using the File
tab.
See also Spot, Point, Direct, Environment, and Relight.
Inputs and Controls

Type Name
Input axis An optional Axis node input. This links the position, rotation, scale,
look An optional input where you can connect a Camera, Light, or Axis that
the Light is automatically rotated to point toward. The Light is
automatically rotated to point towards the connected input
whenever the look input is moved.
REFERENCE GUIDE
774
Light |

(Scripting) Value
Light Tab
read from file read_from_ disabled When disabled, the light is configured using the controls
file on the Light tab. Enable read from file if you want to
read in light information from an .fbx file using the File
tab.
light type light_type point Sets the light type to point, directional, or spot and
disables irrelevant controls.
cone angle cone_angle 40 Sets the spot light cone angle adjusting the spread of
the light. Valid angles are from 0 to 180 degrees.
Note: This control is only enabled for spot

lights.
cone cone_ 0 Controls spot light softness along the edge of the area
penumbra penumbra_ of illumination. A negative value fades inward from the
angle angle circle's edge and vice versa. The cone falloff should be
set to zero or a low amount in order to see the softness.

lights and does not affect the Viewer. Results
are only visible on rendered objects.
cone falloff cone_falloff 0 Sets how much the spot light diminishes from the center
of the circular region out to the edge. The higher the
value, the more focused the light. The falloff is
independent of the falloff type

lights and is not affected by falloff type.
REFERENCE GUIDE
775
Light |

(Scripting) Value
falloff type falloff_type No falloff Sets the amount of light the object gets from the light
source, based on the distance between the light source
and the object.
• No falloff - light does not diminish with distance.
• Linear - diminish the light at a fixed rate as it travels
from the object.
• Quadratic and Cubic - diminish the light at an
exponential rate.
Note: This control is only enabled for point

and spot lights.

Viewer.
geometry.
surface texture.
changed.

REFERENCE GUIDE
776
Light |

(Scripting) Value

Boujou.

z axes.
REFERENCE GUIDE
777
Light |

(Scripting) Value

location.

Local Matrix

World Matrix

Shadows Tab
cast shadows cast_shadows disabled When enabled, the light casts shadow as defined by the
Shadow controls. If using a ScanlineRender node, this
only applies if the light type control was set to
directional or spot. Casting shadows from a point light
REFERENCE GUIDE
778
Light |

(Scripting) Value
type is not supported in the ScanlineRender node.
artifacts.

set to full alpha.
become.

REFERENCE GUIDE
779
Light |

(Scripting) Value
area.


slope_bias is proportional to the slope of the depth map. This
allows you to give a different offset to each value in the
the light.



REFERENCE GUIDE
780
Light |

(Scripting) Value
shadows.


to process.

map resolution.

is disabled.
File Tab (Controls on this tab are only active when read from file is enabled.)
read from file read_from_ disabled When disabled, the light is configured using the controls
file on the Light tab. Enable read from file if you want to
read in light information from an .fbx file using the File
tab.
REFERENCE GUIDE
781
Light |

(Scripting) Value
file file N/A Sets the file path for the .fbx file from which you intend
to import axis information.
reload reload N/A Click to reload the axis information from the specified
file.
animation fbx_take_ N/A When the file control is pointing to a valid .fbx file,
stack name name select the required take name from the dropdown
menu.
node name fbx_node_ N/A When the file control is pointing to a valid .fbx file,
name select the required node name from the dropdown
menu.
input file.
intensity scale intensity_scale 1 Scales the intensity of the light.
Look Tab
look axis look_axis -Z The axis around which the Light is rotated to face the
look input.
look strength look_strength 1 Determines to what extent the Light rotates to the new
orientation. The smaller the value, the less the object is
REFERENCE GUIDE
782
LogGeo |

(Scripting) Value
rotated. Setting the value to 0 produces no rotation.
LogGeo
Move the XYZ of the points by raising the values to a power. This can be useful if you need to visualize rgb
values as a point cloud by treating the rgb values as xyz positions.
Inputs and Controls

Type Name
Input unnamed The 3D geometry object you want to modify. This can be a Card
node, for example. If you want to modify several objects together,
you can also connect a MergeGeo node here.

(Scripting) Value
LogGeo Tab

Viewer.
REFERENCE GUIDE
783
LogGeo |

(Scripting) Value
geometry.
changed.
same settings.
log log 10, 10, 10 Enter the power you want to raise the respective vertex
values to. For example, if you want to raise the vertex z
values to the power of 20, enter 20 in the z field.
swap swap disabled When swap is disabled, the vertex values are calculated
as follows:
vx = vxlogx
vy = vylogy
vz = vzlogz
In the above, vx refers to the x coordinate of vertex v and

logx refers to the x value entered into the log field.
When swap is checked, the values are calculated like
this:
vx = logxvx - 1.0
REFERENCE GUIDE
784
LookupGeo |

(Scripting) Value
vy = logyvy - 1.0
vz = logzvz - 1.0
clamp black clamp_black disabled To clamp the negative x, y, and z values to 0.0, check
clamp black. This prevents your calculated values from
continuing to infinity. This option is only valid if you
have checked swap.
LookupGeo
The LookupGeo node lets you use lookup curves to modify the shape of your 3D geometry objects. It
offers you direct global control over each of the vertex x, y, and z values respectively. You can, for
example, only modify all the y values without touching the x and z values.
You change the different vertex values (x, y, or z) by modifying their associated 2D curves in lookup tables
(LUTs). The x axis in the LUT represents the current vertex value, and the y axis the new vertex value. By
default, the curve is a diagonal line where all the points in the curve have the same value on the y axis (the
new value) as they do on the x axis (the current value). Because both x and y values are the same, there is
no change in the object’s shape.
Inputs and Controls

Type Name
Input unnamed The 3D geometry object whose shape you want to modify. This can
REFERENCE GUIDE
785
LookupGeo |

Type Name
be a Card node, for example. If you want to modify several objects


(Scripting) Value
LookupGeo Tab

Viewer.
geometry.
changed.
same settings.
LUT curves lut N/A Selects the curve you want to modify. Choose X, Y or Z.
For example, you’d select Z to only modify the vertex z
values.
Adjust the curve as necessary by dragging. To insert

reset N/A N/A Resets the selected curve(s) to their initial values.
REFERENCE GUIDE
786
MergeGeo |
MergeGeo
Merges your 3D geometry objects together into one big geometry so you can process all of them at the
same time.
Inputs and Controls

Type Name
Input numbered Any 3D geometry objects you want to merge together.

inputs

(Scripting) Value
MergeGeo Tab
display display unchanged Adjusts the display characteristics of the 3D objects.

Viewer.
geometry.
REFERENCE GUIDE
787
MergeMat |

(Scripting) Value

changed.
render render_mode unchanged Sets how the merged objects will render. This control is
same settings.
shadow shadow_ unchanged Determines whether merged geometry nodes remain

override unchanged, or are adjusted to override inputs. If the
latter option is chosen, you can set shadow options for
the newly merged nodes.
cast shadow cast_shadow disabled If the shadow operation is set to override inputs, cast
shadow enables the geometry to cast shadows on other
objects.
receive receive_ disabled If the shadows operation is set to override inputs,

shadow shadow receive shadow enables the geometry to have shadows
cast on it by other objects, provided a materials shader
has been attached to the object.
MergeMat
Combine two shader nodes together, using compositing algorithms like none, replace, over, and stencil.
The MergeMat node is particularly useful for combining multiple Project3D nodes, allowing you to
composite 2D images projected onto the 3D geometry atop each other.
Inputs and Controls

Type Name
Input A The the foreground material element to merge.
B The the background material element to merge.
REFERENCE GUIDE
788
ModelBuilder |

(Scripting) Value
MergeMat Tab

operation operation over Selects how you want to composite the results of the
two shader nodes together:
• none - only use input B in the composite.
• replace - only use input A in the composite.
• over - composite input A over input B using a mask.
• stencil - use input B outside the mask area.
• mask - use input B inside the mask area.
• plus - add input B to input A.
• max - use input A if it is greater than input B, otherwise
use input B.
• min - use input A if it is less than input B, otherwise use
input B.
ModelBuilder
The ModelBuilder node (NukeX and Nuke Studio only) provides an easy way to create 3D models for 2D
shots. You can build a model by creating shapes and then editing them, and align models over your 2D
footage by dragging vertices to their corresponding 2D location.
To be able to align models, ModelBuilder needs a tracked camera and an input image for visual reference.
You can also use other 3D geometry and point clouds as a reference if you already have these for your
scene.
If you don’t connect a camera, image sequence, or reference geometry, you can still create and edit 3D
models, you just can’t do anything in Align mode.
Creating and editing models using ModelBuilder requires a NukeX license, but the resulting geometry can
also be used in Nuke.
REFERENCE GUIDE
789
ModelBuilder |
Inputs and Controls

Type Name
Input geo 3D geometry to use as a reference when positioning new shapes.
For example, you can use the PointCloudGenerator node to create a

3D point cloud for your shot, and connect that to this input. When
creating new shapes, you can then select vertices on the point cloud
to automatically align the shapes with the point cloud. This gives you
an approximate initial position for your shapes.
cam The camera track of the scene in which you intend to use the
ModelBuilder.
src The source footage from which the camera information was
obtained.
tex An optional 2D texture that you can display as a background in the

UV preview window and on the model in the 3D Viewer.
ModelBuilder Tab
display display wireframe Adjusts the display characteristics of the 3D

shapes. These settings don’t affect the render
output of the scene; these are for display
purposes only in the 3D Viewer.
• off - hides the 3D shape.
• wireframe - displays only the outlines of the
shape’s geometry.
• solid - displays all geometry with a solid
color.
• solid+wireframe - displays the geometry as
solid color with the shape’s geometry
outlines.
REFERENCE GUIDE
790
ModelBuilder |
• textured+wireframe - displays the

wireframe plus the surface texture.
selectable selectable enabled When enabled, you can make selections as

normal in the Viewer. When disabled, points
cannot be selected or changed.
independent from the display selection, but
has the same settings.
cast shadow cast_shadow enabled When enabled, the shape(s) can cast shadows.
receive receive_shadow enabled When enabled, and the geometry has a

shadow material shader attached to it, the shape(s) can
receive shadows.
Scene N/A N/A Shows the hierarchy of shapes and groups of

shapes in the scene.
To rename shapes or groups, click on them

and enter a new name.
To toggle the visibility of shapes or groups,
click . Note that when an item is hidden,

it doesn’t appear in the Viewer or renders.
+ sceneAdd N/A Adds a new group to the scene.
You can use groups to organize and set the

visibility for shapes in the scene.
- sceneDelete N/A Deletes the selected shapes and groups from

the scene.
REFERENCE GUIDE
791
ModelBuilder |
Duplicate sceneCopy N/A Duplicates the selected shapes and groups in

the scene.
Distance shapeDistance 10 Sets the default distance at which shapes are

created. New shapes are positioned at this
distance from the camera.
There are no specific units for the Distance

value. Low values position the shape close to
the camera, and higher values mean further
away.
Textures
Projected textureType current frame Sets the texture method to use:

• current frame - Project the current frame
onto the shape.
• locked frame - Project the frame specified
in the field on the right onto the shape. This
can help you line up the shape against your
source footage.
• frame difference - Subtract the frame
specified in the field on the right from the
current frame and project the resulting frame
onto the shape. This can help you line up the
shape against your source footage.
Note: This control only has an effect

when display is set to textured or
textured+wireframe.
textureFrame N/A When Projected is set to locked frame,

ModelBuilder always projects the texture from
this frame onto the model.
When Projected is set to frame difference,

ModelBuilder subtracts this value from the
current frame and projects the resulting frame
REFERENCE GUIDE
792
ModelBuilder |
on to the model.
Preview previewTexture tex input Sets whether to display the image connected
to the tex input in the UV preview window
and on the 3D model:
• off - Do not display the image in the UV
preview or on the 3D model.
• tex input - Display the image as a
background in the UV preview and set it as
the texture for the currently selected object
in the 3D Viewer. This allows you to line up
your UVs over features in the tex image and
see how well the UVs work on the actual
model.
Export
[bake menu] bakeMenu Selected geometry Sets what ModelBuilder creates when you click
the Bake button:
• Selected geometry - ModelBuilder creates a
geometry node for the selected items in the
scene. You can use this to bake out parts of
the scene to set up projections by baking
Projection.
• Projection - ModelBuilder creates a
projection at the current texture frame. This
creates a Project3D node with a FrameHold
node set to lock the input image and camera
to the texture frame. You can then use the
ApplyMaterial node to apply the projection
to geometry created by baking Selected
geometry.
REFERENCE GUIDE
793
ModelBuilder |
Note: The options in this menu are

implemented in Python, and you can
also use Python to add your own
entries to the menu. For more
information, see Nuke's online help.
Bake bake N/A Bakes out a feature of your model. You can
choose the feature using the bake menu.
This allows you to separate out parts of the

scene for easier processing.
Show srcDisplay enabled When enabled, the src image is displayed in

Source the 3D Viewer whenever the Viewer is locked
Image to the input camera.
When disabled, the src image is not displayed

in the 3D Viewer.
Pass passThroughGeo enabled When enabled, the geometry from the geo
Through input appears both in the Viewer and renders.
Geo This can be useful when you want to use the
geo input as a reference for building your
model. For example, when creating new
shapes, you can select points or vertices on
the geometry to automatically align the
shapes with the geometry.
When disabled, ModelBuilder doesn’t output

the geometry from the geo input.
Shape Defaults Tab
Point
Create createPoint N/A Creates a new point in the scene. You can use
points to define locators on different parts of
REFERENCE GUIDE
794
ModelBuilder |
the scene.
This is the same as selecting Point from the

shape creation menu in the ModelBuilder
toolbar on the left hand side of the Viewer.
Card
Create createCard N/A Creates a new card in the scene.
This is the same as selecting Card from the

Rows newCardRows 4 Sets the default number of rows for new cards.
Columns newCardCols 4 Sets the default number of columns for new

cards.
Cube
Create createCube N/A Creates a new cube in the scene.
This is the same as selecting Cube from the

Rows newCubeRows 1 Sets the default number of rows for new

cubes.
Columns newCubeCols 1 Sets the default number of columns for new

cubes.
Slices newCubeSlices 1 Sets the default number of slices for new

cubes.
Sphere
Create createSphere N/A Creates a new sphere in the scene.
This is the same as selecting Sphere from the

REFERENCE GUIDE
795
ModifyRIB |
Rows newSphereRows 20 Sets the default number of rows for new

spheres.
Columns newSphereCols 20 Sets the default number of columns for new

spheres.
Cone
Create createCone N/A Creates a new cone in the scene.
This is the same as selecting Cone from the

Slices newConeSlices 20 Sets the default number of slices for new

cones.
Cylinder
Create createCylinder N/A Creates a new cylinder in the scene.
This is the same as selecting Cylinder from the

Rows newCylinderRows 2 Sets the default number of rows for new

cylinders.
Columns newCylinderCols 20 Sets the default number of columns for new

cylinders.
ModifyRIB
The ModifyRIB node (NukeX only) lets you insert RIB (RenderMan Interface Bytestream) statements into
your script to modify a RenderMan bytestream before it’s passed to the PrmanRender node.
REFERENCE GUIDE
796
ModifyRIB |
Inputs and Controls

Type Name
Input unnamed The object or scene that you want to modify. Depending on the type
of operation you choose, you may place the ModifyRIB node before
or after the Scene node in your script.

(Scripting) Value
ModifyRIB Tab
operation operation object Sets how RIB statements are modified in the RIB stream.
You can choose from the following operations:
• object - RIB statements are inserted before the
statements for the object connected to the input.
• shader - RIB statements replace all of the shader
definitions for the object connected to the input.
• replace - RIB statements completely replace the
object connected to the input. A transform can then
be applied to the RIB statements. Setting the
operation to replace enables the transform knobs.
• world - RIB statements are inserted at the start of the
world in the RIB stream.
archive ribArchive N/A Specifies the RIB archive (from Nuke or another 3D
application) to add to the scene.
use useRibArchive disabled When, enabled the RIB archive is added to the scene,
otherwise the RIB statements are added.
reload reload N/A Click to reloads the specified RIB archive.
statements ribStatements N/A The RIB statements to add to the scene.
REFERENCE GUIDE
797
ModifyRIB |

(Scripting) Value
Note: Malformed RIB statements (unclosed

quotes, for example) can crash PrmanRender
and in turn Nuke. For more information on RIB
statements, please refer to the documentation
provided with RenderMan.

Cartesian coordinates for every frame of animation in
a given shot. You can create and export them using
Nuke or 3D tracking software, such as 3D-Equalizer,
Maya, or Boujou.
artists.
snap_menu N/A • Snapto - snap your 3D object to a vertex selection.

• Scaleto - scale your 3D object to a vertex selection.
• Translateto - translate your 3D object to a vertex
selection.
axes. You can also adjust translate values by clicking and
REFERENCE GUIDE
798
Normals |

(Scripting) Value
and z axes.

Local Matrix

Normals
The Normals node lets you manipulate the normals of your 3D geometry object. Object normals are
vectors that are perpendicular to the surface. They are used in lighting calculations to determine how the
REFERENCE GUIDE
799
Normals |
light should bounce off a surface at any particular point. By manipulating them, you can control the diffuse
and specular light contributions.
Inputs and Controls

Type Name
Input unnamed The 3D geometry object whose normals you want to manipulate. If
you want to manipulate several objects together, you can also
lookat Any Camera, Axis, or light node. If you set action to lookat, the node
points all normals towards this input. (optional)

(Scripting) Value
Normals Tab

Viewer.
• off - doesn't display geometry outlines.
geometry.
changed.
REFERENCE GUIDE
800
Normals |

(Scripting) Value
same settings.
action action unchanged Sets the mode for manipulating normals:

• unchanged - makes no changes to the object’s
normals.
• set - assigns the normals' values to the normal x, y, and
z fields.
• build - rebuilds each normal based on the surrounding
vertices. Adjust the thresholdangle slider to determine
the break angle where two faces no longer constitute a
smooth surface. An angle of 0 means all faces are flat,
whereas 180 means all faces are smooth. A good
average setting is 60.
• lookat - points all normals at the object connected to
the lookat input. The normal control’s values ares
added to the lookat position.
• delete - removes the named attribute from the object.
For example, if you remove the N attribute, the object
has no normals. See the attribname control below.
invert invert disabled Inverts the resulting normals. This works in all action
modes.
normal normal N/A Constant x, y, and z values. These are used in the set and
lookataction modes.
angle threshold 45 If the angle between two faces is greater than this, break
threshold the normal. This only works with vertex normals and is
only available if build is selected in the action
dropdown. Broken normals create sharp edges when lit.
attrib name N_attrib_ N The name of the normal attribute to delete (this only
name works if you set action to delete). Normals are stored as
a named attribute of a piece of geometry. By default the
attribute name is "N". With this control, you can tell the
REFERENCE GUIDE
801
Phong |

(Scripting) Value
node to perform the delete action on normals from

another attribute instead.
• Cf - surface color
• N - normal
• PW - world-space point
• uv - texture coordinates
• vel - point velocity.
Phong
Uses the Phong shading algorithm to provide accurate shading and highlights. The Phong node has several
map inputs you can use to mask the effect of the node.
Inputs and Controls

Type Name
Input mapsh An optional mask for varying the shininess value. Where the mask is
mapS An optional mask for limiting the effect of the specular component.
Any changes you make to specular are limited to the non-black areas
of the mask.
mapE An optional mask for limiting the effect of the emissive component.
Any changes you make to emission are limited to the non-black areas
of the mask.
mapD An optional mask for limiting the effect of the diffuse component.
REFERENCE GUIDE
802
Phong |

Type Name
Any changes you make to diffuse are limited to the non-black areas
of the mask.
unnamed Either:

(Scripting) Value
Phong Tab

channels.
color color 1 Adjusts the material color.
emission emission 0 The color of the light the material emits. Note that when
you have an image connected to the unnamed input of
the node and adjust this value, you need to look at the
rendered 2D image to see the effect of your changes.
Changing the emission value does not have any effect in
the 3D Viewer.
diffuse diffuse 0.18 The color of the material when illuminated. Note that
input of the node and adjust this value, you need to look
at the rendered 2D image to see the effect of your
changes. Changing the white value does not have any
effect in the 3D Viewer.
specular specular 0.8 Adjusts how bright the highlights on the material seem.
min shininess min_shininess 10 The shininess value controls the width of the highlights.
The higher the value, the wider the highlights.
REFERENCE GUIDE
803
Phong |

(Scripting) Value


max shininess max_shininess 10 The shininess value controls the width of the highlights.


shininess shininess_ luminance Select which channel from the mapSh input is used to
• red to use the red channel for the mapping,
• green to use the green channel,
• blue to use the blue channel,
• alpha to use the alpha channel,
• luminance to use the luminance, or
• averagergb to use the average of the red, green, and
blue channels.
REFERENCE GUIDE
804
Point |
Point
A Point light is a point in 3D space that emits light in every direction. A real world example of a point light
is a light bulb. You can use Point lights to simulate light bulbs, lamps, and candles, for example.
See also Light, Spot, Direct, Environment, and Relight.
Inputs and Controls

Type Name

(Scripting) Value
Point Tab
and the object.
from the object.
exponential rate.

Viewer.
REFERENCE GUIDE
805
Point |

(Scripting) Value

geometry.
surface texture.
changed.

Boujou.

REFERENCE GUIDE
806
Point |

(Scripting) Value

z axes.

location.

Local Matrix
REFERENCE GUIDE
807
Point |

(Scripting) Value

World Matrix

Projection Tab (Controls on this tab do not affect Viewer output. Projection only applies to rendered
objects.)
projection projection_ perspective Sets the projection type for the Point light on rendered
mode objects:
• perspective - objects in front of the light have the
illusion of depth defined by the focal length and
h/vaperture.
• orthographic - objects are viewed using a parallel
projection.
• uv - every object renders its UV space into the output
format. Use this to cook out texture maps.
• spherical - the entire 360° world is rendered as a
spherical map.
focal length focal 50 Adjusts the light’s level of magnification.
horiz aperture haperture 24.576 Sets the horizontal aperture of the light.
vert aperture vaperture 18.672 Sets the vertical aperture of the light.
near near 0.1 Adjusts the position of the light’s forward clipping
plane. Objects in front of this plane are not rendered or
displayed.
far far 10000 Adjusts the position of the light’s rearward clipping
REFERENCE GUIDE
808
Point |

(Scripting) Value
plane. Objects behind this plane are not rendered or

displayed.
window win_translate 0, 0 Translates the light’s output along the uv axes.

translate
window scale win_scale 1, 1 Scales the light’s output along the uv axes.
window roll winroll 0 Rotates the light’s output around the z axis.
focal distance focal_point 2 Controls the distance from the lens where the light
focuses.
fstop fstop 16 Sets the f-stop value (relative aperture) of the light.
Shadows Tab
Shadow controls. However, casting shadows from a
Point light type is not currently supported in the
ScanlineRender node.
REFERENCE GUIDE
809
Point |

(Scripting) Value
artifacts.

set to full alpha.
become.

area.


REFERENCE GUIDE
810
Point |

(Scripting) Value

the light.



shadows.


to process.

map resolution.
REFERENCE GUIDE
811
PointCloudGenerator |

(Scripting) Value

is disabled.
PointCloudGenerator
Using the PointCloudGenerator node (NukeX and Nuke Studio only) you can create a dense point cloud
based on the information generated by CameraTracker and use the dense point cloud further to create 3D
meshes of your 2D footage. Dense point clouds are a useful starting point for 3D modeling and
positioning objects in a scene.
Note: The analysis must be performed in NukeX, but the result can be used in Nuke.
Inputs and Controls

Type Name
Input Camera The camera track of the scene from which you intend to generate the
dense point cloud.
Mask An optional image to use as a mask. By default, the point cloud is

REFERENCE GUIDE
812

Type Name
or set to none.
Source The source footage from which the camera information was
obtained.

Value
PointCloudGenerator Tab
display display solid Adjusts the display characteristics of points. These

• off - hides the points.
points.
• solid - displays all points with a solid color.
• solid+wireframe - displays points as solid color
and includes their vertices.
the Viewer. When disabled, points cannot be
selected or changed.
same settings.
cast shadow cast_shadow enabled When enabled, the object can cast shadows.
receive receive_shadow enabled When enabled, and a material shader is added to the
shadow object, the object can receive shadows.
REFERENCE GUIDE
813

Value
Principal principalView left Sets the principal view where tracks are analyzed and
View the Mask input is applied. The secondary camera is
calculated relative to this Principal View camera.

Ignore Mask ignore None Set the mask type to exclude areas of the sequence
during generation:
• Source Alpha - use the alpha channel of the source
clip to define which areas to ignore.
channel of the source clip to define which areas to
ignore.
• Mask Luminance - use the luminance of the Mask
input to define which areas to ignore.
luminance of the Mask input to define which areas
to ignore.
• Mask Alpha - use the Mask input alpha channel to
• Mask Inverted Alpha - use the inverted Mask
input alpha channel to define which areas to ignore.
Set Keyframes
Analyze analyzeSequence N/A Click to automatically analyse the sequence and set
Sequence the required keyframes to use when generating the
point cloud.
REFERENCE GUIDE
814

Value
Note: Keyframes are selected to provide a

large enough camera baseline to
triangulate points, but you can add and
delete keyframes manually, if necessary.
Keyframes keyframes 0 Displays the frames used to track points in the input
sequence. Both automatic and manual keyframes are
shown.
Add addKeyframe N/A Click to set a keyframe at the current frame.
Delete deleteKeyframe N/A Click to delete the keyframe at the current frame.
Frame frameSpacing 15 Sets the interval between keyframes when you click
Spacing Add All. For example, the default value creates a
keyframe every 15th frame.
Add All addAllKeyframes N/A Click to set keyframes throughout the sequence at
intervals specified by the FrameSpacing control. For
example, for a sequence with 30 frames, the default
spacing value creates 3 keyframes.
Delete All deleteKeyframes N/A Click to delete all keyframes in the sequence.
Calculated accuracy 0 Displays the depth accuracy calculated at the current

Accuracy frame when using the automatic SetKeyframes
method.
The accuracy rating can help you to select frame

ranges when generating a point cloud - higher
accuracy frames produce cleaner point clouds.
Dense Tracking
Point pointSeparation 2 Sets the separation value, in pixels, for points in the
Separation dense point cloud. Reduce the separation to create
denser point clouds and vice-versa.
Track matchThreshold 0.5 Sets the correlation threshold at which point tracks
Threshold are terminated. Use a higher threshold to increase
REFERENCE GUIDE
815

Value
the accuracy in matching points between keyframes.
Track Points track N/A Click to display a frame range dialog to determine
which frames are used to calculate the point cloud.
Use the standard Nuke notation, for example, 1-19,
40-60, 65-100.
Points can also be added to the point cloud in stages

using different frame range selections. Use
CalculatedAccuracy to determine the most accurate
frames for generation.
Clear Points clearPoints N/A Click to clear all tracked points.
Post Filtering
Display displayRejectedPoints enabled When enabled, points that fall outside the Angle and
rejected DensityThresholds are highlighted in the 3D Viewer.
points
When disabled, all rejected points are removed from
the Viewer.
Angle minTriAngle 1 Sets the threshold for the minimum angle to

Threshold triangulate 3D points (in degrees). Points with a large
triangulation angle are more accurate. Set a
threshold of 0 to triangulate all points. Increase the
threshold to retain the more accurate points.
As a rule of thumb, anything below 5 degrees is

likely to be incorrect.
Density densityThreshold 0 Sets the threshold for the minimum density for
Threshold points. Set a threshold of 0 to output all points and
increase the threshold to highlight isolated points.
Delete deleteRejected N/A Click to permanently delete points that are rejected
Rejected by the Angle and DensityThresholds.
Points
REFERENCE GUIDE
816

Value
Tip: To clean up the point cloud, try

enabling Output points per frame and
Display rejected points. You can then tune
the Filtering thresholds and delete points
at different frames.
Output
Point Size pointSize 2 Sets the size, in pixels, of points displayed in the 3D
Viewer.
Output outputPerFrame disabled When enabled, only the points generated at the
points per current frame are displayed in the Viewer.
frame
When disabled, points from all tracked frames are
displayed.
Groups Tab
Display displayGroups enabled When enabled, group colors are displayed as an

groups in overlay in the 3D Viewer.
overlay
Output outputGroups disabled When enabled, only points designated as part of a

visible group on the Groups tab are displayed in the 3D
groups only Viewer.
You can label the point cloud using Groups to help

visualize different parts of the scene.
Create createGroup N/A Click to add the current vertex selections to a new
Group group. You can also use the Viewer right-click menu
to create, delete, and modify groups.
Delete deleteGroup N/A Click to delete all selected groups and deselect the
Selected group in the point cloud.
Group
To delete points use Vertex selection mode in the
3D Viewer and press the delete key.
REFERENCE GUIDE
817

Value
Groups
groups groups None Lists all groups in the point cloud and allows you to
set the name, color, and visibility of each group.
To create groups, enable Vertexselection mode in

the 3D Viewer. You can then select points and use
the right-click menu to create a new group or add
the selected points to an existing group.
Bake bakeGroups N/A Click to bake the selected groups to a separate,

Selected baked point cloud.
Groups
Bake bakeMesh N/A Click to convert the selected groups to a mesh using
Selected Poisson Surface Reconstruction.
Groups to
Mesh Mesh creation is based on the SGP 2006 paper,
Poisson Surface Reconstruction by Michael Kazhdan,
Matthew Bolitho, and Hugues Hoppe. The original
source code and paper can be found at
http://www.cs.jhu.edu/~misha/Code/PoissonRecon/.
Samples samplesPerNode 1 Specifies the minimum number of sample points

used to calculate mesh points, using floating point
values.
For noise-free point clouds, lower values in the range

1.0 - 5.0 can be used. For more noisy samples, higher
values in the range 15.0 - 20.0 may provide a
smoother mesh.

aborts.
REFERENCE GUIDE
818
PoissonMesh |

Value
frame each individual frame. If they throw an exception, the

render aborts.

PoissonMesh
The PoissonMesh node (NukeX and Nuke Studio only) uses information from a dense point cloud to
generate a mesh that you can further use as a 3D object, in 3D modeling for instance. The PoissonMesh
node is based on the Poisson Surface Reconstruction calculation method. The original source code and
paper were created by Michael Kazhdan, Matthew Bolitho, and Hugues Hoppe (for more information, see
http://www.cs.jhu.edu/~misha/Code/PoissonRecon/).
See also PointCloudGenerator and ModelBuilder.
Inputs and Controls

Type Name
Input unnamed The PointCloudGenerator node on which the mesh is based.
REFERENCE GUIDE
819
PoissonMesh |

Value
PoissonMesh Tab
Use Selection useSelection enabled When enabled, selection flags from GeoSelect nodes
further up the tree are used to create the mesh.
Filtering useFiltering enabled When enabled, only faces with normals similar to
those of the input are created.
Depth depth 8 Sets the maximum depth of the calculation tree that is
used for surface reconstruction.
You can also think of it as the density of the voxels

(volumetric pixel) used by the solver. Running at depth
d corresponds to solving on a voxel grid whose
resolution is no larger than 2^d x 2^d x 2^d. Note
that since the reconstructor dynamically adapts the
octree, the specified reconstruction depth is only an
upper bound.
Note: Each increment increases the solver's

potential memory usage by 8 times, so keep
it as low as possible. Due to the scaling of
memory requirements with each increment,
the maximum value is 10.
Scale scale 1.25 Sets the floating point ratio between the 3D cube
used to create the mesh and the bounding box of the
footage it’s created from.
Solver Divide solverDivide 8 Sets the depth at which the solver equation is used in
the creation of your mesh. Using this parameter can
help reduce the memory overhead at the cost of a
small increase in reconstruction time. For example,
with a reconstruction Depth of 9 or higher, a
subdivide depth of 7 or 8 can reduce the memory
usage.
Iso Divide isoDivide 8 Sets the depth at which the iso-surface extractor is
REFERENCE GUIDE
820
PositionToPoints |

Value
used in extraction. This can help you reduce your

memory usage when generating complex meshes. For
example, with a reconstruction Depth of 9 or higher, a
subdivide depth of 7 or 8 can reduce the memory
usage.
Samples per samplesPerNode 1 Sets the minimum number of sample points used to
Node create your mesh.
• For noise-free samples, small values in the range [1.0
- 5.0] are acceptable.
• For more noisy samples, larger values in the range
[15.0 - 20.0] may be needed to provide a smoother,
noise-reduced reconstruction.
Confidence confidence disabled When enabled, the size of the normals is used as
confidence information when generating the mesh.
This may take longer, but can give you better results -
your point cloud point creation is double-checked.
PositionToPoints
Generates a 3D point cloud from an xyz image, such as a position pass. If all the data is in multiple
channels in the input image, then set the surface point to the channels that contain the xyz data in the
input image, and optionally set the surface normal to the channels with the point normals. If you have the
xyz and normals in separate image streams, you can instead connect the xyz stream to the pos input and
optionally connect the normals to the norm input.
REFERENCE GUIDE
821
PositionToPoints |
Inputs and Controls

Type Name
Input unnamed The image from which the point cloud is generated, if all xyz data
resides in one image. This is also where the color information for the
point cloud is taken from.
pos The image containing positional data. Note that this input is hidden
until the unnamed input is connected.
norm The image containing normals data. Note that this input is hidden
until the pos input is connected.

(Scripting) Value
PositionToPoints Tab
display display solid Adjusts the display characteristics of 3D objects. These

geometry.
surface texture.
changed.
REFERENCE GUIDE
822
PrmanRender |

(Scripting) Value
same settings.
receive receive_ enabled When enabled, and a material shader is added to the
shadow shadow object, the object can receive shadows.
node.
node.
surface point P_channel none The point positions are taken from these channels in the
first input. If the pos input is used these channels are
ignored.
surface N_channel none The point normals are taken from these channels in the
normal first input. If the norm input is used these channels are
ignored.
point detail detail 0.5 Controls the number of available points to display in the
Viewer. A value of 0 means no points are displayed. A
value of 1 displays all available points.
point size pointSize 5 Controls the display size of the points in the Viewer.
PrmanRender
PrmanRender (NukeX and Nuke Studio only) is a render node that works together with Pixar’s
PhotoRealistic RenderMan® Pro Server 20, or earlier, to give you an even better quality render result.
PrmanRender is an alternative to Nuke’s ScanlineRender with additional features for rendering 3D scenes.
With PrmanRender, you have control over aspects like shadows and reflections in your render result.
REFERENCE GUIDE
823
PrmanRender |
In order to use the PrmanRender node, you need to have Pixar's RenderMan Pro Server 20, or earlier,
installed and licensed on your machine.
See also ScanlineRender, Reflection, and Refraction.
Inputs and Controls

Type Name
Input cam An optional camera input. The scene is rendered from the
perspective of this camera. If the camera input is not connected,
PrmanRender uses a default camera positioned at the origin and
facing in the negative Z direction.
obj/scn Either:
• A Scene node that is connected to the objects and lights you want
to render, or
• a 3D object or MergeGeo node.
bg An optional background input. This can be used to composite a

background image into the scene and to determine the output
resolution. If not used, this defaults to root.format or root.proxy_
format defined in the Project Settings.

Value
PrmanRender Tab
shadows shadows disabled When enabled, add shadows to your render. You
can adjust the parameters for your shadows using
the Light node’s Shadows tab.
reflections reflections disabled When enabled, add reflections to your render. You
can adjust the parameters for your reflections using
the Reflection node’s Properties panel.
refractions refractions disabled When enabled, add refractions to your render. You
can adjust the parameters for your refractions using
the Refraction node’s Properties panel.
REFERENCE GUIDE
824
PrmanRender |

Value
dof dof disabled When enabled, add depth of field to your render.
projection projection_mode render Sets the projection mode during render:

mode camera • perspective - objects in front of the camera have
the illusion of depth defined by the camera's
focal-length and aperture.
• orthographic - objects are viewed using parallel
projection.
• render camera - take the projection mode from
the camera input.
overscan overscan 0 Sets the number of pixels to render beyond the

left/right and top/bottom of frame, if requested by
subsequent operations.
ambient ambient 0 Sets the global ambient light level.
Sampling Tab
raytrace max raytrace_max_depth 5 Sets the maximum number of specular reflection

depth and refraction bounces PrmanRender uses to trace
your scene and make render calculations.
pixel samples pixel_samples 10 Sets the number of samples to render per pixel.
Having more samples increases your render quality,
but also increases render time.
Note: Setting this value to 1 turns

antialiasing off.
shading rate shading_rate 0.5 Sets the number of shading calculations for
primitives. For example, a shading rate of 1.0
specifies one shading sample per pixel.
This value, along with pixel samples, directly affects

your rendering time and the final quality of your
results. A low shading rate value means your render
will take more time, but the quality will be very
REFERENCE GUIDE
825
PrmanRender |

Value
high. A high value on the other hand means your

render is faster, but the final quality will not be as
good.

values.
Sinc4 the most.
REFERENCE GUIDE
826
PrmanRender |

Value
antialiasing antialiasing_filter Box Sets the antialiasing method used to filter the pixel
filter samples into the final color.
antialiasing antialiasing_filter_ 1 Sets the size of the antialiasing pixel filter to be

filter size size used.
motion blur samples 1 Sets the number of samples to render, per pixel, to
samples produce motion blur. The maximum is 5.
This determines the number of samples of the 3D

scene used so as low a value as possible is best for
setup speed and memory use. If the motion in the
scene over the shutter period is relatively linear,
lower values can be used. If the motion is complex,
then values toward the higher end of the range may
be required.
If the rendered result shows banding or stepping

that looks like too few motion blur samples, first try
increasing the 'pixel samples' setting.
shutter shutter 0.5 Sets the number of frames the shutter stays open
corresponds to half a frame.
options:
frame 30 to 31.
REFERENCE GUIDE
827
PrmanRender |

Value

frame 29 to 30.
shuttercustom shuttercustomoffset 0 If the shutter offset parameter is set to custom,

offset this parameter is used to set the time that the
shutter opens by adding it to the current frame.
randomize temporal_jitter 0 Adds randomness to the distribution of samples in

time time so they don’t produce regularly spaced
images. The larger the value, the larger the time
difference between the samples.
shutter shutter_opening None Sets the shutter opening mode:

opening • None - instantaneous open and close.
• Linear - the shutter opens and closes in linear
intervals.
• Bezier - the shutter opens and closes more
gradually, according to a Bezier curve.
Shader Tab
motion output_motion_ distance Sets the type of vectors you’d like to render:
vectors vectors_type • off - no motion vector information is rendered.
• velocity - store the velocity of every single pixel
in the motion vector channels.
• distance - for every pixel, store the distance (in
pixels) between samples in the motion vector
channels.
REFERENCE GUIDE
828
PrmanRender |

Value
motion vector MB_channels forward When motion vectors is set to anything other than
channels off, the selected vector type is written to this
channel or channels.
output output_shader_ disabled When enabled, shader vectors are output as well as
vectors vectors motion vectors.
surface points P_channel none When output vectors is enabled, the shader vector
surface points are written to this channel or
channels.
surface N_channel none When output vectors is enabled, the shader vector
normal surface normals are written to this channel or
channels.
RIB Tab
filter ribFilter disabled When enabled, filter the scene information

generated by Nuke for RenderMan. Nuke then calls
a Python function called
nukescripts.renderman.filterRIB.
Note: Filtering can make the render start

up slightly slower as a temporary RIB file is
created for each render.
arguments filterArgs none When filter is enabled, specify your filter

arguments. This string is passed by Nuke’s Python
filter function as extra arguments to RenderMan. If
you want to use your own filter, you can also
replace Nuke’s Python function, and have your
arguments passed directly to your own Python
function.
For example, set the filter arguments to -rif

myfilter.so to load your own RI filter.
ouput
REFERENCE GUIDE
829
ProceduralNoise |

Value
file file none Sets the RIB dump file path and name.
Tip: For multiple frames use the ####

pattern.
Execute Execute N/A Click to dump RIB files for a given sequence the
path specified in the file control.
Note: You can execute from the

command line using the -X nodeName
switch.
before render beforeRender none These functions run prior to starting rendering in
execute(). If they throw an exception, the render
aborts.
the render aborts.

ProceduralNoise
REFERENCE GUIDE
830
ProceduralNoise |
The ProceduralNoise (or ProcGeo) node lets you modify your 3D geometry objects using a Perlin noise
function that creates seemingly random noise. For example, you could use this node to generate animated
noise for rippling waves or clouds, or to create a terrain from a flat card.
Inputs and Controls

Type Name
Input unnamed The 3D geometry object you want to modify using a noise function,
for example, a Card node. If you want to modify several objects

(Scripting) Value
ProcGeo Tab

Viewer.
geometry.
or changed.
REFERENCE GUIDE
831
Project3D |

(Scripting) Value
same settings.
ProceduralNoise mode Turbulence Choose the type of noise to use:

Method • fBm (fractional Brownian motion) runs noise()
"octave" times at different scales to get random
numbers between -1 and +1 and adds them up.
• turbulence is the same as fBm except it uses the
absolute value of the noise function so the values it
sums up are all between 0 and 1.
Orientation orient Z Select the values to modify:

• X - only modify the values on the x axis.
• Y - only modify the values on the y axis.
• Z - only modify the values on the z axis.
• All - modify the values on the x, y, and z axes.
X Size x_size 4 The amount of noise on the x axis.
X Offset x_offset 0 Offsets the noise effect on the x axis.
Y Size y_size 4 The amount of noise on the y axis.
Y Offset y_offset 0 Offsets the noise effect on the y axis.
Octaves octaves 5 The amount of detail of the noise.
Lacunarity lacunarity 2.5 The size distribution of the empty space in between
noise instances.
Gain gain 0.5 The amount of noise gain.
Speed speed 1.1 The speed with which the noise changes over time.
Project3D
The Project3D node projects an input image through a camera onto the 3D object.
REFERENCE GUIDE
832
RadialDistort |
See also UVProject.
Inputs and Controls

Type Name
Input cam The Camera controlling the projection.
unnamed The 2D image you’re using for the surface texture.

(Scripting) Value
Project3D Tab
project on project_on both Controls whether the image is projected on the front
facing, back facing, or both polygons.
crop crop enabled When enabled, the input image is extended at its edges
with black. To extend the image with the image edge
colors, disable crop.
occlusion occlusion_ disabled If you want to use raycasting to determine which parts of
mode mode the object are occluded, check self to see the geometry
connected to the shader or check world to see all the
geometries in a scene that causes occlusion.
RadialDistort
The RadialDistort node is a non-linear transformation of the vertices along directions from the object
center, giving either a barrel or pin-cushion distortion.
REFERENCE GUIDE
833
RadialDistort |
Inputs and Controls

Type Name
Input unnamed The 3D geometry object whose shape you want to modify. If you
want to distort several objects together, you can also connect a
MergeGeo node here.

(Scripting) Value
RadialDistort Tab
geometry.
changed.
same settings.
distortion distortion 0 Sets the amount of distortion. Values below 0 produce a
REFERENCE GUIDE
834
RayRender |

(Scripting) Value
barrel distortion, whereas values above 0 produce a pin-

cushion distortion. If you set the value to 0, the 3D
object is not distorted.
power bias power_bias 4 The magnitude of the distortion. The higher the value,
the more distorted the object becomes.
rotation center 0,0,0 Move the center point of the rotation by entering its X,
center Y, and Z coordinates here.
You can also move the center point by dragging its

control in the Viewer.
scale scale 1,1,1 The amount of distortion in the x, y, or z directions.
You can also adjust this by dragging the control in the

Viewer.
preserve preserve_ enabled Check to keep the object centered in its original place in
center center the 3D space. Uncheck to allow the object offset from its
center.
RayRender
When connected to a Scene node, the RayRender node renders all the objects and lights connected to
that scene from the perspective of the Camera connected to the cam input (or a default camera if no cam
input exists). The rendered 2D image is then passed along to the next node in the compositing tree, and
you can use the result as an input to other nodes in the script.
See also PrmanRender, ScanlineRender, Scene, and Camera.
REFERENCE GUIDE
835
RayRender |
Inputs and Controls

Type Name
Input cam An optional camera input. The scene is rendered from the
RayRender uses a default camera positioned at the origin and facing
in the negative Z direction.
obj/scn Either:
to render, or

format defined in the Project Settings.

Value
RayRender Tab

remapping pixels from their original
positions to new positions. This allows
you to avoid problems with image quality,
particularly in high contrast areas of the
frame (where highly aliased, or jaggy,
edges may appear if pixels are not filtered
and retain their original values).
original values.
smoothing.
REFERENCE GUIDE
836
RayRender |

Value

curve).
curve).
smoothing, plus significant sharpening
(as shown by the negative -y portions of
the curve).
• Mitchell - remapped pixels receive
some smoothing, plus blurring to hide
pixelation.
smoothing (which tends to hide moire
patterns).
• Lanczos4, Lanczos6, and Sinc4 -
remapped pixels receive sharpening
which can be useful for scaling down.
Lanczos4 provides the least sharpening
and Sinc4 the most.
projection projection_mode render The projection modes are:

mode camera • perspective - have the camera’s focal
length and aperture define the illusion
of depth for the objects in front of the
camera.
• orthographic - use orthographic
projection (projection onto the
projection plane using parallel rays).
• spherical - have the entire 360-degree
REFERENCE GUIDE
837
RayRender |

Value
world rendered as a spherical map. You

can increase tessellation max to increase
the accuracy of object edges as they are
warped out of line, but this takes longer
to render.
• render camera - use the projection type
of the render camera.
stochastic stochastic_samples 0 Sets the number of samples, per pixel, to

samples use in stochastic estimation (zero is
disabled). Lower values result in faster
renders, while higher values improve the
quality of the final image.
Stochastic sampling is based on Robert L.

Cook’s Stochastic Sampling in Computer
Graphics, available in ACM Transactions on
Graphics, Volume 6, Number 1, January
1996.
Note: It is recommended for

motion blur that the samples
control is adjusted. This also
provides anti-aliasing by jittering
the sample point.
intersection triangle_intersection_ 0.000035 Sets the error threshold for the triangle
epsilon epsilon ray intersection calculations.
MotionBlur Tab
interpolate interpolate_animation disabled When enabled, interpolate between

animation animation keyframes during the shutter
aperture.
When disabled, no interpolation is

calculated.
REFERENCE GUIDE
838
RayRender |

Value
Enabling interpolation can decrease the

number of keyframes and stocastic
samples required to produce motion blur,
but may introduce deviation from the
motion direction.
samples samples 1 Sets the number of keyframes used to

reconstruct motion blur during the
shutter aperture.
uniform uniform_distribution disabled When enabled, use a uniform temporal

distribution distribution of scenes to sample. This
generates more accurate results for
stochastic multisampling.
shutter shutter 0.5 Enter the number of frames the shutter

stays open when motion blurring. For
example, a value of 0.5 corresponds to
half a frame.
shutter offset shutteroffset start This value controls how the shutter
behaves with respect to the current frame
value. It has four options:
• centred - center the shutter around the
current frame. For example, if you set the
shutter value to 1 and your current frame
is 30, the shutter stays open from frame
29,5 to 30,5.
• start - open the shutter at the current
frame. For example, if you set the shutter
value to 1 and your current frame is 30,
the shutter stays open from frame 30 to
31.
• end - close the shutter at the current
frame. For example, if you set the shutter
value to 1 and your current frame is 30,
the shutter stays open from frame 29 to
REFERENCE GUIDE
839
RayRender |

Value
30.
• custom - open the shutter at the time
you specify. In the field next to the
dropdown menu, enter a value (in
frames) you want to add to the current
frame. To open the shutter before the
current frame, enter a negative value. For
example, a value of - 0.5 would open the
shutter half a frame before the current
frame.
shutter custom shuttercustomoffset 0 If the shutter offset parameter is set to

offset custom, this parameter is used to set the
time that the shutter opens by adding it to
the current frame. Values are in frames, so
-0.5 would open the shutter half a frame
match use_scanline_shutter disabled When enabled, assume a sample value of

ScanlineRender 1 and a shutter offset of 0, unless a
shutter offset custom shutter offset is in use.
When disabled, set the required sample

value.
AOVs Tab
output AOV output_shader_vectors disabled When enabled, all the arbitrary output
variables specified are passed into the
specified channels.
remove AOV remove_from_beauty enabled When enabled, the specified AOV are not
from beauty included in the output of the node.
pass
When disabled, all specified AOV channels
are output.
REFERENCE GUIDE
840
RayRender |

Value
surface point AOV_Point none When output vectors is enabled, these

dropdowns allow you to split out the
surface normal AOV_Normal none
various AOV into specific channels for use
motion vector AOV_Motion none later on in the node tree.
solid color AOV_Solid none
direct diffuse AOV_Direct_Diffuse none
direct specular AOV_Direct_Specular none
reflection AOV_Reflection none
emissive AOV_Emissive none
Camera Tab
Stereo stereoScan disabled When enabled, the controls on the

Scan Enable Camera tab are enabled, allowing you to
scan stereo footage.
When disabled, the controls on the

Camera tab are disabled.
Left View leftView N/A Sets the view to use for the left eye in the
output.
Right View rightView N/A Sets the view to use for the right eye in
the output.
Eye Separation eyeSeparation 0.065 Determines how far apart the two views
are, from a viewer's perspective. If you set
the Eye Separation, or interpupillary
distance (IPD), too low, objects in the
scene appear crushed horizontally, but
raising it too high can leave holes in the
stitch.
The IPD is measured in the same units as

the Rig Size control in the upstream C_
REFERENCE GUIDE
841
RayRender |

Value
CameraSolver properties, so adjust it

accordingly.
Convergence convergenceDistance 100 Sets the distance to the zero parallax

Distance point, where the scene is in focus.
Falloff Type falloffType Cosine Determines how pole merging is handled:

• None - no IPD adjustment occurs
towards the poles.
• Linear - the views are merged gradually
from the Start Angle specified toward
the pole. Increasing the angle moves the
start point toward the poles.
• Cosine - the views are merged smoothly
toward the poles. Reducing the
Separation Falloff shifts the transition in
depth towards the poles.
Start Angle separationFalloffStartAngle 0 Sets the point at which falloff begins

when Falloff Type is set to Linear.
Increasing the value pushes the merge

point toward the poles, a value of 90
disables pole merging entirely.
Falloff separationFalloffExponent 1 Sets the rate off falloff for the eye
Exponent separation towards the poles when Falloff
Type is set to Cosine.
A value of 1 produces smooth merging

toward the poles for the left and right
views.
Reducing the value pushes the merge

point toward the poles, a value of 0
disables pole merging entirely.
Sample Ray sampleRayFromCamera disabled When enabled, sample rays with respect
REFERENCE GUIDE
842
ReadGeo |

Value
From Camera to the capture radius for a camera rig.
Enable this control to match stereoscopic

image stitches generated for a horizontal
ring of cameras with a diameter set by the
Rig Size control.
Rig Size rigDiameter 0.1 Sets the diameter of the camera rig used
to generate a corresponding stereoscopic
image stitch, when Sample Ray From
Camera is enabled.
Note: The Rig Size diameter

should always be greater than
the Eye Separation value.
ReadGeo
Imports geometry from a specified location. You can read in:
• Alembic (.abc) files containing meshes or point clouds. The mesh's vertices, normals, UVs, and vertex
colors are read on a per frame basis or at the first frame. Materials and textures are not read in. Any
cameras in the file are passed to Camera nodes.
For more information on Alembic, see http://code.google.com/p/alembic/
• FBX (.fbx) files containing meshes or point clouds. The mesh's vertices, normals, UVs, and vertex colors
are read on a per frame basis or at the first frame. Materials and textures are not read in. Any cameras in
the file are passed to Camera nodes.
• OBJ (.obj) files
• USD (.usd/.usda) files containing meshes and point clouds. The mesh's vertices, normals, UVs, and vertex
colors are read on a per frame basis or at the first frame. Materials and textures are not read in. Cameras
are not currently supported an are grayed-out in the scenegraph.
For more information on USD, see https://graphics.pixar.com/usd/docs/index.html
See also WriteGeo.
REFERENCE GUIDE
843
ReadGeo |
Inputs and Controls

Type Name
Input img The texture to project on to the imported geometry.

Value
ReadGeo Tab
file file none Enter the file path and file name or use the folder
icon to browse to the required location.
Reading .abc and .usd files containing more than one

object displays a scenegraph load dialog that allows
you to create separate ReadGeo nodes from parents
in the scenegraph. See Importing ABC Files or
Importing USD Files for more information.
Localization localizationPolicy from auto- Sets the local file caching behavior. Copies of the
Policy localize path files are stored in a specified local folder for faster
access times:
location.
Update updateLocalization N/A When this ReadGeo node is set to Localization

Policy > on demand, click Update to manually
localize the files from the source files.
REFERENCE GUIDE
844
ReadGeo |

Value
reload reload N/A Click to reload the geometry from disk, overwriting
any changes.
frame range range_first 1 Sets the start frame for the frame range displayed in
the Viewer when the frame slider range is set to
Input. This can be used to display the offset used in
the TimeOffset node.
range_last 1 Sets the end frame for the frame range displayed in
the Viewer when the frame slider range is set to
Input. This can be used to display the offset used in
the TimeOffset node.
display display textured Adjust the display characteristics of the object. These
object’s geometry.
the Viewer. When disabled, points cannot be
selected or changed.
same settings.
receive receive_shadow enabled When enabled, and a material shader has been added
shadow to the geometry, the object can receive shadows.
REFERENCE GUIDE
845
ReadGeo |

Value
Transform Tab

transform the object according to the
contain a set of Cartesian coordinates for every
frame of animation in a given shot. You can create
and export them using Nuke or 3D tracking
software, such as 3D-Equalizer, Maya, or Boujou.
that you’ve applied to the object as a channel file.
This is a useful method of sharing setups between
artists.
snap_menu N/A • Match selection position - the object is snapped

to a new position depending on the points
selected.
• Match selection position, orientation - the object
is snapped to a new position and orientation
object is snapped to a new position, orientation,
and size depending on the points selected.
order translation (T). The possible operation combinations
are SRT, STR, RST, RTS, TSR, TRS.
rotation rot_order ZXY Sets the order of rotation. The possible axial
order combinations are ZXY, XYZ, XZY, YXZ, YZX, ZXY, ZYX.
translate translate 0, 0, 0 Lets you translate the object along the x, y, and z
axes. You can also adjust translate values by clicking
and dragging the object in the 3D Viewer.
you can adjust rotate values by holding down
REFERENCE GUIDE
846
ReadGeo |

Value
uniform scale uniform_scale 1 Lets you scale the object simultaneously on the x, y,
and z axes.

location of the object’s origin point or pivot. The
pivot x, y, and z controls allow you to offset the pivot
point and move it anywhere you like - you can even
move it outside of the object. Subsequent
transformations applied will then occur relative to
the new pivot point location.

Local matrix
specify useMatrix N/A Enable this to specify matrix values for the object
matrix you’re transforming as an alternative to setting
matrix matrix N/A The matrix displays values from the object’s
transform, rotate, scale, skew, and pivot controls.

values, for example, if you wanted to align objects in
a scene.
REFERENCE GUIDE
847
ReadGeo |
ABC

(Scripting) Value
ABC Options - These controls are only shown if you have imported an .abc file.
read on each read_on_ enabled When enabled, ReadGeo bakes each object's transform
frame each_frame into the mesh points and preserves the animation.
This should be enabled for animated objects.
sub frame sub_frame enabled When enabled, objects are read on sub frames which can
be used to render motion blur.
Tip: Disable sub frame for faster UI

interaction.
lock frame lock_frame 1 When read on each frame is disabled, the object is read
at this frame.
frame rate frame_rate 24 The frame rate (frames per second) to use to sample the
animation.
render points point_render_ point clouds Determines how point primitives are rendered:
as mode • point clouds - Render primitives as Nuke point clouds.
• particles - Render primitives as Nuke particles.
use geometry use_ disabled When enabled, apply geometry color attributes read
colors geometry_ from .abc files and apply them to the Nuke geometry.
colors
Note: Disabling use geometry colors can

cause differences in rendered output when
compared to previous versions of Nuke. If this
occurs, enable use geometry colors in the
ReadGeo properties panel.
Scenegraph Tab
scene view scene_view N/A Displays the mesh nodes in the scene. The point cloud
REFERENCE GUIDE
848
ReadGeo |

(Scripting) Value
nodes are not listed.
By default, only meshes imported into this ReadGeo

node are shown. To view all meshes that exist in the
Alembic file, enable view entire scenegraph.
The Load icons on the left indicate the state of each

item in the list:
• A gray dot - The item isn’t loaded into this ReadGeo
node or is hidden.
• An orange bar - The item is loaded into this ReadGeo
node, and is a child of a parent item somewhere
further up the tree. This icon turns gray if you select the
item in the list.
Click the Load icons to show and hide the selected item
and its children.
view entire all_objects disabled When disabled, only meshes imported into this
scenegraph ReadGeo node are shown.
When enabled, all meshes that exist in the Alembic file

are displayed. This allows you to show or hide imported
items by clicking on the icons on the left. Alternatively,
you can choose select or deselect from the right-click
menu.
To select or deselect multiple items, press Ctrl/Cmd or

Shift while clicking them.
FBX

(Scripting) Value
FBX Options - These controls are only shown if you have imported an .fbx file.
REFERENCE GUIDE
849
ReadGeo |

(Scripting) Value
animation fbx_take_ dependent on The take you want to use from the .fbx file. FBX supports
stack name file multiple takes in one file. Usually, one of the takes is a
default take with no animation.
node name fbx_node_ dependent on The mesh node you want to import from the .fbx file.
name file
Point cloud nodes are not listed.
frame rate frame_rate 24 The frame rate (frames per second) to use to sample the
animation curves. To use this rate rather than the one
defined in the FBX file, check use frame rate.
use frame rate use_frame_ disabled When enabled, the frame rate defined above is used to
rate sample the animation curves.
When disabled, the frame rate defined in the .fbx file is

used to sample the animation curves.
object type object_type Mesh Select the object type:

• Mesh - Import a single mesh or all the meshes (or
NURBS curves/patch surfaces converted to meshes)
from the .fbx file.
• Point Cloud - Import a point cloud from the .fbx file.
all objects all_objects disabled When enabled, ReadGeo imports all the meshes in the
.fbx file rather than just one. This overrides whatever
you have selected under node name.
This should be enabled for animated objects.
read read_from_ enabled When enabled, transform properties are imported from
transform file the .fbx file and, while you can view them and use them
from file in expressions, you cannot modify them. Whenever you
reload the script, the transform properties are re-read
from the .fbx file to ensure they match the file content.
When disabled, you can modify the transform properties
REFERENCE GUIDE
850
ReadGeo |

(Scripting) Value
imported from the .fbx file using the transform controls,

and your changes are kept even if the script is reloaded.
Note that geometry data is always read from the FBX

file.
OBJ

(Scripting) Value
OBJ Options - These controls are only shown if you have imported an .obj file.
update mode update_mode all Sets the update mode for the OBJ geometry:
• all - loads all geometry groups for each frame.
• point-cloud - loads only point and point attributes
group data at each frame. The primitives frame
control specifies which frame in the sequence contains
the primitive and vertex attribute group data.
primitives full_ 1 When update mode is set to point-cloud, sets which

frame geometry_ frame in the sequence contains the primitive and vertex
frame attribute group data.
read texture read_texture_ enabled When enabled, w coordinates are read along with uv
w coord w_coord coordinates, where present.
Tip: Occasionally, w values contain garbage,

so you may want to disable this control to use
a value of 1 instead.
REFERENCE GUIDE
851
ReadGeo |
USD

(Scripting) Value
USD Options - These controls are only shown if you have imported a .usd or .usda file.
Note: read on each frame should be enabled

for animated objects.
Scenegraph Tab
scene view scene_graph N/A Displays all items in the file, including currently
unsupported items such as cameras.
The Load column displays the state of each item in the

file:
Note: Grayed-out items are currently

unsupported.
• A gray dot - The item isn’t loaded into this ReadGeo

node or is hidden.
• An orange bar - The item is loaded into this ReadGeo
node, and is a child of a parent item somewhere
further up the tree. This icon turns gray if you select the
item in the list.
• A broken orange bar - The item is partially loaded,
meaning that some children of the item are not
loaded.
You can show or hide imported items by clicking on the

icons in the Load column. Alternatively, you can choose
select or deselect from the right-click menu.
REFERENCE GUIDE
852
Reflection |

(Scripting) Value
The Type column displays at a glance what the file

contains, such as meshes and cameras.
Note: By default, only items imported into

this ReadGeo node are shown. To view all
items that exist in the USD file, enable view
entire scenegraph.
view entire all_objects disabled When disabled, only items imported into this ReadGeo
scenegraph node are shown.
When enabled, all meshes that exist in the USD file are
displayed. This allows you to show or hide imported
items by clicking on the icons on the left. Alternatively,
you can choose select or deselect from the right-click
menu.
To select or deselect multiple items, press Ctrl/Cmd or

Shift while clicking them.
Reflection
Reflection is the familiar physical phenomenon where an image of an object is cast back from a particular
kind of surface, such as glass or water. Using the PositionToPoints node, you can replicate this effect in
your render result of 3D objects, and using the Reflection node, you can adjust the controls for creating
the reflection effect.
Nuke's RayRender node and PrmanRender use raytracing to create this effect and you can use the
Reflection node to adjust the result.
Note: ScanlineRender cannot produce reflections using the Reflection node.
REFERENCE GUIDE
853
Refraction |
Inputs and Controls

Type Name

Output unnamed The 3D geometry object to which you want to apply a reflection
effect. If you want to apply the same effect to several objects, you can
also connect a MergeGeo node here.

(Scripting) Value
Reflection Tab
reflection_ reflection_ 1 Sets the color of the reflection.

color color
value reflection_ 1 Sets the intensity of the reflection.

value
Refraction
Refraction is the familiar physical phenomenon of light traveling differently through different materials
and thus reflecting differently off objects behind that material. For example, if you have a glass of water
with a straw in it, the part of the straw that’s not in water appears to be in a different angle to the part that
is in the water. This is due to water bending the light waves.
PrmanRender uses raytracing to create this effect and you can use the Refraction node to adjust the result.
Without the PrmanRender node and RenderMan Pro Server software though, the Refraction node has no
effect.
REFERENCE GUIDE
854
Relight |
Inputs and Controls

Type Name

Output unnamed The 3D geometry object to which you want to apply a refraction
effect. If you want to apply the same effect to several objects, you can
also connect a MergeGeo node here.

(Scripting) Value
Refraction Tab
refraction refraction_ 1 Changes the type of refraction.

index index
value refraction_ 1 Sets the intensity of the refraction.

value
Relight
The Relight node takes a 2D image containing normal and point position passes and lets you relight it
using 3D lights. Essentially bypassing the need to return to a 3D application and re-render the lighting,
Relight provides a quick and interactive way to relight a 3D scene in a 2D environment.
Relight works by applying a 3D shader to a 2D image using the normal and point position passes stored in
separate image channels, and lets you attach and manipulate a 3D light (or multiple lights via a Scene
node).
See also Light, Spot, Point, Direct, and Environment.
REFERENCE GUIDE
855
Relight |
Inputs and Controls

Type Name
Input color The 2D image to relight using 3D lights. This image should contain
normal and point position passes.
You can create normal and point position passes using the
DepthGenerator node in NukeX, for example.
cam The camera used to render the color image.
Note: The cam input only appears once the lights input
has been connected to a light or scene node.
lights The light node to use to relight the color image. If you want to use
multiple lights, you can also connect a Scene node here.
material The 3D shader to apply to the color image. This can be a Phong
node, for example.
Depending on the type of shader you attach, ensure that you've

defined the necessary properties for it.
Note: The material input only appears once the cam input
has been connected.

(Scripting) Value
Relight Tab
use alpha use_alpha disabled When enabled, the alpha channel of the color image is
used as a mask to limit the effects of Relight on the
original scene.
normal normal none The channel from the color input that contains the
REFERENCE GUIDE
856
ScanlineRender |

(Scripting) Value
vectors normal pass.
point position none The channel from the color input that contains the point
positions position pass.
ambient ambient 0 Sets the global ambient light level for the scene.
ScanlineRender
When connected to a Scene node, the ScanlineRender node renders all the objects and lights connected
to that scene from the perspective of the Camera connected to the cam input (or a default camera if no
cam input exists). The rendered 2D image is then passed along to the next node in the compositing tree,
and you can use the result as an input to other nodes in the script.
The ScanlineRender node also outputs deep data if there is a Deep node downstream.
See also PrmanRender, Scene, and Camera
Inputs and Controls

Type Name
Input obj/scn Either:

to render, or
cam An optional camera input. The scene is rendered from the

ScanlineRender uses a default camera positioned at the origin and
facing in the negative Z direction.
REFERENCE GUIDE
857
ScanlineRender |

Type Name

format defined in the ProjectSettings.
If this input contains a depth channel, ScanlineRender considers it

when doing Z-buffer and Z-blending calculations.

Value
ScanlineRender Tab
transparency transparency enabled When enabled, objects appear transparent where

their alphas are less than 1.
Z-buffer ztest_enabled enabled Enable or disable the Z-buffer which compares

object Z-depth within a scene, assisting with
occlusions.

values.
smoothing.
REFERENCE GUIDE
858
ScanlineRender |

Value

Sinc4 the most.
• Nearest - Fastest and crudest, sample the nearest
texel from the appropriate mip map.
• Bilinear - Remove blockiness, sample and
interpolate the four nearest texels from the
appropriate mipmap level.
• Trilinear - Smooth interpolation of texture quality
according to the distance, bilinearly interpolate
between two closest mipmap levels.
• Anisotropic - Highest quality filtering, gives a
better result when shading surfaces with a high
angle relative to the camera.
antialiasing antialiasing none Sets the level of antialiasing to reduce any aliasing
artifacts in the render. Choose from none, low,
medium and high.
Z-blend zblend_mode none Type of ramp to use to blend two surfaces within
mode the Z-blend range of each other. Smooth looks
better, but linear is provided for back-compatibility.
Z-blend range zblend_range 0.1 Any two surfaces closer together than this distance
on the Z axis are blended together to smooth the
transition between intersecting objects.
REFERENCE GUIDE
859
ScanlineRender |

Value
projection projection_mode render The projection modes are:

mode camera • perspective - have the camera’s focal length and
aperture define the illusion of depth for the
objects in front of the camera.
• orthographic - use orthographic projection
(projection onto the projection plane using
parallel rays).
• uv - have every object render its UV space into
the output format. You can use this option to cook
out texture maps.
• spherical - have the entire 360-degree world
rendered as a spherical map. You can increase
tessellationmax to increase the accuracy of object
edges as they are warped out of line, but this
takes longer to render.
• rendercamera - use the projection type of the
render camera.
tessellation max_tessellation 3 Limits recursive subdivision of polygons by a

max screen-space distance percentage.
This control can be useful in the spherical

projection mode, which sometimes distorts object
edges. If you see such distortions, you can try
increasing this value to tessellate (subdivide)
polygons into smaller polygons. This produces
more accurate object edges, but also takes longer
to render.
overscan overscan 0 The maximum additional pixels to render beyond

the left/right and top/bottom of the frame.
Rendering pixels beyond the edges of the frame
can be useful if subsequent nodes need to have
access outside the frame. For example, a Blur node
down the node tree may produce better results
around the edges of the frame if overscan is used.
REFERENCE GUIDE
860
ScanlineRender |

Value
Similarly, a subsequent LensDistortion node may

require the use of overscan.
ambient ambient 0 Enter a value between 0 (black) and 1 (white) to

change the global ambient color.
MultiSample Tab
samples samples 1 Sets the number of samples to render per pixel, to

produce motion blur and antialiasing. If you use
this, in most cases you can turn off the antialiasing
and filter controls on the ScanlineRender tab.
corresponds to half a frame.
options:
frame 30 to 31.
frame 29 to 30.
REFERENCE GUIDE
861
ScanlineRender |

Value

shuttercustom shuttercustomoffset 0 If the shutter offset parameter is set to custom,

offset this parameter is used to set the time that the
shutter opens by adding it to the current frame.
randomize temporal_jitter 0 Adds randomness to the distribution of samples in

time time so they don’t produce regularly spaced
images. The larger the value, the larger the time
difference between the samples.
sample spacial_jitter 1 The diameter of the circle that the samples for each
diameter pixel are placed in for antialiasing. The larger the
value, the more pixels are jittered.
focus focal_jitter 0 Randomly orbit the camera about a point at the

diameter focal distance in front of it for each sample to
produce depth-of-field effects from multiple
samples.
Note: The focal distance is set in the

Camera node's controls, in the Projection
tab.
REFERENCE GUIDE
862
ScanlineRender |

Value
stochastic stochastic_samples 0 Sets the number of samples, per pixel, to use in

samples stochastic estimation (zero is disabled). Lower
values result in faster renders, while higher values
improve the quality of the final image.
Stochastic sampling is based on Robert L. Cook’s

Stochastic Sampling in Computer Graphics, available
in ACM Transactions on Graphics, Volume 6, Number
1, January 1996.
Note: It is recommended for motion blur

that the samples control is adjusted. This
also provides anti-aliasing by jittering the
sample point.
uniform uniform_ disabled When enabled, use a uniform temporal distribution

distribution distribution of scenes to sample. This generates more accurate
results for stochastic multisampling.
Shader Tab
motion motion_vectors_ distance Select how to render motion vectors:

vectors type • off - do not render motion vectors.
• classic - render motion vectors the classic (pre-
Nuke 6.1) way. This option is only provided for
backwards compatibility, and isn't always accurate.
• velocity - store the velocity of every single pixel
in the motion vector channels (pre-Nuke 7.0 way).
This option is only provided for backwards
compatibility. In order to have the same behavior
as Nuke 6.3, set samples to 1.
• distance - for every pixel, store the distance (in
pixels) between samples in the motion vector
channels. This is the recommended option that
usually produces the best results. It also allows the
VectorBlur node to produce curved vector blur
REFERENCE GUIDE
863
ScanlineRender |

Value
where interpolation between two frames is made

according to a curve rather than linearly.
motion vector MB_channel forward The channels the motion vectors are output to.
channels
output output_shader_ disabled When enabled, shader vectors (surface points and
vectors vectors surface normals) are output as well as motion
vectors. These can be useful if you want to relight
the rendered 3D scene in the compositing phase.
surface points P_channel none The channel to use as the surface point channel.
When output vectors is enabled, ScanlineRender
outputs the surface point positions (in world space
coordinates) into this channel.
surface N_channel none The channel to use as the surface normal channel.
normal When output vectors is enabled, ScanlineRender
outputs the surface point normals (in world space
coordinates) into this channel.
Deep Tab
drop zero drop_zero_alpha_ enabled When enabled, deep samples with an alpha value of
alpha samples samples 0 do not contribute to the output.
When disabled, deep samples with an alpha value 0

contribute to the output.
REFERENCE GUIDE
864
Scene |
Scene
Regardless of its location in your script, the Scene node is the highest-level node in the scene hierarchy
because it references all the elements in a 3D workspace - all the geometric objects, cameras, and lights.
Every Scene node in a script should be connected to a ScanlineRender, RayRender, or PrmanRender node,
which render the results of the scene.
Inputs and Controls

Type Name
Input [numbered 3D objects, Lights, Cameras, and materials. You can connect any
inputs] number of nodes to a Scene node.

(Scripting) Value
Scene Tab
display display unchanged Adjust the display characteristics of the 3D objects

attached to the node. These settings don’t affect the
render output of the scene; these are for display
• wireframe - displays only the outlines of the object
geometry.
color with the object geometry outlines.
• textured - displays only the surface textures.
• textured +wireframe - displays the wireframes plus
the surface textures.
REFERENCE GUIDE
865
Specular |

(Scripting) Value
• unchanged - doesn't change object display modes.
changed.
render render_mode unchanged Sets how objects connected to the scene will render.
This control is independent from the display selection,
but has the same settings.
shadow shadow_ unchanged Determines whether geometry nodes in the scene

override remain unchanged, or are adjusted to override inputs.
If the latter option is chosen, you can set shadow
options for all the geometry in the scene.
cast shadow cast_shadow disabled If the shadow operation is set to override inputs, cast
shadow enables the geometry to cast shadows on other
objects.
receive receive_ disabled If the shadows operation is set to override inputs,

shadow shadow receive shadow enables the geometry to have shadows
cast on it by other objects, provided a materials shader
has been attached to the object.
Specular
The Specular node controls how bright and wide the highlights on a material seem. The location of the
viewpoint is significant: the specular highlights are the brightest along the direct angle of reflection.
REFERENCE GUIDE
866
Specular |
Inputs and Controls

Type Name
Input mapsh An optional mask for varying the shininess value. Where the mask is
map The image used to modulate the specular properties of the material.
unnamed Either:

(Scripting) Value
Specular Tab

white white 0.18 The brightness of the specular highlight. The higher the
value, the shinier the material seems.
min shininess min_shininess 10 The shininess value controls the width of the highlights.

REFERENCE GUIDE
867
Sphere |

(Scripting) Value

max shininess max_shininess 10 The shininess value controls the width of the highlights.


shininess shininess_ luminance Select which channel from the mapSh input is used to
• red to use the red channel for the mapping,
• green to use the green channel,
• blue to use the blue channel,
• alpha to use the alpha channel,
• luminance to use the luminance, or
• average rgb to use the average of the red, green, and
blue channels.
Sphere
REFERENCE GUIDE
868
Sphere |
The Sphere node creates an adjustable sphere in a 3D scene. You can map a texture on to the sphere by
attaching an image to the img input.
Inputs and Controls

Type Name
Input img The texture you want to project on to the object’s face.

(Scripting) Value
Sphere Tab
display display textured Adjust the display characteristics of the 3D object. These
geometry.
or changed.
same settings.
cast shadows cast_shadow enabled When enabled, the sphere object can cast shadows.
shadows shadow the sphere object can have shadows cast onto it by
REFERENCE GUIDE
869
Sphere |

(Scripting) Value
other objects.
node.
node.
rows/columns rows 30 Sets the number of rows on the outer face of the
cause instability.
columns 30 Sets the number of columns on the outer face of the

cause instability.
radius radius 1 Sets the radius of the sphere.
u extent u_extent 360 Sets the extent to which the input image wraps around
the sphere horizontally without extrapolating from the
image edges.
For example, a value of 180 wraps the input image 90°

around the sphere on both sides of the y axis and closes
any remainder by extrapolating from the edges of the
input image.
v extent v_extent 180 Sets the extent to which the input image wraps around
the sphere vertically without extrapolating from the
image edges.
For example, a value of 90 wraps the input image 45°

around the sphere on both sides of the x axis and closes
any remainder by extrapolating from the edges of the
input image.
close top close_top enabled When enabled, the top of the sphere is closed off.
REFERENCE GUIDE
870
Sphere |

(Scripting) Value
close bottom close_bottom enabled When enabled, the bottom of the cylinder is closed off.

Boujou.

REFERENCE GUIDE
871
Spot |

(Scripting) Value
z axes.

location.

Local Matrix
specify matrix useMatrix N/A Enable this to specify matrix values for the object you’re
transforming as an alternative to setting transform,
scale, skew and pivot values above.

Spot
Spot nodes create a point in 3D space that emits a cone-shaped light in a given direction. A real world
example of a spot light is a desk lamp.
See also Light, Point, Direct, Environment, and Relight.
REFERENCE GUIDE
872
Spot |
Inputs and Controls

Type Name

(Scripting) Value
Spot Tab
cone angle cone_angle 40 Sets the spot light cone angle adjusting the spread of
the light. Valid angles are from 0 to 180 degrees.
cone cone_ 0 Controls spot light softness along the edge of the area
penumbra penumbra_ of illumination. A negative value fades inward from the
angle angle circle's edge and vice versa. The cone falloff should be
set to zero or a low amount in order to see the softness.
cone falloff cone_falloff 0 Sets how much the spot light diminishes from the center
of the circular region out to the edge. The higher the
value, the more focused the light. The falloff is
independent of the falloff type
and the object.
from the object.
exponential rate.
display display wireframe Adjust the display characteristics of the 3D object. These
REFERENCE GUIDE
873
Spot |

(Scripting) Value

geometry.
surface texture.
changed.

Boujou.

REFERENCE GUIDE
874
Spot |

(Scripting) Value
z axes.

location.

Local Matrix
REFERENCE GUIDE
875
Spot |

(Scripting) Value

World Matrix

Shadows Tab
Shadow controls.
REFERENCE GUIDE
876
Spot |

(Scripting) Value

artifacts.
This only applies if your shadows are generated using

depth mapping and shadow mode is set to full alpha.
become.
This only applies when shadow mode is set to solid or

clipped alpha.
area.

This only applies when shadow mode is set to solid or

clipped alpha.
the light.
REFERENCE GUIDE
877
Spot |

(Scripting) Value


depth mapping and shadow mode is set to solid or
clipped alpha.
clipping clipping_ 0.5 Objects that are set to cast shadows are considered solid
threshold threshold if their alpha is below the value set here.

depth mapping and shadow mode is set to clipped
alpha.
shadows.


to process.

map resolution.
REFERENCE GUIDE
878
TransformGeo |

(Scripting) Value

is disabled.
TransformGeo
The TransformGeo node lets you translate, rotate, scale, and skew 3D geometry objects.
Inputs and Controls

Type Name
Input unnamed The 3D geometry object you want to translate, rotate, scale, or skew.
If you want to transform several objects together, you can also
look An optional input where you can connect a Camera or 3D object that
the transformed 3D object should face. If a look input exists, the
transformed 3D object is automatically rotated to point towards the
look input whenever the look input is moved.
This can be useful, for example, if you have a 2D matte painting

mapped to a Card in your scene. The look input ensures that the
plane of the painting always faces the camera, regardless of the
REFERENCE GUIDE
879
TransformGeo |

Type Name
camera position, and maintains the illusion depicted by the painting.
axis An optional Axis node input. This links the position, rotation, scale,

(Scripting) Value
TransformGeo Tab
geometry.
changed.
REFERENCE GUIDE
880
TransformGeo |

(Scripting) Value
same settings.
transform transform_ disabled When enabled, transforms normals in the rendered

normals normals output. In most cases this control should be left
disabled.

Cartesian coordinates for every frame of animation in a
Boujou.
artists.

REFERENCE GUIDE
881
TransformGeo |

(Scripting) Value

and z axes.

Local Matrix

Look Tab
look axis look_axis The axis around which the 3D object is rotated to face
the look input.
REFERENCE GUIDE
882
Transmission |

(Scripting) Value
rotate X rotate_x enabled Determines whether the rotation occurs around the X
rotate Y rotate_y enabled Determines whether the rotation occurs around the Y
rotate Z rotate_z enabled Determines whether the rotation occurs around the Z
look strength look_strength 1 Determines to what extent the 3D object rotates to the
new orientation. The smaller the value, the less the
object is rotated. Setting the value to 0 produces no
rotation.
use use_ disabled Uses an alternate scheme to calculate the look rotation.
Transmission
The Transmission node allows you to render semi-transparent, colored shadows by simulating the
absorption of light passing through a surface. The Transmission shader must be used in conjunction with
lights that cast full alpha shadows.
Note: The Transmission node only works if your 3D scene contains a light that casts full alpha
shadows and is rendered using RayRender.
See also RayRender.
REFERENCE GUIDE
883
Transmission |
Inputs and Controls

Type Name
Input map An optional mask for limiting the effect. Any changes you make to
the shadows are limited to the non-black areas of the mask.
unnamed Either:

(Scripting) Value
Transmission Tab
transmittance transmittance 0.18 When absorption mode is set to constant, controls how much
light passes through from the texture image. 0 is equal to no
light passing through and 1 is all light passing through.
intensity intensity 1 Controls the intensity of the light passed through by

modulating the material's absorption.
absorption absorption_ constant Sets the absorption mode used to calculate the shadow:
mode mode • constant - uses the transmittance control to set the amount
of absorption.
• solid - uses the solid component of the input material to set
the amount of absorption.
• diffuse - uses the diffuse component of the input material to
set the amount of absorption.
• map - uses the map input to set the amount of absorption.
REFERENCE GUIDE
884
Trilinear |
Trilinear
The Trilinear node lets you warp your 3D geometry object as a whole by using trilinear interpolation from
the object’s bounding box to the given destination box. For example, you can use this node to create
animated object deformations, such as the squash of a bouncing ball.
Inputs and Controls

Type Name
Input unnamed The 3D geometry object you want to warp. If you want to warp
several objects together, you can also connect a MergeGeo node
here.

(Scripting) Value
Trilinear Tab
geometry.
REFERENCE GUIDE
885
Trilinear |

(Scripting) Value
changed.
same settings.
reset shape to reset_shape N/A Cancels your changes and resets the input to its original
input shape.
copy copy_shape N/A Copies the current positions of the bounding box
corners (p0-p7) into a temporary buffer.
paste paste_shape N/A Pastes previously copied positions of the bounding box
corners (p0-p7).
set key set_key_shape N/A Sets keys for the positions of all the bounding box
corners (p0-p7) on the current frame. This is the same as
selecting Setkey from the animation menu next to each
of the controls.
delete key del_key_shape N/A Deletes the keys on the positions of all the bounding
box corners (p0-p7) on the current frame.
delete anim del_anim_ N/A Deletes the keys on the positions of all the bounding
shape box corners (p0-p7) on all frames.
p0 xyz p0 -0.5, -0.5, -0.5 The x, y, and z coordinates for the p0 corner of the
bounding box. You can also adjust these by dragging
the corner in the Viewer.
p1 xyz p1 -0.5, 0.5, -0.5 The x, y, and z coordinates for the p1 corner of the
p2 xyz p2 0.5, 0.5, -0.5 The x, y, and z coordinates for the p2 corner of the
p3 xyz p3 0.5, -0.5, -0.5 The x, y, and z coordinates for the p3 corner of the
REFERENCE GUIDE
886
Trilinear |

(Scripting) Value

p4 xyz p4 -0.5, -0.5, 0.5 The x, y, and z coordinates for the p4 corner of the
p5 xyz p5 -0.5, 0.5, 0.5 The x, y, and z coordinates for the p5 corner of the
p6 xyz p6 0.5, 0.5, 0.5 The x, y, and z coordinates for the p6 corner of the
p7 xyz p7 0.5, -0.5, 0.5 The x, y, and z coordinates for the p7 corner of the
Source box Tab
use incoming src_use_bbox enabled By default, the node warps the bounding box of the
bounding box input object. If you'd rather warp a custom bounding
box, uncheck this and use the controls below to define
the custom box.
reset source reset_src_ N/A Cancels any changes you've made to the custom
box to input shape bounding box and resets it to the original settings taken
from the input.
This control is only available if you have unchecked

useincomingboundingbox.
box src_color red The color of the custom bounding box in the 3D Viewer.
This control is only available if you have unchecked

useincomingboundingbox.
src0 xyz src0 -0.5, -0.5, -0.5 Sets the x, y, and z coordinates for one of the two
custom bounding box corners you can adjust.
src1 xyz src1 0.5, 0.5, 0.5 Sets the x, y, and z coordinates for one of the two
custom bounding box corners you can adjust.
REFERENCE GUIDE
887
UVProject |
UVProject
The UVProject node sets the UV coordinates for an object, allowing you to project a texture image onto
the object. If the object already has UV coordinates, this node replaces them.
See also Project3D.
Inputs and Controls

Type Name
Input axis/cam The Axis or Camera node controlling the projection attributes.
unnamed The 3D object onto which the texture is projected.

(Scripting) Value
UVProject Tab
• off - doesn't display geometry outlines.
geometry.
REFERENCE GUIDE
888
UVProject |

(Scripting) Value
changed.
same settings.
projection projection_ perspective Sets the projection type for rendered objects. Usually,
mode it’s best to select a type that’s close to the object’s
surface shape.
• off
• perspective
• planar
• spherical
• cylindrical
Note: When projection is set to off, the

UVProject node is essentially disabled.
plane plane XY Sets the projection direction to project the texture

image along the x, y, or z axis.
Note: This dropdown menu is only available if

you selected planar as the projection type.
project on project_on both Sets which surfaces of a 3D object the projection falls on
through the camera:
• both
• front
• back
view frustum frustum_ disabled When enabled, the UVProject node affects only the
REFERENCE GUIDE
889
UVTile |

(Scripting) Value
culling culling vertices inside the camera view frustum. Any vertices
outside the view frustum will not be affected and they
still keep their original uv coordinates.
invert u u_invert disabled When enabled, the u coordinates are inverted on the
horizontal axis.
invert v v_invert disabled When enabled, the v coordinates are inverted on the
vertical axis.
u scale u_scale 1 Stretches or squashes the texture on the horizontal axis.
v scale v_scale 1 Stretches or squashes the texture on the vertical axis.
attrib name uv_attrib_ uv Specifies the name of the attribute to use for the UV
name coordinates of the upstream geometry. If your geometry
has more than one set of UV coordinates, you can
choose the attribute that suits the projection type.
UVTile
When projecting textures onto an object, use the UVTile node to modify the coordinates of a given UV (or
UDIM) patch and adjust the behavior of individual patches.
Inputs and Controls

Type Name
Input unnamed The Read node with the UV (or UDIM) patch whose coordinates or
behavior you want to adjust.
REFERENCE GUIDE
890
Wireframe |

(Scripting) Value
UVTile Tab
udim udim 1001 The patch’s UDIM coordinates. Enter an alternative UDIM
value here to offset the given patch to that point. Check
enable to adjust the value here.
UDIM is a numbering scheme that identifies the first

texture that’s applied to the (0,0) - (1,1) region as 1001,
with numbers increasing by one for each texture in the U
direction, and by ten for each texture in the V direction.
enable udim_enable disabled Check to enable the UDIM numbering scheme. Checking
this disables the u and v fields.
u tile_u 0 The patch’s u coordinate in UV space. Enter a value here

to offset the given patch to that point.
v tile_v 0 The patch’s v coordinate in UV space. Enter a value here

to offset the given patch to that point.
wrap mode wrap_u clamp Sets how existing patches are used to fill in the areas on
the u axis where unique textures are not present:
• clamp - Patch edges are stretched indefinitely.
• repeat - Patches are tiled repeatedly.
• mirror - Patches are mirrored repeatedly.
wrap_v clamp Sets how existing patches are used to fill in the areas on
the v axis where unique textures are not present:
• clamp - Patch edges are stretched indefinitely.
• repeat - Patches are tiled repeatedly.
• mirror - Patches are mirrored repeatedly.
translate translate 0, 0 Translates the entire UV grid map on the x and y axes.
scale scale 1 Scales the width and height of the UV grid map.
Wireframe
REFERENCE GUIDE
891
Wireframe |
The Wireframe node allows you to render a wireframe overlay on the surface of your geometry object.
Note: The Wireframe node only works if you are rendering your 3D scene using ScanlineRender.
Inputs and Controls

Type Name


(Scripting) Value
Wireframe Tab

operation operation opaque Sets how the wireframe overlay is applied to the
geometry:
• opaque - display the wireframe on fully opaque black
input geometry.
• see through - display the wireframe on fully
transparent geometry.
• over - display the wireframe on top of the input
shader or texture.
• multiply - multiply the wireframe by the input shader
or texture and display it on fully transparent geometry.
• modulate - apply standard diffuse shading to the
REFERENCE GUIDE
892
WriteGeo |

(Scripting) Value
wireframe and display it on top of the input shader or

texture. This takes into account any lights in the scene.
line width line_width 0.5 Sets the width of the wireframe lines (in pixels).
line color line_color 1 Sets the color and transparency of the wireframe lines.
WriteGeo
Writes out geometry to a specified file. You can export:
• OBJ files
• FBX files containing meshes, point clouds, cameras, lights, and axes.
• Alembic (.abc) files containing meshes, point clouds, cameras, or axes. For more information on Alembic,
see http://code.google.com/p/alembic/.
You can write the geometry by clicking the Execute button and specifying a range of frames you want
written or by running the Nuke script with the -x switch.
See also ReadGeo.
Inputs and Controls

Type Name
Input unnamed The geometry or scene you want to write out to a specified file.

Value
WriteGeo Tab
file file none Enter the file path and file name or use the folder
icon to browse to the required location.
REFERENCE GUIDE
893
WriteGeo |

Value
Execute Execute N/A Click to write the input geometry to the file
specified in the file control.
frame range first 1 When limit to range is enabled, enter the first
frame to write to the specified file.
last 1 When limit to range is enabled, enter the last

frame to write to the specified file.
limit to use_limit disabled When enabled, enter the first and last frames to
range write to the specified file.
views views main When you’re working with stereo footage, select
the required view.
file type file_type none Select the file type you intend to write:
• none - Nuke attempts to guess the file type from
the file name, but you cannot set any type-
specific controls.
• abc - when the WriteGeo node is connected to a
Scene node, writes to an .abc file and exposes the
Alembic-specific controls (see abc Options).
• fbx - when the WriteGeo node is connected to a
Scene node, writes to an .fbx file and exposes the
FBX-specific controls (see fbx Options).
• obj - write to an .obj file.
abc Options (These controls are only exposed when file type is set to abc.)
geometries writeGeometries enabled When enabled, write the scene geometries into the
.abc file.
pointClouds writePointClouds enabled When enabled, write the scene point clouds into
the .abc file.
cameras writeCameras enabled When enabled, write the scene cameras to the .abc
file.
axes writeAxes enabled When enabled, write the scene axes into the .abc
file.
REFERENCE GUIDE
894
WriteGeo |

Value
storage storageFormat HDF Select the storage format to use when writing the
format file:
• HDF - A storage format that maintains backwards
compatibility.
• Ogawa - A storage format that offers faster file
reading and smaller files.
fbx Options (These controls are only exposed when file type is set to fbx.)
geometries writeGeometries enabled When enabled, write the scene geometries into the
.fbx file.
cameras writeCameras enabled When enabled, write the scene cameras to the .fbx
file.
lights writeLights enabled When enabled, write the scene lights into the .fbx
file.
axes writeAxes enabled When enabled, write the scene axes into the .fbx
file.
point clouds writePointClouds enabled When enabled, write the scene point clouds into
the .fbx file.
ascii file asciiFileFormat disabled When disabled, a binary .fbx file is written. Check
format ascii file format to write the scene as an ascii .fbx
file.
animate animateMeshVertices disabled When enabled, the mesh vertices are animated and
mesh keyframes created at every frame. The animated
vertices meshes use vertex point cache for the data and a
directory with _fpc appended to the file name is
created to contain the point caches.

aborts.
REFERENCE GUIDE
895
WriteGeo |

Value
the render aborts.
frame rendering. They are not called if the render aborts.
If they throw an exception, the render aborts.

REFERENCE GUIDE
896
Particles Nodes
Particles nodes (NukeX and Nuke Studio only) deal with Nuke's built-in particle system, often used for
creating effects like fog, smoke, rain, snow, and explosions.
ParticleBlinkScript
The ParticleBlinkScript node is similar to the BlinkScript node, running Foundry's Blink framework it
enables us to write our code once and run it on any supported device. This is achieved through code
translation, in which the Blink code is turned into specific code for each target device.
Unlike the BlinkScript node which works on images, the ParticleBlinkScript node works on a particle
system. It’s a ParticleOp which lets you write a Blink kernel to modify the particles, allowing custom
behavior that can’t be achieved with Nuke’s built in particle nodes.
Inputs and Controls

Type Name
Input particles The source particles to which the ParticleBlinkScript kernel is

applied.
Properties - BlinkScript Tab

(UI) Value
display display unchanged Adjust the display characteristics of the 3D
REFERENCE GUIDE
897
|

(UI) Value
object. These settings don’t affect the render

output of the scene; these are for display
l off - hides the 3D geometry object.

l wireframe - displays only the outlines
of the object’s geometry.
l solid - displays all geometry with a
solid color.
l solid+wireframe - displays the
geometry as solid color with the
object’s geometry outlines.
l textured - displays only the surface
texture.
l textured+wireframe - displays the
selectable selectable enabled When enabled, you can make selections as

normal in the Viewer. When disabled, points
cannot be selected or changed.
render render_mode unchanged Sets how the object is rendered. This control
is independent from the display selection,
but has the same settings.
l off - hides the 3D geometry object.

l wireframe - displays only the outlines
of the object’s geometry.
l solid - displays all geometry with a
solid color.
l solid+wireframe - displays the
geometry as solid color with the
object’s geometry outlines.
l textured - displays only the surface
texture.
l textured+wireframe - displays the
Kernel File kernelFile none Sets the file path of the kernel used for Load
REFERENCE GUIDE
898
|

(UI) Value
and Save operations. BlinkScript kernels use

the .rpp file extension.
Load loadKernelFile N/A Click to load and compile the kernel

specified in the Kernel File field.
Save saveKernelFile N/A Click to save the current kernel to the

location and file name specified in the
Kernel File field.
Kernel Source
Kernel kernelSource ParticleDrag Enter the kernel you intend to compile or

Editor edit the default ParticleDrag to suit your
purposes.
Click the Recompile button to see the result

of any changes.
Recompile recompile N/A Click to recompile the kernel currently in the

Kernel Editor. Any compile errors present are
displayed on screen.
Create createKnobs N/A Click to create knobs for any parameters

Knobs which don't already have one.
Publish publish N/A Click to create a group containing a copy of

the current node. This can then be saved as a
Gizmo if desired.
Protect protectKernelWhenPublishing disabled When enabled, the kernel in the published

Kernel group will be encoded and will not be
readable when the published node is written
to a script.
REFERENCE GUIDE
899
|
Properties - Conditions Tab

(Scripting) Value
probability probability 1 This defines the probability that any given particle is
affected by this node.
For each incoming particle, the op generates a random

number between 0 and 1. If the number is greater than the
defined probability value, then that particle is ignored and
left unchanged.
0.0 - No particles are ever affected.
1.0 - Every particle is affected
min age min_age 0 Define the minimum required age of a particle for it to be
affected by this node. This node only applies to particles
which are above this age, where the age is normalized to 0-
1.
max age max_age 1 Define the maximum required age of a particle for it to be
affected by this node. This node only applies to particles
which are below this age, where the age is normalized to 0-
1.
random seed 0 Altering this value produces different results for anything
seed that requires random numbers. It is used in particular for
probability but can also be used for the range knobs within
a ParticleEmitter.
channels channels all Select which channels to apply this force to.
Particles can be emitted on one or more channels, only

selected channels will have this force applied to them.
REFERENCE GUIDE
900
|
Properties - Region Tab

region region none Sets the region which you want to use to confine the particle
effect to. For example, if you choose a sphere, only particles
inside that sphere shaped region are affected by particle
effects.
l none - all particles are affected as normal.

l sphere, box, half-space, and cylinder - controls the
region’s boundary shape.
invert region_invert disabled If enabled, particles outside the region are affected, rather
region than those inside it.
l Import chan file - import a channel file and transform

the region marker according to the transformation data
Boujou.
l Export chan file - export the translation parameters
that you’ve applied to the region marker as a channel
artists.
snap_menu N/A l Match selection position - the region marker is

snapped to a new position depending on the points
selected.
l Match selection position, orientation - the region
marker is snapped to a new position and orientation
l Match selection position, orientation, size - the
region marker is snapped to a new position,
orientation, and size depending on the points selected.
REFERENCE GUIDE
901
|

order translation (T). The possible operation combinations are SRT,
STR, RST, RTS, TSR, TRS.
rotation rot_order ZXY Sets the order of rotation. The possible axial combinations
order are ZXY, XYZ, XZY, YXZ, YZX, ZXY, ZYX.
translate translate 0, 0, 0 Lets you translate the region marker along the x, y, and z axes.
You can also adjust translate values by clicking and dragging
the axis in the 3D Viewer.
rotate rotate 0, 0, 0 Lets you rotate the region marker around the x, y, and z axes.
You can adjust rotate values by holding down Ctrl/Cmd and
dragging in the 3D Viewer.
scale scaling 1, 1, 1 Lets you scale the region marker on the x, y, and z axes.
uniform uniform_scale 1 Lets you scale the region marker simultaneously on the x, y,
scale and z axes.
skew skew 0, 0, 0 Lets you skew the region marker on the x, y, and z axes.
pivot pivot 0, 0, 0 When you make changes to the region marker’s position,
scaling, skewing, and rotation, these occur from the location
of the object’s origin point or pivot. The pivot x, y, and z
controls allow you to offset the pivot point and move it
anywhere you like - you can even move it outside of the
object. Subsequent transformations applied will then occur
relative to the new pivot point location.
You can also hold down Ctrl/Cmd+Alt and drag the pivot
point to a new location in the 3D Viewer.
Local Matrix
specify useMatrix N/A Enable this control to specify matrix values for the object
matrix you’re transforming as an alternative to setting transform,
scale, skew and pivot values above.
matrix matrix N/A The matrix displays values from the object’s transform, rotate,
scale, skew, and pivot controls.
REFERENCE GUIDE
902
ParticleBounce |


ParticleBounce
With ParticleBounce, you can make your particles appear to bounce off a 3D shape instead of traveling
through it. Use the ParticleBounce object control in combination with real geometry to make it look as
though the particles are interacting with it. Each particle checks for intersection with the geometry, then
bounces according to the normal at that point. This is currently limited to the primitive Nuke shapes plane,
sphere, and cylinder.
Inputs and Controls

Type Name
Input geometry The geometry to bounce particles off when object is set to input.
particles The particle system you intend to affect.

(Scripting) Value
ParticleBounce Tab
display display unchanged Adjust the display characteristics of the particles. These
• off - hides the particles.
• wireframe - displays only the outlines of the particle.
REFERENCE GUIDE
903
ParticleBounce |

(Scripting) Value
• solid - displays all particles with a solid color.

• solid+wireframe - displays the particles as solid color
with the particles outlines.
surface texture.
• unchanged - doesn't change the particles display
selectable selectable enabled When enabled, you can select the particles in the Viewer
by clicking on them.
render render_mode unchanged Sets how the particles will render. This control is
same settings.
external out_bounce_ bounce Controls particle behavior when a bounce is detected on

bounce mode mode the external surface of the bounce object:
• none - nothing happens and the particle moves as
normal.
• bounce - the particle bounces off the surface.
• kill - the particle is deleted.
new channels out_new_ none Sets the channel where a particle should be assigned to
channels when an external bounce is detected.
bounce out_bounce 1 Sets the strength of the bounce effect. Higher values
cause the bounce angle to be closer to the normal, and
vice versa.
friction out_friction 0 Controls the slow down of particles as they hit the
external surface at an angle to the normal. Higher values
cause a backspin effect - the bounce angle becomes
closer to the normal, and vice versa.
internal in_bounce_ bounce Controls particle behavior when a bounce is detected on

bounce mode mode the internal surface of the bounce object:
REFERENCE GUIDE
904
ParticleBounce |

(Scripting) Value
• none - nothing happens and the particle moves as

normal.
• bounce - the particle bounces off the surface.
• kill - the particle is deleted.
new channels in_new_ none Sets the channel where a particle should be assigned to
channels when an internal bounce is detected.
bounce in_bounce 1 Sets the strength of the bounce effect. Higher values
cause the bounce angle to be closer to the normal, and
vice versa.
friction in_friction 0 Controls the slow down of particles as they hit the
internal surface at an angle to the normal. Higher values
cause a backspin effect - the bounce angle becomes
closer to the normal, and vice versa.
object object plane Sets the geometry used to simulate bounce. Selecting
plane, sphere, or cylinder uses Nuke standard
primitives, but you can select input to use custom
geometry attached to the geometry input.
bounceaxis_ N/A Allows you to to import or export a channel file:

file_menu • Import chan file - import a channel file and transform
the bounce object according to the transformation
data in the channel file. Channel files contain a set of
Boujou.
that you’ve applied to the bounce object as a channel
artists.
bounceaxis_ N/A • Match selection position - the bounce object is

snap_menu snapped to a new position depending on the points
selected.
REFERENCE GUIDE
905
ParticleBounce |

(Scripting) Value
• Match selection position, orientation - the bounce

object is snapped to a new position and orientation
bounce object is snapped to a new position,
transform bounceaxis_ SRT Sets the operation order for scale (S), rotation (R), and
order xform_order translation (T). The possible operation combinations are
rotation order bounceaxis_ ZXY Sets the order of rotation. The possible axial
rot_order combinations are ZXY, XYZ, XZY, YXZ, YZX, ZXY, ZYX.
translate bounceaxis_ 0, 0, 0 Lets you translate the bounce object along the x, y, and z
translate axes. You can also adjust translate values by clicking and
rotate bounceaxis_ 0, 0, 0 Lets you rotate the bounce object around the x, y, and z
rotate axes. You can adjust rotate values by holding down
scale bounceaxis_ 1, 1, 1 Lets you scale the bounce object on the x, y, and z axes.
scaling
uniform scale bounceaxis_ 1 Lets you scale the bounce object simultaneously on the
uniform_scale x, y, and z axes.
skew bounceaxis_ 0, 0, 0 Lets you skew the bounce object on the x, y, and z axes.
skew
pivot bounceaxis_ 0, 0, 0 When you make changes to the bounce object’s

pivot position, scaling, skewing, and rotation, these occur from
the location of the object’s origin point or pivot. The
pivot x, y, and z controls allow you to offset the pivot
point and move it anywhere you like - you can even
move it outside of the object. Subsequent
transformations applied will then occur relative to the
new pivot point location.
REFERENCE GUIDE
906
ParticleBounce |

(Scripting) Value

Local Matrix
matrix bounceaxis_ N/A The matrix displays values from the object’s transform,
matrix rotate, scale, skew, and pivot controls.

Conditions Tab
probability Sets the probability that this node affects your particles.
If you set this to zero, the node won’t affect any
particles, and if the value is 1, the node will affect every
particle.
min age Limits the effect of this node to particles above this
minimum age. The age of the particle is its lifetime
normalized between 0 and 1.
max age Limits the effect of this node to particles below this
maximum age. The age of the particle is its lifetime
random seed Sets the integer to change the results of generated

randomness in your particles. You can achieve slightly
different effects by changing this number.
channels Specifies which particle channels the effect of this node

should be applied to. Channels a and b are arbitrary
REFERENCE GUIDE
907
ParticleBounce |

(Scripting) Value
names for channels which are useful if you want different

ParticleEmitter nodes or other particle force nodes to
have an effect on separate channels.
Region Tab
region region none Sets the region which you want to use to confine the
particle effect to. For example, if you choose a sphere,
only particles inside that sphere shaped region will be
affected by particle effects.
• none - all particles are affected as normal.
• sphere, box, half-space, and cylinder - controls the
invert region region_invert disabled When enabled, particles outside the region are affected
rather than those inside it.

Boujou.
artists.
snap_menu N/A • Match selection position - the region marker is

selected.
• Match selection position, orientation - the region
REFERENCE GUIDE
908
ParticleBounce |

(Scripting) Value

translate translate 0, 0, 0 Lets you translate the region marker along the x, y, and z
rotate rotate 0, 0, 0 Lets you rotate the region marker around the x, y, and z
axes. You can adjust rotate values by holding down
uniform scale uniform_scale 1 Lets you scale the region marker simultaneously on the
x, y, and z axes.
location.

REFERENCE GUIDE
909
ParticleBounce |

(Scripting) Value
Local Matrix

REFERENCE GUIDE
910
ParticleCache |
ParticleCache
The ParticleCache node allows you to store the geometry simulation for a particle system to file. It can
then be read back in different sessions of Nuke or on different machines without the need for
recalculation.
This allows a particle system to be produced by an artist and then used by a render farm without
recalculation, speeding up render times.
Caching does not replace the particle system rendered, as it relies on its inputs.
Note: ParticleCache nodes must be placed at the bottom of a single particle system or multiple
merged particle systems. They cannot be placed beneath a Scene node connected to two
separate streams or in the middle of a string of particle nodes.
Inputs and Controls

Type Name
Input particles The particle system you intend to write to cache.

(UI) Value
Cache Tab
file file none Sets the file path to save the particle system to using
the .nkpc file extension.
REFERENCE GUIDE
911
ParticleCache |

(UI) Value
Note: ParticleCache may need to render up

to 100 sub-frames. To account for this, it
adds decimals to the file name's frame
number. For example, if the file name in the
file field is particle_cache.####.nkpc,
ParticleCache may generate files called
particle_cache.0001.01.nkpc, particle_
cache.0001.02.nkpc, and so on.
padding particle_cache_ 1 The number of extra frames added to the start and
padding end of the ParticleCache render. This is required if
you have nodes downstream requesting frames
outside the normal frame range due to motion blur
(for example, if you have increased the shutter value
in a downstream ScanlineRender node).
If you get a "Particle cache data not found" error

when you enable read from file, you need to
increase this value.
Render Render N/A Click to render the particle system to the location
specified in the file control.
read from particle_cache_ disabled When enabled, use the particle system rendered to
file read_from_file the location specified in the file control.
When disabled, re-evaluate the particle system as

normal.

aborts.
frame each individual frame. If they throw an exception, the
render aborts.
REFERENCE GUIDE
912
ParticleCurve |

(UI) Value

ParticleCurve
With ParticleCurve, you can apply a curve to particle properties (such as size or mass) to change them over
time. You can, for instance, adjust the curve for your particles’ alpha channel so that each particle fades to
invisibility toward the end of its lifetime.
Inputs and Controls

Type Name
Input particles The particle system to which you intend to apply the curve.

(Scripting) Value
ParticleCurve Tab
REFERENCE GUIDE
913
ParticleCurve |

(Scripting) Value

surface texture.
same settings.
curve editor curves N/A Specifies the curves to apply to particles each frame. The
x-axis is the particle's age (expressed as a proportion of
its maximum lifetime, so 0.0 is a new particle and 1.0 is a
particle about to expire).
Note: Changes in the curve editor are limited

by the Apply curve to checkboxes. For
example, changing the size curve has no effect
if size is not checked under Apply curve to.
reset N/A N/A Click to reset all curves to their original values.
Apply curve to:
rgb affect_rgb enabled When enabled, changes to the curves are applied in the
red, green, and blue channels.
alpha affect_alpha enabled When enabled, changes to the curves are applied in the
alpha channel.
size affect_size disabled When enabled, changes to the curves are applied to the
size of the particles.
REFERENCE GUIDE
914
ParticleCurve |

(Scripting) Value
mass affect_mass disabled When enabled, changes to the curves are applied to the
mass of the particles.
Conditions Tab
probability probability 1 Sets the probability that this node affects your particles.
particle.
min age min_age 0 Limits the effect of this node to particles above this
max age max_age 1 Limits the effect of this node to particles below this
random seed seed 0 Sets the integer to change the results of generated
channels channels all Specifies which particle channels the effect of this node
Region Tab
REFERENCE GUIDE
915
ParticleCurve |

(Scripting) Value

Boujou.
artists.

selected.
REFERENCE GUIDE
916
ParticleDirectionalForce |

(Scripting) Value
x, y, and z axes.
location.

Local Matrix

ParticleDirectionalForce
Use ParticleDirectionalForce to apply a directional force, rather like gravity, to your particles.
REFERENCE GUIDE
917
Inputs and Controls

Type Name
Input particles The particle system to which you intend to apply the force.

(Scripting) Value
ParticleDirectionalForce Tab
• solid+wireframe - displays the particles as solid
color with the particles outlines.
selectable selectable enabled When enabled, you can select the particles in the
Viewer by clicking on them.
same settings.
Strength xyz strength 0, 0, 0 Sets the direction of force applied to the particles on
the x, y, and z axes.
Uniform uniformStrength 1 Sets the strength applied to the directional force

Strength specified in the Strength control.
REFERENCE GUIDE
918

(Scripting) Value
Conditions Tab
probability probability 1 Sets the probability that this node affects your
particles. If you set this to zero, the node won’t affect
any particles, and if the value is 1, the node will affect
every particle.
channels channels all Specifies which particle channels the effect of this
node should be applied to. Channels a and b are
arbitrary names for channels which are useful if you
want different ParticleEmitter nodes or other particle
force nodes to have an effect on separate channels.
Region Tab
invert region region_invert disabled When enabled, particles outside the region are
affected rather than those inside it.

REFERENCE GUIDE
919

(Scripting) Value
transform the region marker according to the

of animation in a given shot. You can create and
export them using Nuke or 3D tracking software, such
as 3D-Equalizer, Maya, or Boujou.
file. This is a useful method of sharing setups
between artists.

selected.
orientation, and size depending on the points
selected.
translate translate 0, 0, 0 Lets you translate the region marker along the x, y, and
z axes. You can also adjust translate values by clicking
and dragging the axis in the 3D Viewer.
axes. you can adjust rotate values by holding down
REFERENCE GUIDE
920
ParticleDrag |

(Scripting) Value
x, y, and z axes.
pivot pivot 0, 0, 0 When you make changes to the region marker’s

position, scaling, skewing, and rotation, these occur
from the location of the object’s origin point or pivot.
The pivot x, y, and z controls allow you to offset the
pivot point and move it anywhere you like - you can
even move it outside of the object. Subsequent

Local Matrix
specify useMatrix N/A Enable this control to specify matrix values for the
matrix object you’re transforming as an alternative to setting

values, for example, if you wanted to align objects in a
scene.
ParticleDrag
With ParticleDrag, you can apply drag on your particles to gradually alter their velocity over time. This
means particles will start off fast and gradually slow down, unless you use negative values to apply the
reverse effect.
REFERENCE GUIDE
921
ParticleDrag |
The formula is: particle velocity*(1-drag). If the drag value is 0.01, the particles lose 1% of their velocity
per frame.
Inputs and Controls

Type Name
Input particles The particle system to which you intend to apply drag.

(Scripting) Value
ParticleDrag Tab
surface texture.
same settings.
drag drag 0 Sets the proportion of each particle's velocity that is lost
per frame, where 0.01 = 1 percent and 1 = 100 percent.
REFERENCE GUIDE
922
ParticleDrag |

(Scripting) Value
Note: Negative drag values increase the

particle velocity.
rotational rotational_ 0 Sets the proportion of each particle's rotation that is lost
drag drag per frame, where 0.01 = 1 percent and 1 = 100 percent.
Note: Negative values increase the particle

rotation velocity.
Conditions Tab
particle.
Region Tab
REFERENCE GUIDE
923
ParticleDrag |

(Scripting) Value


Boujou.
artists.

selected.
REFERENCE GUIDE
924
ParticleDrag |

(Scripting) Value
x, y, and z axes.
location.

Local Matrix

REFERENCE GUIDE
925
ParticleEmitter |
ParticleEmitter
ParticleEmitter is the only required node for creating particles, but with no geometry input the normal
along which particles travel is the y-axis. Once you’ve connected a Viewer and your geometry, click play on
the timeline to see the default set of particles emitting from your geometry.
Inputs and Controls

Type Name
Input particle The image or geometry to emit as particles - when you connect up
this input, another particle input is created allowing you to mix
particle types. You can also use a PositionToPoints point cloud as the
emitted particle.
emit The geometry from which the particles are emitted. You can also
emit from PositionToPoints point clouds using any normals
information present.
See PositionToPoints for more information.
merge Another particle system, including another ParticleEmitter.

(Scripting) Value
ParticleEmitter Tab
REFERENCE GUIDE
926
ParticleEmitter |

(Scripting) Value

surface texture.
same settings.
start at start_frame 0 Sets which frame the particles start emitting. Negative
values cause the particles to start before frame 0, and
positive values delay the particles.
For example, you might want your particles to start

before time zero, allowing them to spread out before
rendering - the snow is already there, not just starting to
fall.
channels channels a The particles only exist in the selected channel(s).
emit from emit_from points Sets where particles emit from:

• points - emit from geometry points, including
PositionToPoints point clouds.
• edges - emit from geometry edges.
• faces - emit from geometry faces.
REFERENCE GUIDE
927
ParticleEmitter |

(Scripting) Value
Note: Enabling only emit from selected

allows you to emit from points selected in a
preceding GeoSelect node.
• bbox - emit from within the volume defined by the

bbox of the geometry, which can reduce the need for
pre-roll, start at, when creating volumetric particle
simulations.
emit_order randomly Sets how particles are emitted:

• randomly - particles are emitted from a random point
on each face, though the same seed is used each time
once generated.
• uniformly - particles are emitted from every vertex on
every frame, ignoring the emission rate setting.
• in order - particles are emitted from the vertices of
the geometry input in ascending order.
randomize_ no_random_ Sets how the initial particle velocity directions are
type direction randomized:
• no random direction - directions are not randomized.
The emit object's normals are used to determine
direction.
• randomized direction - the initial directions are
completely randomized.
• randomized outwards - a randomly selected
direction, depending on emit from:
bbox - particles move away from the center.
points, edges, and faces - particles move away at no
more than 90 degrees from the nearest normal.
emission rate rate 10 Sets the exact number of particles emitted per frame
and is affected by the rate channel control. If your rate
channel is less than 1 all the way through, the emission
rate lessens.
REFERENCE GUIDE
928
ParticleEmitter |

(Scripting) Value
only emit selection_only disabled When enabled, only points selected in a preceding
from selected GeoSelect node are used as emit from sites. See
GeoSelect for more information.
threshold selection_ 0.5 When only emit from selected is enabled and emit
threshold from is set to something other than points, this control
indicates the threshold used for selection testing. For
example, if set to 0.5 then the boundary is exactly
halfway between selected and unselected points.
vertex vertex_rate 1 When emit_order is set to uniformly, sets the number

emission rate of particles emitted per vertex per frame. It is an exact
number, not an average, and decreases if you change the
probability.
Note: This control is affected by the setting of

the rate channel. If your rate channel is not at a
value of 1 all the way through, the emission
rate is less.
rate variation rate_variation 0 Produces a random emission variation by adding a

Gaussian-distributed random number to the emission
rate value multiplied by the rate variation value.
Roughly speaking, 0 is equal to the emission rate and 1

is very random variation.
rate channel N/A disabled Enables the associated rate channel to the right.
channel to none.
rate_channel none Sets the channel of the input geometry texture which is
used to modulate the emission rate. You can use this
control to emit particles from certain areas of the input
geometry.
For example, if you are emitting from a Card which has a

Ramp texture, more particles are emitted from the light
REFERENCE GUIDE
929
ParticleEmitter |

(Scripting) Value
part of the Ramp (values closer to 1) than from the dark

parts (values closer to 0).
max lifetime lifetime 10 Sets the number of frames that each particle exists for.
max lifetime lifetime_ 0 Produces a random lifetime variation by adding a

range variation Gaussian-distributed random number to the max
lifetime value multiplied by the max lifetime range
value.
Roughly speaking, 0 is equal to the max lifetime and 1

lifetime N/A disabled Enables the associated lifetime channel to the right.
channel Disabling this checkbox is the same as setting the
channel to none.
lifetime_ none Sets the channel of the input geometry texture which is
channel used to modulate the max lifetime.

Ramp texture, particles emitted from the light parts of
the ramp (values closer to 1) have a lifetime value closer
to that set in the max lifetime than particles emitted
from the dark parts (values closer to 0).
halflife halflife 0 Sets a halflife causing a fraction of the particles to die

randomly each frame such that on average, half of the
particles that existed at a given time will continue to
exist after the halflife (in the same way as radioactive
decay).
For example, after twice the halflife value in frames,

only a quarter of the original particles will be left, and so
on.
velocity velocity 1 Sets the speed in units per frame of the particles as they
leave the emitter.
REFERENCE GUIDE
930
ParticleEmitter |

(Scripting) Value
velocity range velocity_ 0 Produces a random velocity variation by adding a

variation Gaussian-distributed random number to the velocity
value multiplied by the velocity range value.
Roughly speaking, 0 is equal to the velocity and 1 is very

random variation.
velocity N/A disabled Enables the associated velocity channel to the right.
channel Disabling this checkbox is the same as setting the
channel to none.
velocity_ none Sets the channel of the input geometry texture which is
channel used to modulate the velocity.

Ramp texture, particles are emitted from the light parts
of the ramp at a higher velocity (values closer to 1) than
from the dark parts (values closer to 0).
rotation rotation_ 0 Sets how quickly particles spin around their local y axis
velocity velocity in 3D space.
The y axis points in the direction the particles were

initially emitted, but then stays unchanged. You can use
the ParticleMotionAlign node to align particles along
their direction of motion, or the ParticleLookAt node to
cause all particles to align in a particular direction.
Note: Rotation velocity is only applicable to

particles with geometry, sprite rotation is not
currently supported.
rotation rotation_ 0 Produces a random rotation velocityby adding a

velocity range velocity_ Gaussian-distributed random number to the rotation
variation velocity value multiplied by the rotationvelocity range
value.
Roughly speaking, 0 is equal to the rotationvelocity and
REFERENCE GUIDE
931
ParticleEmitter |

(Scripting) Value
1 is very random variation.
rotation N/A disabled Enables the associated rotation velocity channel to the
velocity right. Disabling this checkbox is the same as setting the
channel channel to none.
rotation_ none Sets the channel of the input geometry texture which is
velocity_ used to modulate the rotation velocity.
channel
Ramp texture, particles are emitted from the light parts
of the ramp at a higher rotation velocity (values closer
to 1) than from the dark parts (values closer to 0).
size size 0.1 Sets the size of each particle. If the particle input is
geometry, the instance of that geometry at each particle
is affected by this value.
size range size_variation 0 Produces a random variation in particle sizeby adding a

Gaussian-distributed random number to the size value
multiplied by the size range value.
Roughly speaking, 0 is equal to the size and 1 is very

random variation.
size channel N/A disabled Enables the associated size channel to the right.
channel to none.
size_channel none Sets the channel of the input geometry texture which is
used to modulate the size of the particles.

Ramp texture, the size of the particles emitted from the
light part of the Ramp (values closer to 1) will be greater
than that of those emitted from the dark parts (values
closer to 0).
REFERENCE GUIDE
932
ParticleEmitter |

(Scripting) Value
mass mass 1 Sets the mass of each particle when a force is applied,
for example when using the ParticlePointForce node.
mass range mass_ 0 Produces a random in particle massby adding a

variation Gaussian-distributed random number to the mass value
multiplied by the mass range value.
Roughly speaking, 0 is equal to the mass and 1 is very

random variation.
mass channel N/A disabled Enables the associated mass channel to the right.
channel to none.
mass_channel none Sets the channel of the input geometry texture which is
used to modulate the mass of the particles. You can use
this control to emit particles with different mass from
different areas.

Ramp texture, the particles from the light part of the
Ramp (values closer to 1) will have a higher mass value
(based on a percentage of the mass value) than from the
dark parts (values closer to 0).
transfer transfer_ 0 Sets the transfer strength of any velocity that the initial
velocity velocity emitter had to the particles.
Setting this to 0 transfers no velocity to the spawned

particles. At value 1, full velocity is transferred from the
originating particle.
transfer transfer_ 1 Sets the time, in frames, to look forward and backward
window window to determine the transfer velocity.
spread spread 0 Applies a spread to the particles - by default, a cone

shape around the direction of emission.
color color 1 Sets the initial color of particles when the particle input
is not connected.
REFERENCE GUIDE
933
ParticleEmitter |

(Scripting) Value
color from color_from_ disabled When enabled, the particles take their initial color from
texture texture the color of the emitter geometry texture at the point of
emission. The color modulates as the texture changes.
input order input_order randomly Sets the order of particle inputs used for the
representation:
• randomly - a random input is picked for each particle
emitted.
• in order - the inputs are cycled through for each
particle emitted.
start at start_frame_ first Determines which frame a newly emitted particle's

animation representation starts from:
• first - uses the first frame from the input for each new
particle.
• in order - uses successive frames for each new particle.
• current - uses the frame from which the particle was
emitted.
• random - uses a random frame within the
representation's frame range.
limit to range start_wrap disabled When enabled, particle start frames are forced to be
within the range of the representation input. This causes
the start at > in order and current options to wrap
round back to the start of the frame range once it has
been exceeded, in a continuous loop.
advance frame_ in steps Determines whether particle representations animate

advance after emission:
• constant - particles do not animate, and keep the
same representation for their entire lifetime.
• in steps - the particle’s representation advances frame
by frame.
• randomly - a different random frame is picked each
time.
max clip max_clip_ 100 Sets the number of frames used in animation from the
REFERENCE GUIDE
934
ParticleEmitter |

(Scripting) Value
length length representation input.
random seed seed 0 Sets a random seed to create more varied results, for
example in conjunction with the range controls in the
emitter such as velocity range. You can achieve slightly
different effects by changing the seed number.
Region Tab

Boujou.
artists.

selected.
REFERENCE GUIDE
935
ParticleEmitter |

(Scripting) Value

x, y, and z axes.
location.

REFERENCE GUIDE
936
ParticleExpression |

(Scripting) Value
Local Matrix

ParticleExpression
With ParticleExpression, you can adjust your particles by setting expressions on their attributes. Using
expressions gives you a vast variety of ways of adjusting the way your particles behave. You can use a
similar expression syntax as you would elsewhere in Nuke, with the exception that some functions which
work in normal Nuke expressions aren't available in particle expressions and vice versa. The main
difference between Nuke’s Expression node and ParticleExpression is that particle expressions can return a
3D vector instead of just a single floating point number. If a particle expression returns a single number N
in a field that expects a vector (such as velocity or acceleration) it will be converted into a vector with N
for each of its components. For more information about functions you can use with ParticleExpression, see
Nuke's online help.
Inputs and Controls

Type Name
Input particles The particle system to which you intend to apply expressions.
REFERENCE GUIDE
937

(Scripting) Value
ParticleExpression Tab
[temp name temp_name0 N/A If you need to use a long expression in several fields,
field] you can use this row for assigning the expression
temporarily to a variable. Enter the variable name here,
and the expression on the right of the = sign. You can
then use the variable to represent the expression in the
= fields next to the other particle controls.

expression in the = fields next to the other particle
controls.
per-particle temp_expr0_pp enabled When enabled, this temporary expression is evaluated

per-particle.

controls.

per-particle.
REFERENCE GUIDE
938

(Scripting) Value

controls.

per-particle.

controls.

per-particle.
color colexpr N/A Sets expressions controlling particle color.
only on new colexpr_onlynew disabled When disabled, the expression affects all particles.
When enabled, the expression only affects new
REFERENCE GUIDE
939

(Scripting) Value
particles.
opacity opacity_expr N/A Sets expressions controlling particle opacity.
only on new opacityexpr_ disabled When disabled, the expression affects all particles.
onlynew When enabled, the expression only affects new
particles.
size sizeexpr N/A Sets expressions controlling particle size.
only on new sizeexpr_ disabled When disabled, the expression affects all particles.
particles.
mass massexpr N/A Sets expressions controlling particle mass.
only on new massexpr_ disabled When disabled, the expression affects all particles.
particles.
accel accelexpr N/A Sets expressions controlling particle acceleration.
only on new accelexpr_ disabled When disabled, the expression affects all particles.
particles.
force forceexpr N/A Sets expressions controlling particle force.
only on new forceexpr_ disabled When disabled, the expression affects all particles.
particles.
pos posexpr N/A Sets expressions controlling particle position.
only on new _ disabled When disabled, the expression affects all particles.
onlynewposexpr When enabled, the expression only affects new
particles.
vel velexpr N/A Sets expressions controlling particle velocity.
only on new velexpr_onlynew disabled When disabled, the expression affects all particles.
When enabled, the expression only affects new
particles.
REFERENCE GUIDE
940

(Scripting) Value
frame frameexpr N/A Sets expressions controlling the frame used for
particle geometry or texture.
only on new frameexpr_ disabled When disabled, the expression affects all particles.
particles.
chans channelsexpr N/A Sets expressions controlling whether or not the

channels specified in the set to control are applied to
the particles.
only on new channelsexpr_ disabled When disabled, the expression affects all particles.
particles.
set to new_channels b Sets the new channels to assign to particles matching

the chans expression.
Note: Particles are deleted if none is

selected.
Conditions Tab
probability probability 1 Sets the probability that this node affects your
particles. If you set this to zero, the node won’t affect
any particles, and if the value is 1, the node will affect
every particle.
REFERENCE GUIDE
941

(Scripting) Value
channels channels all Specifies which particle channels the effect of this
node should be applied to. Channels a and b are
arbitrary names for channels which are useful if you
want different ParticleEmitter nodes or other particle
force nodes to have an effect on separate channels.
Region Tab
invert region region_invert disabled When enabled, particles outside the region are
affected rather than those inside it.

transform the region marker according to the
of animation in a given shot. You can create and
export them using Nuke or 3D tracking software, such
as 3D-Equalizer, Maya, or Boujou.
file. This is a useful method of sharing setups
between artists.

selected.
REFERENCE GUIDE
942

(Scripting) Value

orientation, and size depending on the points
selected.
translate translate 0, 0, 0 Lets you translate the region marker along the x, y, and
z axes. You can also adjust translate values by clicking
and dragging the axis in the 3D Viewer.
x, y, and z axes.
pivot pivot 0, 0, 0 When you make changes to the region marker’s

position, scaling, skewing, and rotation, these occur
from the location of the object’s origin point or pivot.
The pivot x, y, and z controls allow you to offset the
pivot point and move it anywhere you like - you can
even move it outside of the object. Subsequent

REFERENCE GUIDE
943
ParticleGravity |

(Scripting) Value
Local Matrix
specify useMatrix N/A Enable this control to specify matrix values for the
matrix object you’re transforming as an alternative to setting

values, for example, if you wanted to align objects in a
scene.
ParticleGravity
With ParticleGravity, you can apply gravity to your particles. Unlike our familiar gravity, particle gravity
works in any or all of the x, y and z directions, and you can also set negative values to them. When you
connect the ParticleGravity node to your particle stream, an arrow appears in the Viewer, which you can
then use to control the gravity. The bigger and longer the arrow, the stronger the gravity effect.
Inputs and Controls

Type Name
Input particles The particle system to which you intend to apply gravity.

(Scripting) Value
ParticleGravity Tab
REFERENCE GUIDE
944
ParticleGravity |

(Scripting) Value
surface texture.
same settings.
from xyz from N/A Sets the point of origin for the vector on the x, y, and z
axes.
to xyz to N/A Sets the acceleration to be applied to all particles,

depending on the direction and magnitude of the
vector applied - the bigger and longer the arrow, the
stronger the gravity effect.
The position of the vector has no effect and Gravity

affects all particles equally, irrespective of mass.
Conditions Tab
particle.
REFERENCE GUIDE
945
ParticleGravity |

(Scripting) Value
Region Tab

Boujou.
REFERENCE GUIDE
946
ParticleGravity |

(Scripting) Value

artists.

selected.
x, y, and z axes.
REFERENCE GUIDE
947
ParticleLookAt |

(Scripting) Value
location.

Local Matrix

ParticleLookAt
With ParticleLookAt, you can determine a 3D point at which all the particles within a certain region are
looking toward.
REFERENCE GUIDE
948
ParticleLookAt |
Inputs and Controls

Type Name
Input particles The particle system you intend to align with a certain point.

(Scripting) Value
ParticleLookAt Tab
surface texture.
same settings.
position xyz position 0, 0, 0 Sets the xyz coordinates that all particles look toward.
strength strength 1 Sets the speed at which particles turn to look at the
specified position. A strength of 0 has no effect on the
particles and a strength of 1 forces them to look at the
position instantly.
REFERENCE GUIDE
949
ParticleLookAt |

(Scripting) Value
Conditions Tab
particle.
Region Tab

REFERENCE GUIDE
950
ParticleLookAt |

(Scripting) Value

Boujou.
artists.

selected.
REFERENCE GUIDE
951
ParticleMerge |

(Scripting) Value
x, y, and z axes.
location.

Local Matrix

ParticleMerge
If you have more than one set of particle nodes, and you want to combine them into one stream,
ParticleMerge is your node. Attach your particle streams to ParticleMerge’s numbered inputs and you’re all
set.
REFERENCE GUIDE
952
ParticleMotionAlign |
Inputs and Controls

Type Name
Input numbered inputs The first particle system you intend to merge. Adding a particle
system spawns another numbered input.
ParticleMotionAlign
You can add ParticleMotionAlign in your particle stream to realign all the particles along their direction of
motion.
Inputs and Controls

Type Name
Input particles The particle system you intend to align.

(Scripting) Value
ParticleMotionAlign Tab
REFERENCE GUIDE
953

(Scripting) Value

surface texture.
same settings.
Conditions Tab
particle.
REFERENCE GUIDE
954

(Scripting) Value

Region Tab

Boujou.
artists.

selected.
REFERENCE GUIDE
955

(Scripting) Value
x, y, and z axes.
location.

REFERENCE GUIDE
956
ParticlePointForce |

(Scripting) Value
Local Matrix

ParticlePointForce
With ParticlePointForce, you can attract or repel particles to or from a certain point in the 3D space. You
can use an animated or still axis expression-linked to the position value, or a value you have chosen to
enter manually.
Inputs and Controls

Type Name
Input particles The particle system to which you intend to apply forces.

(Scripting) Value
ParticlePointForce Tab
REFERENCE GUIDE
957

(Scripting) Value

surface texture.
same settings.
strength strength 0 Sets the strength of the attraction, measured in inverse

square. A negative value is an attraction and a positive
value is a repulsion.
falloff falloff inverse square How quickly the strength of the attraction falls off with
respect to distance:
• none - attraction or repulsion is constant, regardless of
the distance from the position values.
• inverse and inverse square - falloff is calculated with
respect to the selected law.
radius radius 0 The radius of influence. Particles outside the radius are
ignored, particles inside will be attracted to or repelled
from the chosen point position.
position xyz position 0, 0, 0 Sets the position of the point which attracts or repels
the particles.
REFERENCE GUIDE
958

(Scripting) Value
Conditions Tab
particle.
Region Tab

REFERENCE GUIDE
959

(Scripting) Value

Boujou.
artists.

selected.
REFERENCE GUIDE
960
ParticleSettings |

(Scripting) Value
x, y, and z axes.
location.

Local Matrix

ParticleSettings
Use the steps per frame control in the ParticleSettings control panel to adjust how many steps of particle
simulation take place per animation frame.
REFERENCE GUIDE
961
ParticleSpawn |
Inputs and Controls

Type Name
Input particles The particle system for which you intend modify settings.

(Scripting) Value
ParticleSettings Tab
steps per steps_per_ 1 Sets how many steps of particle simulation take place
frame frame per animation frame.
Sometimes simulations cannot generate enough

accuracy by calculating once per frame and the resulting
particle movement can appear jagged. Increasing steps
per frame causes the simulation to asses the movement
of the particles multiple times per frame to smooth
movement.
ParticleSpawn
If you’re looking to have your existing particles emit even more particles, you should turn to
ParticleSpawn. This node emits particles from existing particles.
Inputs and Controls

Type Name
Input particle The image or geometry to emit as particles - when you connect up
REFERENCE GUIDE
962
ParticleSpawn |

Type Name
this input, another particle input is created allowing you to mix

particle types.
merge Another particle system, including another ParticleEmitter, on which

to base the spawned particles.

(Scripting) Value
ParticleSpawn Tab
display display unchanged Adjust the display characteristics of the spawned

particles. These settings don’t affect the render output
of the scene; these are for display purposes only in the
3D Viewer.
surface texture.
selectable selectable enabled When enabled, you can select the spawned particles in
the Viewer by clicking on them.
same settings.
channels channels b The spawned particles only exist in the selected channel
(s).
emission rate rate 10 Sets the exact number of particles spawned per frame
and is affected by the rate channel control of the
REFERENCE GUIDE
963
ParticleSpawn |

(Scripting) Value
ParticleEmitter. If your rate channel is less than 1 all the

way through, the spawn rate lessens.
rate variation rate_variation 0 Produces a random emission variation by adding a

Gaussian-distributed random number to the emission
rate value multiplied by the rate variation value.
Roughly speaking, 0 is equal to the emission rate and 1

lifetime lifetime 10 Sets the number of frames that each spawned particle
exists for.
lifetime range lifetime_ 0 Produces a random lifetime variation by adding a

variation Gaussian-distributed random number to the lifetime
value multiplied by the lifetime range value.
Roughly speaking, 0 is equal to the lifetime and 1 is very

random variation.
halflife halflife 0 Sets a halflife causing a fraction of the spawned

particles to die randomly each frame such that on
average, half of the particles that existed at a given time
will continue to exist after the halflife (in the same way
as radioactive decay).
For example, after twice the halflife value in frames,

only a quarter of the original spawned particles will be
left, and so on.
velocity velocity 1 Sets the speed in units per frame of the particles as they
spawn.
velocity range velocity_ 0 Produces a random velocity variation by adding a

variation Gaussian-distributed random number to the velocity
value multiplied by the velocity range value.
Roughly speaking, 0 is equal to the velocity and 1 is very

random variation.
rotation rotation_ 0 Sets how quickly spawned particles spin around their
REFERENCE GUIDE
964
ParticleSpawn |

(Scripting) Value
velocity velocity local y axis in 3D space.
The y axis points in the direction the particles were

initially spawned, but then stays unchanged. You can use
the ParticleMotionAlign node to align particles along
their direction of motion, or the ParticleLookAt node to
cause all particles to align in a particular direction.
rotation rotation_ 0 Produces a random rotation velocity by adding a

velocity range velocity_ Gaussian-distributed random number to the rotation
variation velocity value multiplied by the rotationvelocity range
value.
Roughly speaking, 0 is equal to the rotationvelocity and

1 is very random variation.
size size 0.1 Sets the size of each spawned particle. If the particle
input is geometry, the instance of that geometry at each
particle is affected by this value.
size range size_variation 0 Produces a random variation in particle size by adding a

Gaussian-distributed random number to the size value
multiplied by the size range value.
Roughly speaking, 0 is equal to the size and 1 is very

random variation.
mass mass 0.1 Sets the mass of each spawned particle when a force is
applied, for example when using the ParticlePointForce
node.
mass range mass_ 0 Produces a random in particle mass by adding a

variation Gaussian-distributed random number to the mass value
multiplied by the mass range value.
Roughly speaking, 0 is equal to the mass and 1 is very

random variation.
transfer transfer_ 1 Transfers velocity from the initial emitter to the spawned
velocity velocity particles on a sliding scale - all inherited velocity at 1,
REFERENCE GUIDE
965
ParticleSpawn |

(Scripting) Value
and zero inherited velocity at 0.
spread spread 0 Applies a spread to the particles - by default, a cone

shape around the direction of emission.
conservation conserve_ enabled When enabled, the mass of any spawned particle is
of mass mass subtracted from the originating particle. If the mass of
the originating particle is zero at the end of a frame, it is
deleted.
conservation conserve_ enabled When enabled, the momentum of any spawned particle
of momentum is subtracted from the momentum of the originating
momentum particle, in correspondence with Newton's third law of
motion.
align velocity velocity_ enabled When enabled, spawned particles inherit the velocity
to direction direction from the parent particle, but the direction of motion is
of motion dictated by the ParticleSpawn controls.
When disabled, spawned particles inherit the direction

of motion from the parent particle, but with increased
velocity.
color color 1 Sets the initial color of spawned particles when the
particle input is not connected.
inherit color inherit_color disabled When enabled, the spawned particle’s color is taken
from the originating particle. When disabled, spawned
particles derive their color from the particle input, if
connected, or the color control.
input order input_order randomly Sets the order of particle inputs used for the
representation:
• randomly - a random input is picked for each particle
emitted.
• in order - the inputs are cycled through for each
particle emitted.
start at start_frame_ first Determines which frame a newly emitted particle's
REFERENCE GUIDE
966
ParticleSpawn |

(Scripting) Value
animation representation starts from:

• first - uses the first frame from the input for each new
particle.
• in order - uses successive frames for each new particle.
• current - uses the frame from which the particle was
emitted.
• random - uses a random frame within the
representation's frame range.
limit to range start_wrap disabled When enabled, particle start frames are forced to be
within the range of the representation input. This causes
the start at > in order and current options to wrap
round back to the start of the frame range once it has
been exceeded, in a continuous loop.
advance frame_ in steps Determines whether particle representations animate

advance after emission:
• constant - particles do not animate, and keep the
same representation for their entire lifetime.
• in steps - the particle’s representation advances frame
by frame.
• randomly - a different random frame is picked each
time.
max clip max_clip_ 100 Sets the number of frames used in animation from the
length length representation input.
Conditions Tab
particle.
REFERENCE GUIDE
967
ParticleSpawn |

(Scripting) Value
Region Tab

Boujou.
REFERENCE GUIDE
968
ParticleSpawn |

(Scripting) Value
artists.

selected.
x, y, and z axes.
REFERENCE GUIDE
969
ParticleSpeedLimit |

(Scripting) Value
location.

Local Matrix

ParticleSpeedLimit
ParticleSpeedLimit restricts particles to a specified minimum and maximum speed.
Inputs and Controls

Type Name
Input particles The particle system in which you intend limit particle speeds.
REFERENCE GUIDE
970

(Scripting) Value
ParticleSpeedLimit Tab
surface texture.
same settings.
minimum minimum 0 Sets the minimum speed at which a particle can travel.
maximum maximum 100 Sets the maximum speed at which a particle can travel.
strength strength 1 Sets how quickly the particles change speed toward the
minimum and maximum values. A strength of 0 has no
effect on the particles and a strength of 1 forces them
to change speed instantly.
Conditions Tab
particle.
REFERENCE GUIDE
971

(Scripting) Value
Region Tab

Boujou.
REFERENCE GUIDE
972

(Scripting) Value

artists.

selected.
x, y, and z axes.
REFERENCE GUIDE
973
ParticleToGeo |

(Scripting) Value
location.

Local Matrix

ParticleToGeo
With ParticleToGeo, you can control particles in a simulation by channel, giving you the ability to isolate
certain particles at any point in the Node Graph. For example, you might want to freeze certain particles
within a simulation while allowing others to emit as normal or apply a particle effect to the particles in a
single channel.
REFERENCE GUIDE
974
ParticleToGeo |
Inputs and Controls

Type Name
Input particles The particle system for which you intend to create particle
geometry.

(Scripting) Value
ParticleToGeo Tab
surface texture.
render render_mode unchanged Sets how the particles render. This control is
same settings.
channels channels all Sets which particles are affected, by channel. For
example, setting channels to a only affects particles
emitted into the a channel in the ParticleEmitter node's
Properties panel.
REFERENCE GUIDE
975
ParticleTurbulence |

(Scripting) Value
align mode align none Sets the way sprite particles are aligned:
• none - sprite particles are not aligned.
• spin - sprite particles are aligned according to their
rotation.
• velocity - sprite particles are aligned according to
their velocity.
Note: This control has no effect on particles

generated from geometry.
frame frame 50 When specify frame is enabled, the frame control

freezes the particles in the selected channels at the
specified frame in the simulation.
For example, you could freeze particles in a channel

called fog at frame 50 while allowing particles in the
other channels to simulate as normal.
specify frame use_frame disabled When enabled, the frame control can be used to freeze
particles at a certain frame in the simulation.
ParticleTurbulence
ParticleTurbulence applies noise to the particle movement, dispersing the particles in the x, y and/or z
directions.
REFERENCE GUIDE
976
Inputs and Controls

Type Name
Input particles The particle system to which you intend to apply turbulence.

(Scripting) Value
ParticleTurbulence Tab
surface texture.
same settings.
strength xyz strength 0, 0, 0 Sets the strength of the turbulence force, which can be
different along each axis.
scale xyz scale 1, 1, 1 Sets the scale of the noise lattice, which can be different
along each axis.
REFERENCE GUIDE
977

(Scripting) Value
offset xyz offset 0, 0, 0 Sets the offset applied to the noise lattice, which can be
different along each axis.
Conditions Tab
particle.
Region Tab
REFERENCE GUIDE
978

(Scripting) Value

Boujou.
artists.

selected.
REFERENCE GUIDE
979
ParticleVortex |

(Scripting) Value
x, y, and z axes.
location.

Local Matrix

ParticleVortex
ParticleVortex applies a circular force to the particles and attracts them to an imaginary line, thus creating a
whirlpool of particles. When you connect the ParticleVortex node to your particle stream, an arrow
REFERENCE GUIDE
980
ParticleVortex |
appears in the 3D Viewer, which you can then use to determine direction and velocity of the vortex effect.
The bigger and longer the arrow, the stronger the effect.
Inputs and Controls

Type Name
Input particles The particle system to which you intend to apply the vortex.

(Scripting) Value
ParticleVortex Tab
surface texture.
same settings.
parallel parallel 0 Applies a force to particles in a direction parallel to the

vector line. Negative values flow with the vector line and
positive values against it.
REFERENCE GUIDE
981
ParticleVortex |

(Scripting) Value
parallel falloff parallel_falloff none Sets which falloff law to use when calculating the effect
of the force on each particle:
• none - no falloff occurs, regardless of distance from
the origin.
tangential tangential 0 Applies a force to particles in a direction tangential to

the vector line. Negative values produce clockwise
movement around the vector line and positive values
counter-clockwise.
tangential tangential_ inverse Sets which falloff law to use when calculating the effect
falloff falloff of the force on each particle:
• none - no falloff occurs regardless of distance from
the origin.
radial radial 0 Applies a force which attracts particles to (positive

values), or repels them from (negative values) the vector
line.
radial falloff radial_falloff inverse Sets which falloff law to use when calculating the effect
of the force on each particle:
• none - no falloff occurs regardless of distance from
the origin.
from xyz from 0, 0, 0 Sets the point of origin for the arrow on the x, y, and z
axes.
The arrow determines the direction and velocity of the

vortex effect. The bigger and longer the arrow, the
stronger the effect.
to xyz to 0, 1, 0 Sets the point of destination for the arrow on the x, y,
REFERENCE GUIDE
982
ParticleVortex |

(Scripting) Value
and z axes.

vortex effect. The bigger and longer the arrow, the
Conditions Tab
particle.
Region Tab
REFERENCE GUIDE
983
ParticleVortex |

(Scripting) Value

Boujou.
artists.

selected.
REFERENCE GUIDE
984
ParticleWind |

(Scripting) Value

x, y, and z axes.
location.

Local Matrix

ParticleWind
REFERENCE GUIDE
985
ParticleWind |
With ParticleWind, you can simulate a wind blowing on the particles. When you connect the ParticleWind
node to your particle stream, an arrow appears in the Viewer, which you can then use to determine
direction and velocity of the wind. The bigger and longer the arrow, the stronger the wind effect.
Inputs and Controls

Type Name
Input particles The particle system to which you intend to apply wind.

(Scripting) Value
ParticleWind Tab
surface texture.
same settings.
from xyz from 0, 0, 0 Sets the point of origin for the arrow on the x, y, and z
axes.
REFERENCE GUIDE
986
ParticleWind |

(Scripting) Value

wind effect. The bigger and longer the arrow, the
to xyz to 0, 1, 0 Sets the point of destination for the arrow on the x, y,

and z axes.

wind effect. The bigger and longer the arrow, the
Note: Particles are gradually accelerated, they

aren’t emitted at full speed.
drag drag 0.2 Sets the amount of simulated drag applied to particles.
Higher values drag the particles further from their origin,
and vice versa.
air resistance air_resistance disabled When enabled, simulated resistance is added to the drag
control progressively slowing particles down over time
in a similar way to ParticleDrag.
Conditions Tab
particle.
REFERENCE GUIDE
987
ParticleWind |

(Scripting) Value
Region Tab

Boujou.
artists.

selected.
REFERENCE GUIDE
988
ParticleWind |

(Scripting) Value

x, y, and z axes.
location.

REFERENCE GUIDE
989
ParticleWind |

(Scripting) Value
Local Matrix

REFERENCE GUIDE
990
Deep Nodes
Deep nodes deal with Deep image compositing, where each pixel can have multiple values.
DeepColorCorrect
This is a ColorCorrect node for deep compositing. It applies the color correction to all samples at each
pixel. You can use DeepColorCorrect to create a matte by setting the gain value for your alpha channel to
0 and setting the offset value to 1 in the range you want the matte for.
There are control sets for adjusting shadows, midtones and highlights. You can use the lookup curves on
the Ranges tab to control these.
Inputs and Controls

Type Name
Input unnamed The deep image sequence to receive the color correction. This can
also be a DeepMerge node with merged deep data.

Value
DeepColorCorrect Tab
master
saturation saturation 1 Adjusts the saturation in the entire image.
contrast contrast 1 Adjusts the contrast in the entire image.
gamma gamma 1 Adjusts the gamma in the entire image.
REFERENCE GUIDE
991
|

Value
gain gain 1 Adjusts the gain in the entire image.
offset offset 0 Adjusts the offset in the entire image.
shadows
saturation shadows.saturation 1 Adjusts the saturation in the darkest parts of the

image.
contrast shadows.contrast 1 Adjusts the contrast in the darkest parts of the

image.
gamma shadows.gamma 1 Adjusts the gamma in the darkest parts of the

image.
gain shadows.gain 1 Adjusts the gain in the darkest parts of the

image.
offset shadows.offset 0 Adjusts the offset in the darkest parts of the

image.
midtones
saturation midtones.saturation 1 Adjusts the saturation in the middle range of the

image.
contrast midtones.contrast 1 Adjusts the contrast in the middle range of the

image.
gamma midtones.gamma 1 Adjusts the gamma in the middle range of the

image.
gain midtones.gain 1 Adjusts the gain in the middle range of the

image.
offset midtones.offset 0 Adjusts the offset in the middle range of the

image.
highlights
saturation highlights.saturation 1 Adjusts the saturation in the highlights of the

image.
contrast highlights.contrast 1 Adjusts the contrast in the highlights of the
REFERENCE GUIDE
992
|

Value
image.
gamma highlights.gamma 1 Adjusts the gamma in the highlights of the

image.
gain highlights.gain 1 Adjusts the gain in the highlights of the image.
offset highlights.offset 0 Adjusts the offset in the highlights of the image.
Masking Tab
mix mix 1 Dissolves between the original image at 0 and

the full color correction at 1.
zmap A, B, C, D zmap 0, 1, 1, 1 Sets how much of the color correction to use at

various depths. The x axis represents the distance
from the camera, and the y axis the fraction of
the color correction to use (where 0 is none and
1 is the full effect).
limit_z limit_z disabled When enabled, allows dissolving between the

original image and the full color correction
depending on the depth of the samples, using
the zmap control.
Ranges Tab
test test disabled Lets you overlay the output with black, gray, or
white to show what is considered to be in the
shadows, midtones, or highlights. Green and
magenta indicate a mixture of ranges.
curve editor lookup N/A You can use the shadow and highlight lookup
curves to edit the range of the image that is
considered to be in the shadows or highlights.
You can also look up color information for the
current pixel in the Viewer.
REFERENCE GUIDE
993
DeepCrop |

Value
Warning: DO NOT adjust the midtone

curve. Midtones are always equal to 1
minus the other two curves.
reset N/A N/A Return the selected curve(s) to the default values.
DeepCrop
You can use the DeepCrop node to:
• crop deep data in front of or behind certain planes in depth, or
• crop deep images inside or outside of the crop box in the Viewer (much like the regular Crop node).
Inputs and Controls

Type Name
Input unnamed The deep image sequence to crop. This can also be a DeepMerge
node with merged deep data.

(Scripting) Value
DeepCrop Tab
znear znear 1 Sets the near depth value of the image. By default, any
samples in front of this plane in depth (that is, below this
value) are cropped.
use use_znear enabled Whether to crop samples below the znear value.
zfar zfar 2 Sets the far depth value of the image. By default, any
samples behind this plane in depth (that is, above this
REFERENCE GUIDE
994
DeepCrop |

(Scripting) Value
value) are cropped.
use use_zfar enabled Whether to crop samples above the zfar value.
keep outside outside_ disabled When enabled, samples between the znear and zfar
zrange zrange values are cropped.
When disabled, samples outside the znear to zfar range

are cropped.
bbox bbox N/A Set the size of the crop box in the Viewer. By default,
any areas outside this box are cropped.

• x - the distance (in pixels) between the left edge of the
image and the left side of the crop box.
of the image and the bottom edge of the crop box.
• r - the distance (in pixels) between the left edge of the
image and the right side of the crop box.
the image and the top edge of the crop box.
• h - the height of the crop box. This is only available if

dragging its edges.
use use_bbox enabled Set whether to use the crop box in the Viewer.
keep outside outside_bbox disabled When enabled, any samples inside the crop box are
bbox cropped.
When disabled, any samples outside the crop box are

cropped.
REFERENCE GUIDE
995
DeepExpression |
DeepExpression
The DeepExpression node allows you to apply complex mathematical formulae to deep data using
expressions.
Inputs and Controls

Type Name
Input unnamed The deep image sequence to which you want to apply expressions
to. This can also be a DeepMerge node with merged deep data.

(Scripting) Value
DeepExpression Tab
[variable name temp_name0 N/A If you need to use a long expression in several fields,
fields below the channels.

expression in the fields below the channels.
REFERENCE GUIDE
996
DeepExpression |

(Scripting) Value


- temp_expr3 N/A If you need to use a long expression in several fields,

REFERENCE GUIDE
997
DeepFromFrames |

(Scripting) Value
chans0 chans0 red, green, The channel(s) to which you want to apply expressions
blue, alpha to.

chans1 chans1 front, back The channel(s) to which you want to apply expressions
to.

chans2 chans2 none The channel(s) to which you want to apply expressions
to.

chans3 chans3 none The channel(s) to which you want to apply expressions
to.

[expression N/A N/A Expression fields for each channel selected above.
fields]
DeepFromFrames
The DeepFromFrames node copies multiple input frames from a regular 2D image to samples in a single
deep frame.
REFERENCE GUIDE
998
DeepFromFrames |
Inputs and Controls

Type Name
Input unnamed The image sequence from which you want to copy the frames from.

(Scripting) Value
DeepFromFrames Tab
samples samples 5 Sets the number of frames to sample the input at.
frame range range_first 1 Sets the start frame for sampling the input. For example,
with the default samples value (5) and frame range
value (1-9) DeepFromFrames samples at frames 1, 3, 5, 7
and 9.
range_last 9 Sets the end frame for sampling the input. For example,
with the default samples value (5) and frame range
value (1-9) DeepFromFrames samples at frames 1, 3, 5, 7
and 9.
premult premult enabled Set whether the input needs premultiplying.
When disabled, DeepFromFrames assumes that the input

is already premultiplied.
split alpha alpha_mode multiplicative Set how the alpha channel is split:
mode • additive - performs a straight division by the number
of samples. This does not result in the original alpha
values if the deep samples are flattened later on.
• multiplicative - the samples are split so that each is
1-pow(1-alpha,1/numSamples). If the deep samples
are flattened later on (for example, using
DeepToImage), the alpha values match the originals.
zmin zmin 1 Sets the depth at which the first sampled frame is
placed. This is the front-most sample in depth.
REFERENCE GUIDE
999
DeepFromImage |

(Scripting) Value
zmax zmax 2 Sets the depth at which the last sampled frame is
placed. This is the furthermost sample in depth.
DeepFromImage
Using DeepFromImage you can convert a standard 2D image to a deep image with a single sample for
each pixel at the depth defined by the depth.z channel.
Inputs and Controls

Type Name
Input 1 The 2D image sequence to convert to a deep image.

(Scripting) Value
DeepFromImage Tab
premult input premult disabled Set whether the input needs premultiplying.
When disabled, DeepFromImage assumes that the

input is already premultiplied.
specify z set_z disabled When enabled, you can use the z field below to
manually specify the Z depth for the single sample at
each pixel.
When disabled, the Z depth for the single sample at

each pixel is defined by the input’s depth.z channel.
z z 0 Set the Z depth value to use for the single sample at

each pixel.
REFERENCE GUIDE
1000
DeepMerge |

(Scripting) Value
This control only has an effect when you have enabled

specify z.
DeepMerge
DeepMerge merges samples from multiple deep images, so that each output pixel contains all of the
samples from the same pixel in each input. You can also use it to holdout samples in the B input that are
occluded by samples in the A input or plus overlapping samples from the A and B inputs.
Inputs and Controls

Type Name
Input A, B The deep images that you want to merge.

(Scripting) Value
DeepMerge Tab
operation operation combine Sets how to merge the inputs:

• combine - combine the samples from the A and B
inputs.
• holdout - hold out samples from the B input by the
samples in the A input. This removes or fades out
samples in input B that are occluded by samples in
input A.
• plus - adds A and B overlapping samples, which can be
useful when recombining data after a holdout.
drop hidden drop_hidden disabled When enabled, samples that are completely obscured by
samples other samples are discarded (that is, samples behind
REFERENCE GUIDE
1001
DeepMerge |

(Scripting) Value
other samples that have an alpha of 1).
This control is only available when operation is set to

combine.
drop zero drop_zero_ 0.0000001 Any samples whose alpha value falls below this threshold
threshold threshold are discarded from the holdout. This can help filter out
samples with very low alpha values caused by floating
point inaccuracy.
This control is only available when operation is set to

holdout.
metadata metainput B Sets which input’s metadata is passed down the node
from tree.
Note: When metadata from is set to All and

there are keys with the same name in both
inputs, keys in B override keys in A.
volumetric volumetric_ disabled Enable volumetric holdout if you want Nuke to calculate
holdout holdout occlusion using the values of the holdout samples in
front of samples from main. This is a more accurate
representation of occlusion at depth, but can take longer
to process.
For example:
M = main sample
H = holdout sample
M0 remains unchanged since there are no holdout

samples before it. M2 is affected by the combined H0,
REFERENCE GUIDE
1002
DeepRead |

(Scripting) Value
H1, and H2 holdout samples and M4 is affected by all

holdout samples.
You can now view the result, which is a holdout with red,
green, blue, and alpha channels. Note that the output
image is still a deep image.
Note: When disabled, deep samples that

coincide with a holdout are removed and
everything 'deeper' is also removed.
DeepRead
The DeepRead node loads deep images from disk in two formats:
• DTEX (generated from Pixar’s PhotoRealistic RenderMan® Pro Server).
• Scanline OpenEXR 2.3, or above (tiled OpenEXR 2.3, or above, files are not supported).
Note: In order to load DTEX files, you need to have Pixar's RenderMan Pro Server 20, or earlier,
installed on your machine. You don't need a RenderMan license, however. See Nuke's online help
Unlike standard 2D images that contain a single value for each channel of each pixel, deep images contain
multiple samples per pixel at varying depths. Each sample contains per-pixel information, such as color,
opacity, and camera-relative depth.
See also DeepWrite.
REFERENCE GUIDE
1003
DeepRead |
Inputs and Controls

Value
DeepRead Tab
file file N/A The image to be loaded from disk. This file should
be in:
• DTEX format, generated from RenderMan Pro
Server, or
• Scanline OpenEXR 2.3 format.
For frame numbers, you can use # for each digit or

alternatively, use printf-style formatting (%04d).
Localization localizationPolicy from auto- Sets the local file caching behavior. Copies of the
Policy localize path files are stored in a specified local folder for faster
access times:
location.
Update updateLocalization N/A When this DeepRead node is set to Localization

Policy > on demand, click Update to manually
localize the files from the source files.
format format N/A Set the format of the image file. This is automatically
set based on the file header, but you can manually set
it to any of the available formats. If the format does
not yet exist, you can select new to create a new
format from scratch.
REFERENCE GUIDE
1004
DeepRead |

Value
proxy proxy N/A The location of the proxy image. A proxy image is
used if proxy mode is enabled and the required
resolution is less than or equal to the file size.
proxy format proxy_format root.proxy_ The proxy image is read if the format of the lower
format resolution image is smaller or equal to this. This is
automatically set based on the file header but you
can manually set it to any of the available formats. If
the format does not yet exist, you can select new to
create a new format from scratch.
frame range first N/A Sets the first frame in the range of frames to use for
the image sequence.
before hold Sets how the sequence displays before the first frame
you’ve defined.
• hold - select to show a still picture of the first
frame in the frame range.
span of the frame range before the first frame in
the frame range.
limits.
• black - select to display a black frame before the
first frame.
last N/A Sets the last frame in the range of frames to use for
the image sequence.
after hold Sets how the sequence displays after the last frame
you’ve defined.
• hold - select to show a still picture of the last frame
in the frame range.
span of the frame range after the last frame in the
frame range.
REFERENCE GUIDE
1005
DeepRead |

Value

limits.
• black - select to display a black frame after the last
frame.
frame frame_mode expression By default, Nuke assumes an exact relation between

the current frame processed and the frame read in.
For example, at frame 15, Nuke reads in
image.0015.exr. However, you can change this
behavior using this control. For instance, if you have a
sequence that runs from image.0500.exr to
image.1000.exr, you may want to read in
image.0500.exr at frame 1. You can do so in several
ways:
• expression - enter an expression in the field on the
right. For example, if your clip begins from
image.0500.exr and you want to place this first
the expression frame+499.
• start at - enter a start frame number in the field on
the right. For example, if your sequence begins
from image.0500.exr and you enter 1 in the field,
image0500.exr is read in at frame 1. Similarly, if you
enter 100 in the field, image0500.exr is read in at
frame 100.
• offset - enter a constant offset in the field on the
right. This constant value is added to the current
frame to get the number of the frame that’s read in.
For example, if your clip begins from
image.0500.exr and you want to place this first
499 as the constant offset.
frame N/A The expression, start frame, or offset you want to use
REFERENCE GUIDE
1006
DeepRead |

Value
to alter the relation between the current frame

processed and the frame read in.
original origFirst N/A Set the first frame in the frame range you want to use
range from the input clip. For example, if you set the
original range to be from frame 40 to 50, then the
DeepRead node indicator in the Dope Sheet only
shows that clip as ten frames long. This is for visual
reference, rather than a change in what is read in, and
it may make it easier to work with many DeepRead
nodes in the Dope Sheet.
origLast N/A Set the last frame in the frame range you want to use
from the input clip. For example, if you set the
original range to be from frame 40 to 50, then the
DeepRead node indicator in the Dope Sheet only
shows that clip as ten frames long. This is for visual
reference, rather than a change in what is read in, and
it may make it easier to work with many DeepRead
nodes in the Dope Sheet.
missing on_error error Sets what to do if there is an error opening the file:
frames • error - display an error on the node and in the
Viewer at any missing frames.
• black - set any missing frames to black.
• checkerboard - replace any missing frames with a
checkerboard image.
• nearest frame - replace any missing frames with
their nearest frame.
reload reload N/A Reload the deep image from the disk.
dtex Options (These controls are only exposed when loading a .dtex file.)
type type auto Sets how to determine the type of .dtex file:
• auto - Nuke automatically detects the file type by
looking at the subimage name. If the name is either
Deep Shadow or ends with (or is) .deepopacity,
REFERENCE GUIDE
1007
DeepRead |

Value
Nuke treats the file as a deep opacity file.

• deepopacity - forces Nuke to treat the file as an
accumulated deep opacity file, corresponding to a
RenderMan Display Driver configuration of:
Display "Filename.dtex" "deepshad" "deepopacity"

• alpha - forces Nuke to treat the file as the newer
point-sampled alpha or color, corresponding to a
RenderMan Display Driver configuration of either:
Display "Filename.dtex" "deepshad" "a" or
Display "Filename.dtex" "deepshad" "rgba"
discrete discrete disabled When enabled, Nuke treats the .dtex file as discrete
samples, with the front and back being the same.
This control is only relevant for deep opacity files, as

color deep compositing files are always discrete.
premultiply premult disabled When enabled, Nuke premultiplies the values from
the .dtex file.
When disabled, Nuke assumes the values have

already been premultiplied.
raw values raw disabled When enabled, deep samples are read "as is", without
any processing.
exr Options (These controls are only exposed when loading a .exr file.)
do not attach noprefix disabled When enabled, metadata keys are read as they are,
prefix without attaching a prefix to them.
When disabled, the prefix exr is attached to

metadata keys to make them distinct from other
metadata in the tree.
REFERENCE GUIDE
1008
DeepRecolor |
DeepRecolor
DeepRecolor merges:
• a standard 2D image (color input) and
• a deep image file that only contains opacity for each sample (depth input).
It spreads the color at each pixel of the color input across all the samples of the corresponding pixel in
the depth input, producing a deep output image.
Inputs and Controls

Type Name
Input color A standard 2D image sequence whose channels you want to copy
onto a deep image. This should be an unpremultiplied image. If
your input image is premultiplied, you can add an Unpremult node
between the image and the DeepRecolor node.
depth A deep image to recolor.

(Scripting) Value
DeepRecolor Tab
channels channels rgb The channels to copy from the color input.

target input targetInputAlpha disabled When disabled, Nuke distributes the color to each
alpha sample by unpremultiplying by the color image’s
alpha and then remultiplying by the alpha of each
sample.
However, sometimes the alpha from the final high-

quality flat render might not match the alpha
REFERENCE GUIDE
1009
DeepReformat |

(Scripting) Value
represented by a given set of deep samples. This

could be the result of the compression that usually
happens to deep files on disk, or some change to the
shader.
When target input alpha is enabled, the color input’s

alpha is distributed amongst the samples so that the
final resulting alpha after flattening of the deep data
matches the color input’s alpha.
DeepReformat
DeepReformat is the Reformat node for deep data. You can use it to set your deep image’s dimensions,
scale, and so on.
Inputs and Controls

Type Name
Input unnamed The deep image to resize. This can also be a DeepMerge node with
merged deep data.

(Scripting) Value
DeepReformat Tab
type type to format Select how to resize the image:

• to format - Set the output width and height to the
selected format. Choose the format in the output
format dropdown menu. If the format does not yet
exist, you can select new to create a new format from
REFERENCE GUIDE
1010
DeepReformat |

(Scripting) Value
scratch. The default setting, root.format, resizes the

image to the format indicated on the Project Settings
dialog.
• to box - Set the output width and height to
dimensions you define (in pixels). Enter values in the
width/height and pixel aspect fields to specify the
dimensions. This option is useful for creating a
thumbnail of the input image.
• scale - Set the output width and height to a multiple
of the input size. Use the scale slider to define the
factor. The scale factor is rounded slightly, so that the
output image is an integer number of pixels in the
direction chosen under resize type.
output format format root.format The format to which you want to output the deep image
sequence. If the format does not yet exist, you can select
new to create a new format from scratch. The default
setting, root.format, resizes the image to the format
indicated in the Project Settings.

format.
width/height box_width 200 The output width for your deep image. The units are in
pixels.

box.
box_height 200 The output height for your deep image. The units are in
pixels.

box.
force this box_fixed disabled When enabled, the output image matches the
shape width/height fields exactly, even if the original image is
a different shape. To achieve this, one direction gets
either clipped or padded.
REFERENCE GUIDE
1011
DeepReformat |

(Scripting) Value
When disabled, the output image is approximately the

same shape as the original, rounded to the nearest
integer number of pixels.

box.
pixel aspect box_pixel_ 1 Sets the pixel aspect ratio for the output image.
aspect
box.
scale scale 1 The scale factor for the width and the height. To scale
each direction separately using different scale factors,
click the 2 button.
This control is only available if you have set type to

scale.
resize type resize width Choose the method by which you preserve or override
the original pixel aspect ratio. Select:
• none - to not resize the original.
• width - to scale the original until its width matches the
output width. Height is then scaled in such a manner as
to preserve the original aspect ratio.
• height - to scale the original so that it fills the output
height. Width is then scaled in such a manner as to
preserve the original aspect ratio.
• fit - to scale the original so that its smallest side fills
the output width or height. The longest side is then
aspect ratio.
• fill - to scale the original so that its longest side fills
the output width or height. The smallest side is then
aspect ratio.
• distort - to scale the original so that both sides fill the
REFERENCE GUIDE
1012
DeepReformat |

(Scripting) Value
output dimensions. This option does not preserve the

original aspect ratio, so distortions may occur.
center center enabled When enabled, DeepReformat translates the image to

center it in the output.
When disabled, DeepReformat translates the image so

that the lower left corners line up.
flip flip disabled When enabled, flips the image upside down.
flop flop disabled When enabled, flops the image left and right.
turn turn disabled When enabled, rotates the image 90 degrees counter-
clockwise.
black outside black_outside disabled When enabled, pixels outside the image boundary are
set to black.
When disabled, the outside area is filled with the

outermost pixels of the image area.
preserve pbb disabled When enabled, pixels outside the output format are
bounding box preserved.
When disabled, pixels outside the output format are

clipped off.
REFERENCE GUIDE
1013
DeepSample |
DeepSample
Use this node to sample any given pixel in a deep image. Move the pos indicator over pixels in the Viewer
to produce deep sample information in the DeepSample control panel.
Inputs and Controls

Type Name
Input unnamed The deep image to sample. This can also be a DeepMerge node with
merged deep data.

(Scripting) Value
DeepSample Tab
pos x, y pos 0, 0 The pixel to sample.
You can also set this by moving the pos indicator in the
Viewer.
accumulate accumulate disabled When enabled, DeepSample shows cumulative sample

values of the sample pixel (the final composited values
as they get built up).
When disabled, DeepSample shows the individual

sample values of the sample pixel.
channels channels rgba The channels to display in the sample list.

[sample list] list N/A The deep samples at the pos pixel. The values shown
include:
• deep.front - The front-most depth value.
REFERENCE GUIDE
1014
DeepToImage |

(Scripting) Value
• deep.back - The furthermost depth value.

• any channels you’ve selected using the channels
controls.
You can use the right-click menu to copy the currently

selected value (Copy) or all values in the table (Copy All).
DeepToImage
You can use this node to flatten a deep image - in other words, merge all the samples in a deep image
into a regular 2D image.
Inputs and Controls

Type Name
Input unnamed The deep image to flatten. This can also be a DeepMerge node with
merged deep data.

(Scripting) Value
DeepToImage Tab
volumetric volumetric_ enabled When enabled, DeepToImage calculates both the front
composition composition and back depths of each sample.
When disabled, DeepToImage only calculates the front

depth of each sample and assumes the samples do not
overlap. The calculation takes less time, but if you have
overlapping samples in your deep image, the resulting
image might not represent every pixel as expected.
REFERENCE GUIDE
1015
DeepToPoints |
DeepToPoints
You can use this node to transform the deep pixel samples into points in 3D space that you can see in
Nuke's 3D view, like a point cloud. This node is useful for position reference.
Inputs and Controls

Type Name
Input camera An optional camera through which you can look at the point cloud.
deep The deep image to view in 3D. This can also be a DeepMerge node
with merged deep data.

(Scripting) Value
DeepToPoints Tab
display display solid The display type for the point cloud:
• off - hides the points.
• wireframe - displays only the outlines of the points.
• solid - displays all points with a solid color.
• solid+wireframe - displays the points as solid color
with their outlines.
• textured - displays the only the surface texture.
surface texture.
selectable selectable enabled When enabled, you can select the points in the Viewer
same settings.
REFERENCE GUIDE
1016
DeepTransform |

(Scripting) Value
Point detail detail 0.25 Sets the density of the point cloud.
Point size pointSize 2 Sets the size of the points.
DeepTransform
You can use this node to reposition deep data. It allows you to:
• translate the samples along the x, y, and z axes, and
• scale the samples’ Z depth values.
Inputs and Controls

Type Name
Input mask An optional image to use as a mask. The z translate and z scale
effects are limited to the non-black areas of the mask.
deep The deep image to transform. This can also be a DeepMerge node
with merged deep data.

(Scripting) Value
DeepTransform Tab
translate x, y, translate 0, 0, 0 Moves all samples at each pixel along the x, y, and z
z axes.
zscale zscale 1 Scales the z depth of all samples at each pixel. Values
above 1 decrease the depth, whereas values below 1
increase it.
mask N/A enabled Enables the associated mask channel to the right.
REFERENCE GUIDE
1017
DeepWrite |

(Scripting) Value

channel to none.
mask_channel rgba.alpha The channel to use as a mask. The z translate and z scale
effects are limited to the non-black areas of this channel.
DeepWrite
This node renders the result of all upstream deep nodes and saves the result to disk in the scanline
OpenEXR 2.3 format (tiled OpenEXR files are not supported).
Note that this node executes all renders at the currently active scale: either full- or proxy-resolution. To
toggle between these, press Ctrl/Cmd+P.
See also DeepRead.
Inputs and Controls

Value
DeepWrite Tab
channels channels rgba Sets the channels to render.

you can use the controls on the right to select
file file none Sets the file path and name of the file to render.
proxy proxy none Sets the file path and name of a relevant proxy
image. This proxy image is used if proxy mode is on
and the required resolution is less than or equal to
the proxy format.
REFERENCE GUIDE
1018
DeepWrite |

Value

relation between the currently processed frame
and the numbering of the frame written out. The
resulting file name for the current frame is
displayed on the DeepWrite node in the Node
Graph. For example, if your deep clip begins from
frame 500 and you want to name that frame
image.0001.exr rather than image.0500.exr, you can
use the expression frame-499. This way, 499
frames are subtracted from the current frame to
get the number for the frame written out. Frame
500 is written out as image.0001.exr, frame 501 is
written out as image.0002.exr, and so on. Another
example of an expression is frame*2. This
expression multiplies the current frame by two to
get the number of the frame that’s written out. At
frame 1, image.0002.exr is written out; at frame 2,
image.0004.exr is written out; at frame 3,
image.0006.exr is written out; and so on.
• startat - Lets you enter a start frame number in the
field on the right. This specifies the frame number
given to the first frame in the sequence. The
numbering of the rest of the frames is offset
accordingly. For example, if your sequence begins
from frame 500 and you enter 1 in the field, frame
500 is written out as image.0001.exr, frame 501 as
image.0002.exr, and so on. Similarly, if you enter
100 in the field, frame 500 is written out as
image.0100.exr.
• offset - Lets you enter a constant offset in the field
on the right. This constant value is added to the
current frame to get the number for the frame
that’s written out. For example, if your clip begins
REFERENCE GUIDE
1019
DeepWrite |

Value
from frame 500 and you want to render this first

frame as image.0001.exr rather than
image.0500.exr, you can use -499 as the constant
offset. This way, 499 is subtracted from the current
frame to get the number for the frame that’s
written out. At frame 500, image.0001.exr is written
out; at frame 501, image.0002.exr is written out,
and so on.
frame none Depending on the frame mode, you can enter a

start frame, an offset, or an expression here.
views views dependent When you’re working with stereo footage, select the
on Project required view to render.
Settings
file type file_type none Sets the rendered file format manually, enabling
type-specific controls.
Note: If file type is left blank, Nuke

disables any file-type-specific controls.
exr Options (These controls are only exposed when file type is set to exr.)
datatype datatype 16 bit half Sets the bit depth of the rendered .exr files:
• 16-bit half
• 32-bit float
compression compression Zip (1 Sets the compression type to apply to the rendered
scanline) file.
metadata metadata default Determines what metadata is included with the

metadata rendered file:
• no metadata - No custom attributes are created,
and only metadata that fills required header fields
is written out.
• default metadata - The optional timecode,
REFERENCE GUIDE
1020
DeepWrite |

Value
edgecode, frame rate, and exposure header fields

are also filled using metadata values.
• default metadata and exr/*
• all metadata except input/*
• all metadata
do not attach noprefix disabled When enabled, unknown metadata keys are written
prefix into the file as they are.
When disabled, unknown metadata keys have the

prefix nuke attached to them when they are written
into the file.
render order render_order 1 When multiple nodes are rendered at once, they are
sorted into increasing order by this number.
Render Render N/A Click to display the pre-Render setup window.
frame range first 1 Sets the first frame of a sequence to render.
last 1 Sets the last frame of a sequence to render.
limit to use_limit disabled When enabled, only frames within the frame range
range are rendered.
Note: If the specified frames are outside

the sequence range, the DeepWrite node
behaves as if it is disabled.

aborts.
REFERENCE GUIDE
1021
DeepWrite |

Value
the render aborts.

REFERENCE GUIDE
1022
Views Nodes
Views nodes deal with stereoscopic or multi-view compositing.
Anaglyph
If you’re working on a stereoscopic or multi-view project, you can use this node to convert your input into
an anaglyph image. By default, the left input is filtered to remove blue and green, and the right view to
remove red.
Anaglyph images produce a 3D effect when viewed with two-color anaglyph glasses.
See also MixViews and SideBySide.
Inputs and Controls

Type Name
Input unnamed The image you want to convert into an anaglyph image. This input
needs to contain at least two views, one for the left eye and another
for the right eye:
• If your input is an .exr file, the file may already contain both views.
This is because .exr files support multiple views in a single file.
• With any other file types, you need to have separate files for the
left and right views. You can combine these files into a single input
using a JoinViews node.
REFERENCE GUIDE
1023
JoinViews |

(Scripting) Value
Anaglyph Tab
views views N/A The views to use for the left and the right eye.
amtcolor amtcolour 0 By default, Anaglyph converts the input images into

grayscale images. To add color into the images, select a
value between 0 (grayscale) and 1 (colored) here.
Note: If the images include areas that are very

red, green, or blue, adding more color into
them may not produce the best possible
results.
(right=red) swap disabled When enabled, the colors are inverted so that the red
channel is used from the right input and the blue and
green channels from the left.
horizontal offset 0 Sets where the images appear in relation to the screen
offset when viewed with anaglyph glasses. To have the images
appear in front of the screen, you would usually enter a
negative value. To have the images appear further away,
you would usually enter a positive value. (This is not the
case if you have swapped the left and right views
around.)
JoinViews
If you have separate files for the views that exist in your Project Settings (for example, one file for the left
view and another for the right), JoinViews lets you combine these files into a single output. This can be
useful if:
• You want to perform the same actions on all views. By default, Nuke applies any changes you make to all
views of the processed node.
• Your subsequent nodes (for example, Split and Join) require a single input that contains two views.
See also Split and Join and OneView.
REFERENCE GUIDE
1024
MixViews |
Inputs and Controls

Type Name
Input There’s one The views that you want to combine into a single output.
input for each
view defined in
the Project
Settings.
MixViews
If you’re working on a stereoscopic or multi-view project, you can use this node to display a blend
between two views in the Viewer. This allows you to check how elements in these views are aligned.
See also Anaglyph and SideBySide.
Inputs and Controls

Type Name
Input unnamed The image that contains the two views you want to dissolve between:
• With any other file types, you need to have separate files for the
left and right views. You can combine these files into a single input
REFERENCE GUIDE
1025
OneView |

(Scripting) Value
MixViews Tab
views views N/A The two views to blend between.
mix mix 0 Adjust the blend between the views. Setting this to 0 or
1 displays only one of the views. Values between 0 and 1
produce a gradual transition from one view to the other.
OneView
OneView lets you separate one view for processing. This is useful if you need to perform different actions
on different views, for example if you want to color correct one view but not the other. Any changes you
make using subsequent nodes are only applied to the extracted view (regardless of what view you are
displaying in the Viewer).
If you want to perform the same action on both views but use different values for each, you can split the
view off in the node’s controls instead.
To merge the views back together further down the node tree, use the JoinViews node.
See also Split and Join.
Inputs and Controls

Type Name
Input unnamed The image that contains the view you want to separate for
processing.

(Scripting) Value
OneView Tab
view view N/A The view to extract.
REFERENCE GUIDE
1026
ReConverge |
ReConverge
If you’re working on a stereoscopic or multi-view project, the ReConverge node lets you shift
convergence (the inward rotation of the eyes or cameras) so that any selected point in the image appears
at screen depth when viewed with 3D glasses. This point is called the convergence point. It is the point
where the lines of sight from the two cameras meet.
At the convergence point, the different views in the image are aligned and appear at screen depth when
viewed with 3D glasses. Anything behind the convergence point appears behind the screen, while anything
in front of it seems to pop out of the screen.
To calculate the convergence shift, the ReConverge node needs a disparity field that maps the location of
a pixel in one view to the location of its corresponding pixel in the other view. To create the disparity
field, you can use O_DisparityGenerator plug-in, which is part of the Ocula plug-in set. Alternatively, you
can create the disparity field in a 3D application. Once you have the disparity field, you can store it in the
channels of an .exr file, or use the Shuffle node to add the disparity channels in the data stream where you
need them.
It is also possible to use the same element as the convergence point throughout the image sequence. To
do so, link the ReConverge node with a Tracker node.
To better view the effect of the ReConverge node, you can insert an Anaglyph node between the
ReConverge node and the Viewer.
Note: The ReConverge node only shifts views horizontally, not vertically.
Inputs and Controls

Type Name
Input unnamed The image whose convergence you want to adjust. This input should
contain at least two views (one for the left eye and another for the
right) and a disparity field:
REFERENCE GUIDE
1027
ShuffleViews |

Type Name
This is because .exr files support multiple views in a single file. With
any other file types, you need to have separate files for the left and
right views. You can combine these files into a single input using a
JoinViews node.
• Make sure there is a disparity field upstream from the image
sequence. If the image sequence is an .exr file, the disparity field
can be included in its channels. Otherwise, you can use a Shuffle
node or Ocula’s O_DisparityGenerator plug-in to add it in the data
stream.

Value
ReConverge Tab
Converge convergencepoint 0, 0 The x and y coordinates of the convergence point (the

upon xy point you want to appear at screen level when viewed
with 3D glasses).
You can also move the point by dragging the

convergence point overlay in the Viewer.
Convergence offset 0 Adjusts the offset for convergence (in pixels). To bring
offset all elements of your image forward from the screen
level, enter a positive value. To move all elements
further away, enter a negative value.
Mode convergemode shift right The view to move to achieve the convergence shift:
• shift right - move the right view.
• shift left - move the left view.
• shift both - move both views.
ShuffleViews
You can use this node to rearrange the views in your script. For example, you can swap the left and right
views around in the pipeline, so that Nuke uses the left input for the right eye and vice versa.
REFERENCE GUIDE
1028
SideBySide |
Inputs and Controls

Type Name
Input unnamed The image that contains the views you want to rearrange.

(Scripting) Value
ShuffleViews Tab
views to use views N/A 1. Click add to add rows of controls to the control
panel, or delete to remove unnecessary rows.
2. Use the buttons or dropdown menus to select which
view to replace with which. For example, to swap the
left and right views around, you need to make the
following selections:
• On one row, select left under get, and right under
from (“get left from right”). The left view is now
replaced with the right view.
• On another row, select right under get, and left
under from (“get right from left”).
SideBySide
If you’re working on a stereoscopic or multi-view project, you can use this node to display two views next
to each other in the Viewer.
See also Anaglyph and MixViews.
REFERENCE GUIDE
1029
Split and Join |
Inputs and Controls

Type Name
Input unnamed The image that contains the two views you want to display side by
side:
• With any other file types, you need to have separate files for the left
and right views. You can combine these files into a single input

(Scripting) Value
SideBySide Tab
vertical vertical disabled Enable this to display one view on top of another.
Disable this to display the two views next to each other.
view1 view1 N/A Select the view to display on the left (or the top if you
have checked vertical).
view2 view2 N/A Select the view to display on the right (or the bottom if
you have checked vertical).
swap swap N/A Swaps the views around in the Viewer.
Split and Join

Split and Join is a combination of OneView and JoinViews nodes. It allows you to extract all the views in
your Project Settings, process them individually, and then merge them back together. It’s no different to
using several OneView nodes together with a JoinViews node, but it makes working faster because you do
not need to add each node separately.
REFERENCE GUIDE
1030
Split and Join |
Once you’ve used Split and Join, you can add any necessary nodes, such as color corrections, between the
OneView and JoinViews nodes.
REFERENCE GUIDE
1031
Metadata Nodes
Metadata nodes deal with information embedded in your images, such as the image’s original bit depth,
width, and height.
AddTimeCode
Adds a time code to the metadata passed down from an input node. The default settings add the time
code 01:00:00:00 to the first frame. This time code is updated throughout the frame range according to
the input clip’s playback speed, which in turn is controlled by the fps (frames per second) control in the
Project Settings. If you change the fps value in the Project Settings, the time code in the metadata is
updated to reflect the change.
Inputs and Controls

Type Name
Input unnamed The node to receive the time code information.

(UI) (Scripting)
TimeCode Tab
startcode startcode 01:00:00:00 Enter the time code value to add to the first frame of the
input.
fps fps 24 If you want to specify the playback speed manually rather
than get it from the Project Settings, disable get FPS
from metadata and enter a new value in this field.
get FPS metafps enabled By default, frames per second is read from the input’s
REFERENCE GUIDE
1032
CompareMetaData |

(UI) (Scripting)
from playback speed, which is controlled by the Project

metadata Settings. If you want to override the value in the input,
disable this control and enter the required value in the fps
field.
start frame frame 1 If you want to specify a different start frame than the first
frame, enable use start frame? and enter a new value in
this field.
use start useFrame disabled By default, the start frame is always 1. If you want to
frame? override this value, enable this control and enter the
required frame in the start frame field.
CompareMetaData
Connecting nodes to the CompareMetaData node enables you to quickly determine metadata differences
between files. The CompareMetaData node only shows keys and values where there are differences
between the two inputs. For example, if you were having trouble with QuickTime frame rates, you could
connect up two QuickTime files and immediately see differences between the input/frame_rate key
values.
Inputs and Controls

Type Name
Input A The first node whose metadata you want to compare.
B The second node whose metadata you want to compare.
REFERENCE GUIDE
1033
CopyMetaData |

(Scripting) Value
CompareMetaData Tab
metadata metadata N/A Displays the differences between the metadata of any
viewer two nodes connected to the inputs.
CopyMetaData
This node copies metadata from one image to another, filtering the metadata to exclude some keys if
required.
Inputs and Controls

Type Name
Input Meta The node whose metadata you want to copy to the output.
Image The node whose image you want to pass down the tree.

Value
CopyMetaData Tab
metadata mergeMode Image + Set the image and metadata combination you
from Meta intend to pass down the tree:
• Image+Meta - add the metadata from the
Meta input to the metadata from the Image
input. If the inputs share any common
metadata keys, the values taken from the Meta
input override those taken from the Image
input.
• Meta only - only use the metadata from the
REFERENCE GUIDE
1034
CopyMetaData |

Value
Meta input.
• Meta+Image - add the metadata from the
Image input to the metadata from the Meta
input. If the inputs share any common
metadata keys, the values taken from the
Image input override those taken from the
Meta input.
• Image only - only use the metadata from the
Image input. This produces the same result as
not using a CopyMetaData at all: both the
image and metadata are taken from the
Image input. This option can be useful if you
want to filter the metadata passed down the
tree.
Meta copy metadatafilter N/A Enter search criteria to select the metadata to
only copy from the Meta input.
For example, if you enter f in the copy only

field under Meta filtering, only the keys and
values that include the letter f are copied from
the Meta input.
within metadatafilterMode keys and Limit the metadata you want to pass down the
values tree to certain fields:
• keys and values
• keys only
• values only
Image copy imageMetadatafilter N/A Enter search criteria to select the metadata to
only copy from the Image input.
For example, if you enter f in the copy only

field under Image filtering, only the keys and
values that include the letter f are copied from
the Image input.
within imageMetadatafilterMode keys and Limit the metadata you want to pass down the
REFERENCE GUIDE
1035
ModifyMetaData |

Value
values tree to certain fields:

• keys and values
• keys only
• values only
ModifyMetaData
The ModifyMetaData node enables you to add, edit, or remove the metadata associated with a file.
Inputs and Controls

Type Name
Input unnamed The node whose metadata you want to modify.

(Scripting) Value
ModifyMetaData Tab
metadata metadata N/A 1. First add, remove, and reorder metadata

viewer placemarkers using the +/- and up/down arrow
buttons.
2. Double-click the key field in the placemarker to
select the metadata you want to edit or remove. The
Pick metadata key dialog opens. If you want to add
metadata, simply enter a new key name in the field
at the bottom of this dialog.
3. Double-click the action field in the placemarker to
toggle the operation between set and remove.
4. If you want to add or edit metadata, double-click the
REFERENCE GUIDE
1036
ViewMetaData |

(Scripting) Value
value field in the placemarker and enter a new value

for the selected key.
ViewMetaData
ViewMetaData lets you inspect metadata (such as the image’s original bit depth, width, and height) passed
down by an input node. The data is split into two columns, keys and the values associated with them.
Inputs and Controls

Type Name
Input unnamed The node whose metadata you want to inspect.

Value
ViewMetaData Tab
metadata shownmetadata N/A Displays the metadata of any node connected to the
viewer input. Once you know which keys exist on the input,
you can reference them in expressions.
The metadata displayed depends on input file type.

For example, a JPEG might only contain input/ keys,
whereas QuickTimes contain input/ and quicktime/
keys.
REFERENCE GUIDE
1037
ViewMetaData |

Value
search metadatafilter N/A Filter the metadata by entering text in this field. By
metadata for default, only metadata containing the keys and
values entered are displayed. You can change this
behavior using the within control.
within metadatafiltermode keys and Defines how to filter the search metadata for
values control. You can search keysandvalues, keysonly, or
valuesonly.
REFERENCE GUIDE
1038
ToolSets Nodes
Tool sets deal with custom tools created in Nuke. You can create your own ToolSets from scratch or
modify the existing 2D and 3D examples that ship with Nuke. The example scripts also contain a link to
footage you can download to use with the examples:
http://thefoundry.s3.amazonaws.com/products/nuke/toolsets/toolset_examples.zip
2D ToolSets
Nuke ships with the following 2D ToolSets:
ToolSet Description Result
CG_Beauty
TB_CG_Beauty A basic approach to splitting

out and merging back
together several layers from
multi-channel EXRs using
some common third-party
renderers, such as Arnold,
Modo, and RenderMan.
Keying
TK_Keying An example of combining

different keys from blue or
greenscreen footage to
produce a better matte. This
ToolSet creates a key for the
edges of the alpha channel,
ignoring fine detail, and then
combines it with a hard, core
matte. Finally the edges are
keyed separately, masked,
and combined with the result.
REFERENCE GUIDE
1039
|
SmartPaint
TSVSmartVectorFix An example of rendering

vectors for a plate and then
using VectorDistort to warp
the source. An example
image is then merged on top
of the source.
TSVSmartVectorGrade An example of rendering

vectors for a plate and then
using VectorDistort to warp a
black and white mask over
the source. The warped mask
is then used to grade the
original plate.
TSVSmartVectorSTMap An example of writing the

results of a SmartVector to
disk as an STMap sequence to
save processing time. This
STMap can then be used
throughout the shot as
needed, rather than
recalculating the warp.
3D ToolSets
Nuke ships with the following 3D ToolSets:
Particles
REFERENCE GUIDE
1040
|
P_DustHit Creates an adjustable dust

particle system. Double-click
the particle system to open
its Properties panel and then
click the S icon to display the
contents of the group.
The image shows the 3D

particle system rendered
through a ScanlineRender
node.
P_FogBox Creates an adjustable fog

particle system. Double-click
the particle system to open
its Properties panel and then

node.
P_SnowRain Creates an adjustable snow or

rain particle system. Double-
click the particle system to
open its Properties panel and
then click the S icon to
display the contents of the
group.

node.
REFERENCE GUIDE
1041
|
P_Sparks Creates an adjustable trail of

sparks. Double-click the
particle system to open its
Properties panel and then

node.
P_Trail Creates an adjustable comet

effect. Double-click the
particle system to open its
Properties panel and then

node.
Projection and Rig Removal
TP_RRCleanProject An example showing painting

out the rig on a single frame,
projecting the paint onto 3D
geometry and rendering it
through the moving scene
camera, and then merging the
correction back over the
source plate.
TP_RRCleanTexture An example showing

projecting a single frame
through a static camera onto
geometry, 'unprojecting' that
REFERENCE GUIDE
1042
|
frame to the geometry's UV

space to remove perspective,
painting the correction and
reapplying it to the geometry
camera. The correction is then
merged back over the source
plate.
TP_RRExistingProject An example showing

projecting an existing clean
frame through a camera onto
3D geometry, rendering it
correction back over the
source plate.
TP_RRExitsingTexture An example applying a pre-

existing clean texture to 3D
geometry, rendering it
texture back over the source
plate.
TP_RRPlanarTrack An example showing creating

a cleaned frame using a
RotoPaint node, tracking a
plane using the Roto node's
Tracking tab, and then using
the resultant CornerPin node
to track this frame into place
across the entire sequence.
Relighting
REFERENCE GUIDE
1043
Create |
TR_Relighting An example script showing

the creation of a point cloud
from a position pass, which is
then used to relight the 2D
image from 3D light
information.
Create
Creates a node tool set in the Nuke toolbar. This can be useful if you find yourself creating the same set of
nodes repeatedly. Once you have created a tool set, you can easily create the nodes that it contains by
selecting it in the ToolSets menu (rather than creating each node separately).
To create a tool set, select the nodes you want to include in the Node Graph and click ToolSets > Create.
In the Create ToolSet dialog that appears, use the ToolSets menu to select the menu where you’d like to
place your new tool set. Then give it a name in the Menu item field and click Create.
By default, your new tool set goes under the ToolSets menu, but if you’d like to create a subfolder for the
tool set, you can do that by specifying the folder name before the tool set name, separated by a forward
slash, in the Menu item field. For example, entering Roto/BasicRoto would create a subfolder called
Roto in the ToolSets menu, and place a new tool set by the name of BasicRoto in it.
Like gizmos, tool sets can be shared between artists if they are using a centralized .nuke folder. This needs
to be accessed through a NUKE_PATH environment variable that you can set up.
See also Delete.
Delete
Allows you to delete a tool set you’ve created earlier. Simply click ToolSets > Delete in the toolbar and
select the tool set you want to remove. A dialog displays asking you to confirm the removal. Click Yes.
See also Create.
REFERENCE GUIDE
1044
Other Nodes
Other nodes contain additional nodes for script and Viewer management.
Assert
You can use this node to test the validity of a Tcl expression. If the expression resolves to false, this node
raises an error.
The Assert node does not alter the input image in any way.
For more information on Tcl expressions, see Nuke's online help or select Help > Documentation > Knob
Math Expressions in Nuke.
AudioRead
You can use this node to read in an audio file and view it in the Curve Editor and Dope Sheet in order to
line up keyframes of your composition with the waveform of the sound. You can then flipbook the audio
with your footage to preview your comp with sound.
Inputs and Controls

(Scripting) Value
AudioRead Tab
file file N/A Sets the file path of the audio file you’re reading in. This
can be an uncompressed WAV or AIFF file.
REFERENCE GUIDE
1045
Backdrop |

(Scripting) Value
time range begintime 0, 0 Sets the start and end times for the audio in Nuke.
file time filebegintime 0, 0 Sets the start and end times of the audio file read in.
range These are automatically set to the values in the file, but
you can change them to trim the data used in Nuke.
reload reload N/A Click to refresh the read buffer.
ratesource ratesource file Sets the source for the sample rate. File reads the rate
from the audio file, custom lets you specify a custom
rate in the rate field.
rate rate Sets the sample rate of the audio clip.
curves
left left 0 Adjusts the left level on the current frame.
right right 0 Adjusts the right level on the current frame.
key interval keyinterval 0 Sets the key interval you want to use when creating the
curves.
generate generate N/A Click to generate the audio data as a curve that you can
use in the Curve Editor and Dope Sheet.
Backdrop
This node visually groups nodes in the Node Graph. Inserting a Backdrop node creates a box behind the
nodes. When you move the box, all the nodes that overlap the box are moved too. By inserting several
Backdrop nodes, you can group the nodes in your node tree onto boxes of different colors and titles. This
makes it easier to find a particular node in a large node tree, for example.
You can also use the Z Order control in the Properties panel to layer-up Backdrop nodes. Backdrops with
lower Z Order values appear underneath those with a higher value.
See also StickyNote.
REFERENCE GUIDE
1046
BlinkScript |
BlinkScript
The BlinkScript node runs Foundry's Blink framework enabling us to write our code once and run it on any
supported device. This is achieved through code translation, in which the Blink code is turned into specific
code for each target device. Code is generated and compiled on-the-fly, allowing you to switch between
devices at will.
BlinkScript runs a Blink "kernel" over every pixel in the output, where a Blink kernel is similar to a C++ class,
but with some special parameter types and functions. Through translation, the code in the BlinkScript node
can be turned into normal C++ or SIMD code for the CPU, or OpenCL for the GPU.
The Blink framework streamlines plug-in development workflow significantly, as you no longer have to
exit Nuke to compile your code.
Note: To use the GPU when rendering from the command line, add --gpu to your command.
See the Nuke Online Help installation pages for more information on GPU support.
You can publish kernels in Group nodes which can then be saved as gizmos, if required. Published kernels
can be encoded to protect your IP using BlinkScript's built-in kernel protection. Protected kernels are not
readable when the published node is saved to a script.
Warning: BlinkScript is very flexible, as there are no restrictions on the code you can write
within a kernel. As a result, code complied from the Kernel Source can cause Nuke to crash, so
please use caution!
Inputs and Controls

Type Name
Input src The image to which the BlinkScript kernel is applied.
REFERENCE GUIDE
1047
BlinkScript |
Control Knob (Scripting) Default Value Function

(UI)
BlinkScript Tab
Kernel File kernelSourceFile none Sets the file path of the kernel used for Load
and Save operations. BlinkScript kernels use the
.rpp file extension.
Load reloadKernelSourceFile N/A Click to load and compile the kernel specified
in the Kernel File field.
Save saveKernelFile N/A Click to save the current kernel to the location
and file name specified in the Kernel File field.
Clear clearKernelSource N/A Click to clear the Kernel Editor.
Recompile recompile N/A Click to recompile the kernel currently in the

Kernel Editor. Any compile errors present are
displayed on screen.
Kernel Source
Kernel kernelSource SaturationKernel Enter the kernel you intend to compile or edit
Editor the default SaturationKernel to suit your
purposes.
Click the Recompile button to see the result of

any changes.
Kernel Parameters Tab
Not available when:
the GPU.
REFERENCE GUIDE
1048
BlinkScript |

(UI)

dropdown.

requires you to restart Nuke before
the change takes effect.
Use GPU if useGPUIfAvailable enabled When enabled, rendering occurs on the Local

the GPU whenever the script is
opened on a machine that does have
a GPU available.
You should also select this if you wish
to render from the command line
with the --gpu option.

Vectorize vectorize enabled When enabled, use SIMD instructions on the

on CPU CPU where possible.
Note: This is currently an

experimental feature and
vectorization might fail for some
valid Blink kernels.
Saturation SaturationKernel_ 1.2 Adjusts the result of the SaturationKernel by
REFERENCE GUIDE
1049
BurnIn |

(UI)
Saturation this factor.
Note: Saturation is the only control

exposed by the default
SaturationKernel. For more
information on writing kernels and
some examples, see Help >
Documentation.
Publish publishButton N/A Click to create a Group node containing a copy

of the current node. The Group can then be
saved as a gizmo.
The kernel parameters and GPU controls are

exposed as user controls in the published
version of the node.
Protect protectKernel disabled When enabled, the kernel in the published

kernel WhenPublishing group is encoded and is not readable when the
published node is written to a script.
format format Dependent on Sets the output format when Specify output
Project format is enabled, but otherwise defaults to
Settings the input format size.
Specify specifiedFormat disabled When enabled, allows you to set the output
output format using the format control.
format
Enabling this control also exposes the format
dropdown in the published Group node.
BurnIn
The BurnIn node allows you to add text to certain areas of your footage quickly, without relying on the
Text node.
REFERENCE GUIDE
1050
BurnIn |
Inputs and Controls

Type Name
Input 1 The image to which the burn-in is applied.

(UI)
BurnIn Tab
color burnIn_color 1 Sets the color of the burn-in text.
opacity burnIn_opacity 1 Sets the opacity of the burn-in text.
x- burnIn_xpadding 30 Controls the position of the burn-in fields

padding on the x and y axes, in pixels.
y- burnIn_yPadding 30
padding
font font family Utopia Sets the burn-in font and style.
font style Regular
text burnIn_textScale 0.5 Sets the size of the burn-in text.

scale
REFERENCE GUIDE
1051
DiskCache |

(UI)
top left burnIn_topLeft hiero/clip Determines the metadata used to create

the burn-in text in the six positions listed.
top burnIn_topMiddle hiero/sequence
middle Setting a dropdown to none removes
burn-in from that position.
top burnIn_topRight hiero/shot
right
bottom burnIn_bottomLeft hiero/track

left
bottom burnIn_ hiero/project

middle bottomMiddle
bottom burnIn_ hiero/sequence/timecode

right bottomRight
Background Tab
enable burnIn_ disabled When enabled, the burn-in text has a

backgroundEnable background determined by the color,
opacity, and border controls.
color burnIn_ 0,0,0,1 Sets the color of the burn-in background.

backgroundColor
opacity burnIn_ 1 Sets the opacity of the burn-in background.

backgroundOpacity
x- burnIn_ 10 Controls the width and height of the burn-

border backgroundXBorder in background, in pixels.
y- burnIn_ 5
border backgroundYBorder
DiskCache
DiskCache caches to disk scanlines from its input as they are requested by its output. It does this at full 32-
bit floating point. This allows Nuke to reference the cached data instead of constantly recalculating the
REFERENCE GUIDE
1052
DiskCache |
output of the preceding nodes.
By default, DiskCache caches the selected channels of the current frame at the current zoom level.
DiskCache can be useful, for example, if:

• you are working on a large, complex node tree. Using the DiskCache node, you can break the node tree
into smaller sections and cache any branches that you are no longer working on.
• you are reading in images from a network. If you insert a DiskCache node after a Read node, the image
will be cached locally and displayed faster.
• you are painting or rotoscoping. If you insert a DiskCache node before a RotoPaint node, flipping frames
becomes faster.
The cached images are saved in the same directory as the images the Nuke Viewer caches automatically.
You can set the location and size of this directory in the Preferences. A large, fast local disk is DiskCache's
best friend.
If you make a change in the nodes upstream, the affected cached images are discarded and automatically
recalculated.
Inputs and Controls

Type Name
Input unnamed The node tree to cache locally.

Value
DiskCache Tab
channels channels all Sets the channels to cache from the current frame at
the current zoom level.
Precache Precache n/a Click to force all lines in the frame to be calculated
and cached.
REFERENCE GUIDE
1053
Dot |

Value

aborts.
the render aborts.

Dot
You can use the Dot node to make bends in the connecting arrows between other nodes. This can help
you keep your script easy to read.
The Dot node does not alter the input image in any way.
You can also create this node by pressing . (period) on the Node Graph.
Inputs and Controls

Type Name
Input unnamed The image sequence after which you want to add a bend in the
connecting arrow.
REFERENCE GUIDE
1054
Group |
Group
You can use the Group node to nest multiple nodes inside a single node. The original nodes are replaced
with the Group node. When you create a Group node, its internal structure is shown in a separate Node
Graph that opens.
You can also create this node by pressing Ctrl/Cmd+G on the Node Graph.
See also Precomp.
Inputs and Controls

Type Name
Input unnamed The image sequence to which you want to apply the nodes nested
inside the Group node.

(Scripting) Value
Group Tab (this includes the controls common to all nodes plus the following)
N/A N/A Shows the internal structure of the Group node in a

separate Node Graph. This can be useful if you want to
see what the Group does or ungroup the nodes nested
inside it.
export as N/A N/A Exports the group of nodes as a gizmo that can be
gizmo... reused by other artists. Gizmos are saved in a separate
.gizmo file in your Nuke plug-in folder. Nuke scripts can
use gizmos just like any other node type. Saved scripts
only contain the name and control settings for the
gizmo; the definition is in the gizmo file and it is read at
the same time the script is loaded into Nuke. Thus, you
can alter the implementation of the gizmo and change
all the scripts that are using it.
REFERENCE GUIDE
1055
Input |

(Scripting) Value
lock all lock_ disabled When enabled, the connections between the grouped
connections connections nodes are locked so that they cannot be accidentally
disconnected.
Input
Adds an input arrow to a Group node. This allows you to add multiple inputs to your Group or gizmo (for
example a main input and a mask input). By default, the inputs are numbered, but if you want to name
them, you can do so by renaming the Input nodes.
Note that you should always connect the main input first and any optional inputs after.
See also Group and Output.
LiveGroup
LiveGroup nodes can be used in conjunction with LiveInput nodes so that multiple artists can work on
different parts of the same shot as separate scripts, without the need for rendering. See LiveInput for more
information.
LiveGroups are a hybrid of the functionality offered by Precomps, Groups, and Gizmos combining all the
functionality that they lack individually. Just like Precomps, LiveGroups can store independent .nk files,
allowing you to save a subset of the node tree as a separate .nk script, render the output of this saved
script, and read the rendered output back into the master comp as a single image input.
You can also use LiveGroups like Group nodes to nest multiple nodes inside a single node. The original
nodes are replaced with the LiveGroup node. When you create a LiveGroup node, its internal structure is
shown in a separate Node Graph tab.
See also Precomp and Group.
REFERENCE GUIDE
1056
LiveGroup |
Inputs and Controls

Type Name
Input LiveInput1 The image sequence to which you want to apply the nodes or script
inside the LiveGroup node.

(Scripting) Value
LiveGroup Tab
Publish publish N/A Click to change the LiveGroup to the 'Published' state
and save the current LiveGroup as a .nk script to the
location specified in the file control.
Publishing does not add the contents of the LiveGroup

to the current script, instead nodes within the LiveGroup
are read from the external script.
Make Editable make_ N/A Click to change the LiveGroup to the editable state.
editable
Make Editable adds the contents of the LiveGroup to
the current script when the script is saved.
file file none Sets the location that the contents of the LiveGroup are
read from or written to, depending on whether you're
publishing or editing the group.
Reload reload_script n/a Click to recalculate the .nk script specified in the file
control to create the output.
Open open n/a Click to open the .nk script specified in the file control
in a new Nuke session.
Render render n/a Click to render the .nk script specified in the file control
and output the results from the LiveGroup node.
read file for reading disabled When enabled, the LiveGroup node reads in the
output rendered image rather than calculating the output of the
script.
REFERENCE GUIDE
1057
LiveInput |

(Scripting) Value
When disabled, Nuke calculates the output of the

LiveGroup script.
missing on_error error Sets the behavior when there is an error opening the
frames output file:
• error - display an error message on any missing
frames.
• black - replace any missing frames with black.
• checkerboard - replace any missing frames with a
checkerboard image.
• read input - display the result of the input tree rather
than the rendered file on any missing frames.
Reload reload_write n/a Click to re-read the image written to disk.
advanced
output node output none Sets the name of the node to use instead of the Output
node in the LiveGroup.
You can use this control to override what is set as the

Output node in the LiveGroup script.
enable useOutput disabled When enabled, the node specified in the output node
control is used instead of the Output node in the
LiveGroup.
LiveInput
The LiveInput node is a type of Input node. Adding a LiveInput to a container type node, such as a Group,
Precomp, or LiveGroup adds a corresponding input connection to that container node, just like an Input
node.
LiveInput differs from Input nodes in that they can reference Nuke scripts external to the current script,
allowing you to work collaboratively with other artists in the context of a master script without the need
for rendering or reading footage from the master script.
REFERENCE GUIDE
1058
NoOp |
Inputs and Controls

(Scripting) Value
LiveInput Tab
liveGroup liveGroup N/A Sets the LiveGroup from which the input is taken.
Only LiveGroups from the script entered in the file

control can be referenced.
file file N/A Defines the location of Nuke script containing the
referenced LiveGroup.
Reload reload_script N/A Click to re-read the referenced Nuke script and
recalculate the contents of the LiveGroup.
Open open N/A Click to open the script defined in the file control in a
new Nuke session.
NoOp
NoOp passes the input image through unchanged. It does not alter the image in any way, but can be used
as an interface for controlling other nodes. To do so, right-click on the NoOp control panel, select
Manage User Knobs, and add the control you need (for example, a checkbox). Then, Ctrl+Cmd+drag this
control on top of a control in another node (for example, the disable checkbox of a Blur node). This
creates a linking Tcl expression that parents the Blur node’s disable checkbox to the checkbox in the NoOp
controls, allowing you to use the NoOp controls to disable the Blur node.
You could also use a NoOp node to change the file path of multiple Read nodes using a single control.
For more information on Tcl expressions, see Nuke's online help or select Help > Documentation > Knob
Math Expressions in Nuke.
REFERENCE GUIDE
1059
Output |
Inputs and Controls

Type Name
Input unnamed This image sequence is passed through unchanged.
This input is optional - you don’t need to connect NoOp to the node
you want it to control.
Output
Adds an output arrow to a Group node. This is necessary if the node tree that forms the Group has several
branches and thus several possible output nodes. It can also be useful if you accidentally delete an Output
node from a Group.
The output arrows are always unnamed.
See also Group and Input.
PostageStamp
You can add this node to your script to display a thumbnail render of the output at the current frame or a
frame of your choice. This node does not alter the input image in anyway.
By default, the postage stamp is always updated to match the current frame. To change this behavior, press
Shift+S to open the Preferences dialog, go to the Node Graph tab, and set postage stamp mode to
Static frame. Then, set static frame in the PostageStamp node properties to the frame you want to
display.
Note: If the frame number you use is outside the frame range for the node, it is clamped to the
first or last frame in the range.
REFERENCE GUIDE
1060
Precomp |
Precomp
The Precomp node is like a Group node, but its content is stored in an independent .nk file. This allows
you to save a subset of the node tree as a separate Nuke script, render the output of this saved script, and
read the rendered output back into the main comp as a single image input. Precomp nodes can be useful
in at least two ways:
• They can be used to reduce portions of the node tree to pre-rendered image inputs. This speeds up
render time, as Nuke only has to process the single image input instead of all the nodes that were used
to create it. Because the original nodes are saved in a separate .nk script, you also maintain access to
them and can adjust them later if necessary.
• They enable a collaborative workflow. While one artist works on the main comp, others can work on the
sections that have been exported using the Precomp node. These sections can be edited, versioned, and
managed independent of the main comp.
You can also create this node by pressing Ctrl/Cmd+Shift+P on the Node Graph.
See also Group.
Inputs and Controls

Precomp Tab
file file N/A Sets the .nk script to use as the contents
of the precomp.
Reload reload_script N/A Click to reload the .nk script specified in

the file control.
Open open N/A Click to open the .nk script specified in

the file control in a new Nuke session.
Render render N/A Click to render the .nk script specified in

the file control and output the results
from the Precomp node.
read file for output reading disabled When enabled, the Precomp node reads in
REFERENCE GUIDE
1061
Precomp |
the rendered precomp image rather than

calculating the output of the precomp
script.
When disabled, Nuke calculates the output

of the precomp script.
missing frames on_error error Sets the behavior when there is an error
opening the output file:
• error - display an error message on any
missing frames.
• black - replace any missing frames with
black.
• checkerboard - replace any missing
frames with a checkerboard image.
• read input - display the result of the
input tree rather than the rendered file
on any missing frames.
Reload reload_write N/A Click to re-read the image from disk.
advanced
output node output N/A Sets the name of the Write node to use
instead of the Output node in the
precomp.
You can use this control to override what

is set as the Output node in the precomp
script.
enable use_output disabled When enabled, the Write node specified

in the output node control is used instead
of the Output node in the precomp.
REFERENCE GUIDE
1062
Root |
Root
The Project Settings in the Compositing environment are contained within the Root node, but the node
itself is never visible in the Node Graph. When you start working on a script, you should first define the
settings for it in the Project Settings.
You can display the Project Settings by navigating to Edit > Project Settings or by pressing S in the
Node Graph.
Inputs and Controls

Value
Root Tab
name name none Allows you to set the script name

and path. This field is completed
automatically the first time you
save the script.
project directory project_directory none Sets the default location where

project files are saved. For
example, if you only specify a file
name in a Write node, the files
are rendered to this location.
Script Directory Script_directory N/A Click to set the project directory

using an expression so that if the
.nk file is relocated, the project
directory updates automatically.
comment label none Adds a comment to the script,

which is only visible in the plain
text of the .nk file in the label
field.
lock all connections lock_connections disabled When enabled, all node

connections are locked, making it
REFERENCE GUIDE
1063
Root |

Value
easier to move nodes around the

Node Graph without
disconnecting them accidentally.
frame range first_frame 1 Sets the frame range for the script,
which is used for nodes that don't
last_frame 100 include their own frame range
controls.
lock range lock_range disabled When enabled, the frame range is

locked so that adding new image
sequences does not alter the
frame range.
When disabled, adding new

image sequences alters the frame
range to encompass the new
range.
fps fps 24 Sets the project frame rate.

Adjusting this control affects the
Viewer's frame rate as well.
full size format format 2K_Super_ Sets the default format for the
35 project.
proxy mode proxy disabled When enabled, images are

rendered at either the proxy scale
or format as specified by the
proxy type control.
proxy_type scale Sets the proxy type when proxy

mode is enabled:
• format - the proxy resolution is
determined by the proxy
format control.
• scale - the proxy resolution is
determined by the proxy scale
control.
REFERENCE GUIDE
1064
Root |

Value
proxy scale proxy_scale 0.5 Sets the proxy resolution when

the proxy type control is set to
scale.
proxy format proxy_format 1K_Super_ Sets the proxy resolution when

35 the proxy type control is set to
format.
read proxy files proxySetting if larger Sets when Read nodes fallback to
the proxy resolution when proxy
mode is enabled:
• never - the proxy resolution files
are never used.
• if larger - the proxy resolution
files are used if the files are
equal to or larger than the scaled
size.
• if nearest - the files closest to
the scaled size is used.
• always - the proxy resolution
files are always used.
is live group is_live disabled When enabled, the script is saved

as a LiveGroup script, allowing you
to nest it within other scripts using
the LiveGroup node.
Color Tab
color management colorManagement Nuke Sets which colorspaces are

available to Read/Write nodes
and Viewer processLUTs in the
current script.
• Nuke - only the Nuke root
colorspaces are available.
• OCIO - the specified OCIO
config colorspaces are available
REFERENCE GUIDE
1065
Root |

Value
as well as Nuke root colorspaces.
Enable OCIO GPU path OCIOGPUSupport disabled When enabled, any Viewers using
for GPU Viewer the GPU to process data also
compute OCIO data on the GPU,
rather than the CPU.
Note: The GPU path in

OCIO is not completely
accurate, so you may see
banding or color
inaccuracy when using
OCIO on the GPU.
This control only affects the

Viewer when the Preferences
> Panels > Viewer (Comp) > use
GPU for Viewer when possible is
enabled.
OCIO config OCIO_config nuke- When color management is set to

default OCIO, sets the configuration to
use and populates the default
LUT settings appropriately.
Selecting custom allows you to

reference a custom OCIO color
file using the custom OCIO
config control.
custom OCIO config customOCIOConfigPath none When OCIO config is set to

custom, enter the file path and
name of the .config file.
default LUT settings
working space workingSpaceLUT linear Sets the working colorspace
REFERENCE GUIDE
1066
Root |

Value
within NukeHiero. Files are

converted to this colorspace when
they are read into NukeHiero. At
render time, the files are
converted from this colorspace to
the selected output colorspace.
If color management is set to

Nuke, this control is always set to
linear.
monitor monitorLUT sRGB Sets the default LUT for postage

stamps, OpenGL textures, and all
non-Viewer image and color
displays.
If your Viewer process is a simple

LUT, set this control to the same
LUT, although sRGB provides an
acceptable result is most cases.
Note: When color

management is set to
OCIO, these colorspaces
are grouped by the
Display and saved in the
Project Settings.
Other colorspace types

are grouped by the
family name.
8-bit files int8LUT sRGB Sets the default LUT used when
reading and writing image data
containing 8 bits.
16-bit files int16LUT sRGB Sets the default LUT used when
REFERENCE GUIDE
1067
Root |

Value

containing 16 bits.
log files logLUT Cineon Sets the default LUT used when
containing Cineon/DPX style log
data. These files are typically 10-
bit.
float files floatLUT linear Sets the default LUT used when
containing floating point values.
luts luts none Displays the lookup curves used to

convert between file/device
colorspaces and the linear
working colorspace used for
internal operations.
Note: SLog has been

deprecated in favor of
SLog1, but remains in
the list to ensure
backward compatibility.
+ N/A N/A Click to add a new curve to the list

of LUTs.
reset N/A N/A Click to reset any points that you

have adjusted on a curve to their
default position.
- N/A N/A Click to delete a custom curve

from the list of LUTs. You can't
delete the default LUTs that ship
with Nuke.
Views Tab
REFERENCE GUIDE
1068
Root |

Value
+ views N/A Click to add a new view to the

script.
- views N/A Click to remove a view from the

script.
Note: You cannot

remove a view if it is the
only view in the script.
up views N/A Click to move a view up or down

the views list. The view at the top
down views N/A is the first listed above the Viewer.
views views main Lists all the views currently in the

script and the color associated
with each view.
Note: The colors are

only visible in the script
when Use colors in UI?
is enabled.
hero hero_view main Sets the view that is loaded by

default when the script is loaded.
Set up views for stereo setlr N/A Click to automatically create left
and right views for stereographic
projects.
View selection uses views_button enabled When enabled, the views in the
buttons? script are displayed as buttons
above the Viewer.
When disabled, the views in the

script are hidden in a dropdown
menu.
REFERENCE GUIDE
1069
Root |

Value
Use colors in UI? views_colours disabled When enabled, the colors

specified per view in the views list
are displayed above the Viewer
and in Node Graph for clarity.
onScriptLoad onScriptLoad none These functions run when a script

is loaded.
onScriptSave onScriptSave none These functions run when a script

is saved.
onScriptClose onScriptClose none These functions run when a script

is closed.
Live Group Callbacks
onLiveGroupMakeLocal onLiveGroupMakeLocal none These functions run after a

LiveGroup node goes from the
Published state to the Local state.
onLiveGroupPrePublish onLiveGroupPrePublish none These functions run before a

LiveGroup is published. If this
callback throws an error, the
publish operation is interrupted.
onLiveGroupPublish onLiveGroupPublish none These functions run after a

LiveGroup is published.
onLiveGroupReload onLiveGroupReload none These functions run after a

LiveGroup is reloaded.
Font Tab
project font path free_type_font_path none Allows you to specify a directory

from which to read font files.
include system fonts free_type_system_fonts enabled When enabled, all the system
fonts are available in addition to
those that ship with Nuke.
REFERENCE GUIDE
1070
StickyNote |

Value
When disabled, only the fonts that

ship with Nuke are available.
Rescan font path RescanFreeTypeMappingPaths N/A Click to rescan the specified

project font path for new font
files.
StickyNote
Lets you add notes to the Node Graph. The notes can be any text or HTML mark-up. Usually, they are made
as annotations to the elements in the node tree.
See also Backdrop.
All plugins
This menu contains all nodes available in Nuke, including any unsupported nodes (such as earlier versions
of the currently supported nodes). Note that some of these nodes may be unstable or cause unexpected
behavior.
If you have any gizmos installed, they also appear in this menu.
To refresh the contents of the menu, click Update.
REFERENCE GUIDE
1071
FurnaceCore Nodes
FurnaceCore nodes contain the most popular Furnace plug-ins incorporated in NukeX and Nuke Studio.
F_Align
F_Align takes two sequences that were shot of the same scene and lines them up spatially. It uses Global
Motion Estimation (GME) to calculate a four-corner pin so that each frame in one shot (the source input)
will be aligned with the corresponding frame in a second shot (the reference input). The result is the
source image which has been repositioned to line up with the reference image.
Inputs and Controls

Type Name
Input Ref The reference image used to align the source image.
Src The source image to align.

(Scripting) Value
F_Align Tab
Analyse analyse N/A Click to begin analysis of the input clips and calculate a
corner pin. Interrupting the analysis will not delete the
corner pin keys that have already been calculated.
Render renderOn enabled When enabled, this toggle causes the effect to update
During the time line and render a freshly analyzed frame in the
Analysis Viewer so you can see the progress of the effect.
REFERENCE GUIDE
1072
|

(Scripting) Value
Note: Disabling this control may speed up the

general analysis.
Clear Analysis clear N/A Click to delete all key frames from the corner pin
controls, allowing you to force a re-analysis if you feel
the need to.
Analysis range Source Clip This controls the range of frames any analysis runs over:
Range Range • Specified Range - reads the Analysis Start and
Analysis Stop fields for the range of frames to analyze.
• Source Clip Range - automatically determines the
range of frames to analyze from the length of the clip
attached to the Src input.
• Current Frame - the analysis occurs only on the
current frame. This is useful for correcting any errors
that may have occurred while analyzing the entire clip.
Analysis Start start 0 The first frame analyzed if AnalysisRange is set to

SpecifiedRange.
Analysis Stop stop 100 The last frame analyzed if AnalysisRange is set to
SpecifiedRange.
Scale scale disabled Indicates whether the calculated corner pin can include a
scaling factor.
Rotate rotate enabled Indicates whether the calculated corner pin can include
rotations.
Translate translate enabled Indicates whether the calculated corner pin can include
translations on the x and y axes.
Perspective perspective disabled Indicates whether the calculated corner pin can include
perspective transforms.
Analysis Region
Analysis regionBL N/A The region analyzed to calculate the four-corner pin.
Region BL This is especially useful when doing any form of frame
REFERENCE GUIDE
1073
|

(Scripting) Value
locking, in which case, go to the lock frame, look at the

reference clip and position the box over the area you
want locked.
• Analysis Region BL - controls the position of the
Analysis regionTR N/A
bottom left corner of the analysis region.
Region TR
• Analysis Region TR - controls the position of the top
right corner of the analysis region.
Advanced
Accuracy accuracy 0.9 Controls the time/accuracy trade off. Higher values slow
the analysis, but can produce better result.
Filtering filtering Medium Sets the filtering quality:

• Low - low quality but quick to render.
• Medium - uses a bilinear filter. This gives good results
and is quicker to render than High quality filtering.
• High - uses a sinc filter to interpolate pixels giving a
sharper repair. This gives the best results but takes
longer to process.
Invert invert disabled Enable this control to use the inverse of the calculated
four-corner pin during render.
Advanced > Four Corner Pin
Bottom Left pinBL N/A The bottom left corner pin calculated during the analysis
pass.
Bottom Right pinBR N/A The bottom right corner pin calculated during the
analysis pass.
Top Left pinTL N/A The top left corner pin calculated during the analysis
pass.
Top Right pinTR N/A The top right corner pin calculated during the analysis
pass.
About about N/A Displays a dialog containing information about this

node.
REFERENCE GUIDE
1074
F_DeFlicker2 |
F_DeFlicker2
F_Deflicker2 is used to remove flicker - particularly flicker that is localized and dependent on the
geometry of the scene (that is, flicker that is not present across the whole of the image), such as that
caused by an un-synchronized fluorescent light in a shot.
Note: F_DeFlicker2 looks at input frames outside the current frame when performing
calculations, and therefore can be a computationally expensive node. As such, using more than
two instances of F_DeFlicker2 in a node tree will dramatically increase render times. It is strongly
advised therefore, that you render each instance out separately.
Inputs and Controls

Type Name
Input Src The sequence requiring deflicker.

(Scripting) Value
F_DeFlicker2 Tab
DeFlicker amount 0.3 Reduces flicker without removing it entirely; lower

Amount values leave more flicker behind.
Block Size blockSize 9.6 Defines the width and height of the control block (in
pixels) centered around a particular pixel analysed by the
deflicker algorithm.
Note: The default value avoids the loss of

detail and noisy motion fields associated with
higher and lower BlockSize, and rarely needs
adjusting.
REFERENCE GUIDE
1075
F_ReGrain |

(Scripting) Value
Use Motion useMotion enabled Enables a second deflicker pass using motion-
compensated frames. This can improve results in areas
where there is fast motion, where the initial deflicker
pass may have introduced blurring.
Vector Detail vectorDetail 0.2 Set the density of the motion vectors used when
UseMotion is turned on. The maximum value of 1
generates one vector per pixel, producing the most
accurate vectors, but taking longer to render. The default
value of 0.2 generates a vector at every fifth pixel.
Analysis range 2 The number of frames searched each side of the current
Range frame when calculating the flicker. Higher values may
give better results, but can also bring in erroneous
information and take longer to process.

node.
F_ReGrain
F_ReGrain is used to add grain to a sequence. It is designed to sample an area of grain from one image
and then to generate unlimited amounts of this grain with exactly the same statistics as the original. This
new grain can then be applied to another image.
See also Grain and ScannedGrain.
Inputs and Controls

Type Name
Input Grain The image from which grain is sampled. When a Grain input is
supplied, F_ReGrain automatically switches to using grain sampled
from this input. However, the supplied grain stocks are still available.
REFERENCE GUIDE
1076
F_ReGrain |

Type Name
Src The image to which the grain is added.
F_ReGrain Tab
Grain Type grainType Preset Stock Selects whether the grain is sampled from the Grain
input or from a set of standard stocks.
• Preset Stock - grain characteristics are sampled
from the film stock specified in the Preset Stock
field.
• From Grain Clip - samples and reconstructs the
grain characteristics from the Grain input.
Preset Stock presetStock FUJIF250under The film stock that grain characteristics are sampled
2K from when GrainType is set to PresetStock.
Grain amount 1 Adjusts the brightness of the grain. Setting this to 0

Amount adds no grain.
Grain Size size 1 Adjusts the size of the grain granules. The larger the
value, the bigger and softer the granules.
Output output Result Sets whether to render the result or a test image.
• Result - shows the Src input with the grain applied.
• Grain Plate - shows a test image with the grain
applied. This test image is composed from a
section of the input image surrounded by a uniform
solid color sampled from the image with the grain
applied. If the inner area is indistinguishable from
the outer area, then you have a good grain sample.
Analyse analyse N/A Click to begin analysis of the source. Press this button
if the input clip from which the grain was analyzed
has changed, but you do not want to move the
analysis region to trigger re-analysis.
REFERENCE GUIDE
1077
F_ReGrain |
Note: A warning displays when the input

clip changes.
Analysis Region
Analysis regionBL N/A A selection box that marks the region of image used
Region BL to analyze the grain when GrainType is set to From
Grain Clip. This part of the frame must contain no
image detail, only grain.
Analysis regionTR N/A • Analysis Region BL - controls the position of the
Region TR bottom left corner of the analysis region.
• Analysis Region TR - controls the position of the
top right corner of the analysis region.
Analysis frame 1 Sets the frame to sample the grain from.

Frame
Grain Colour grainColourSpace sRGB This tells F_ReGrain what color space the grain
Space sample clip was in when the grain originated. Setting
this correctly ensures that the grain is not
exaggerated by any color space conversions prior to
sampling.
• Cineon
• sRGB
• Linear
Advanced
Process Red processRed enabled Disable this if you do not want to process the red
channel.
Red Amount redAmount 1 Sets the brightness of the grain in the red channel.
Red Size redSize 1 Adjusts the size of the grain granules in the red
channel.
Process processGreen enabled Disable this if you do not want to process the green
Green channel.
Green greenAmount 1 Sets the brightness of the grain in the green channel.
REFERENCE GUIDE
1078
F_ReGrain |
Amount
Green Size greenSize 1 Adjusts the size of the grain granules in the green
channel.
Process Blue processBlue enabled Disable this if you do not want to process the blue
channel.
Blue Amount blueAmount 1 Sets the brightness of the grain in the blue channel.
Blue Size blueSize 1 Adjusts the size of the grain granules in the blue
channel.
Advanced > Grain Response
Apply Grain srcColourSpace Grain Colour Sets what color space the grain sample is re-applied
In Space to the image:
• Cineon / sRGB / Linear - the grain sample is
applied in the specified space.
• Grain Colour Space - the grain sample is applied in
the color space set in the Analysis Range > Grain
Colour Space field.
Low Gain lowGain 1 Adjusts the gain of the grain in the lowlights.
Mid Gain midGain 1 Adjusts the gain of the grain in the midtones.
High Gain highGain 1 Adjusts the gain of the grain in the highlights.
Use Sampled useResponse disabled Enable this control to scale the brightness of the
Response grain as a function of the luminance of the Grain
image.
Sampled responseMix 1 Decreasing the Sampled Response Mix control

Response reduces the effect of the response curves until, at 0,
Mix they have no effect on the output.
Note: This control is only available if Use

Sampled Response is enabled.
Sample Grain sample N/A Click to update the response curves from the current
REFERENCE GUIDE
1079
F_RigRemoval |
Response frame. Multiple clicks accumulate the grain response

rather than resetting every time.

Reset Grain reset N/A Click to reset the grain curves to their default (flat)
Response response.

Draw drawResponse disabled Overlay the response curves on the bottom left
Response corner of the viewer.


node.
F_RigRemoval
F_RigRemoval eliminates unwanted objects, or rigs, from image sequences without the need for accurate
rotoscoping or keying to produce a clean plate. The rig removal algorithm works by estimating the
background motion between successive frames, ignoring the foreground object, and then using the
motion information to look forward and backward in the sequence in order to find the correct piece of
background to fill in the missing region.
See also RotoPaint and Project3D.
REFERENCE GUIDE
1080
F_RigRemoval |
Inputs and Controls

Type Name
Input RigMask An optional mask to designate the rig area to remove.
Src The source image containing the unwanted object or rig. The source
may contain an alpha channel to define the rig area.

(Scripting) Value
F_RigRemoval Tab
Rig Region rigRegion Box Defines the area to repair:

• Box - repair the area inside a rectangular box,
controlled by the Rig Region Box controls or the on-
screen box.
• Src Alpha - repair the region defined by the alpha of
the Src input.
• Src Inverted Alpha - repair the region defined by
the inverted alpha of the Src input.
• RigMask Luminance - repair the region defined by
the luminance of the Rig Mask input.
• RigMask Inverted Luminance - repair the region
defined by the inverted luminance of the Rig Mask
input.
• RigMask Alpha - repair the region defined by the
alpha of the Rig Mask input.
• RigMask Inverted Alpha - repair the region defined
by the inverted alpha of the Rig Mask input.
Frames framesSearched Forward and Select whether to search forwards, backwards, or in

Searched Backward both directions to find missing data.
• Forward and Backward - searches before and after
the current frame.
REFERENCE GUIDE
1081
F_RigRemoval |

(Scripting) Value
• Forward - searches frames after the current frame.

• Backward - searches frames before the current
frame.
Frame Range frameRange 4 Sets the number of frames the algorithm looks
forwards and backwards in the sequence to find the
missing data. If you are getting red pixels, then
increase this value.
Frames Used framesUsed Half of If the Frame Range is set to a large number, the
in Range Frames rendering time can be prohibitive. Frames Used in
Range may speed up the repair by not using every
frame to fill the foreground region, effectively
skipping frames. However, this may reduce the quality
of the result.
• All Frames - use every frame in the specified frame
range to construct the repair.
• Half of Frames - use every other frame in the
specified frame range to construct the repair.
• Quarter of Frames - use every fourth frame in the
• 10% of Frames - use every tenth frame in the
• Max 25 Frames - use no more than 25 frames from
the specified frame range to construct the repair. This
option can be useful if Frame Range has been set to
a very large number.
Max Rig maxRigMove 30 To avoid perspective changes, F_RigRemoval searches

Movement for the missing data inside an area immediately around
the RigRegion. Max Rig Movement defines the width
of this area (in pixels). Fast movement in the Src input
may require a higher value than slow movement.
Rig Region Box
REFERENCE GUIDE
1082
F_RigRemoval |

(Scripting) Value
Rig Region regionBL N/A The rectangular area used to define the repair when
BL Rig Region is set to Box.
• Rig Region BL - controls the position of the bottom
Rig Region regionTR N/A left corner of the rig region.
TL • Rig Region TR - controls the position of the top
right corner of the rig region.
Advanced
Filtering filtering Medium Sets the filtering quality.

• Medium - uses a bilinear filter. This gives good
results and is quicker to render than high filtering.
longer to process.
Luminance lumCorrect disabled Enable this to correct for luminance changes from
Correct information taken from other frames. This is
particularly important if the lighting changes
throughout the sequence.
Perspective perspCorrect disabled Enable this to correct for minor perspective changes.
Correct
Overlap overlapCorrect 1 The repair is built up using slices of information from

Correct other frames in the sequence which are then
overlapped and blended to give a more natural
looking repair. This value controls how much the
regions overlap. Increasing this value too much
degrades image sharpness.
Repair Fail failOpacity 0.5 Sets the level of transparency of the red pixels used to
Marker show where the repair has failed.
Opacity
Preserve preserveAlpha disabled Enable this to preserve the original alpha channel. By
Alpha default, the alpha channel is set to white where the
REFERENCE GUIDE
1083
F_Steadiness |

(Scripting) Value
repair has failed and black everywhere else.

node.
F_Steadiness
F_Steadiness uses Global Motion Estimation (GME) to calculate a four-corner pin, so that camera motion
within a single shot can be smoothed out over a range of frames or removed by locking to a specific
frame.
Inputs and Controls

Type Name
Input Src The source clip to stabilize.

(Scripting) Value
F_Steadiness Tab
Analyse analyse N/A Click to begin an analysis of the input clip and calculate
a corner pin. Interrupting the analysis will not delete the
corner pin keys that have already been calculated.
Render renderOn enabled When enabled, this toggle causes the effect to update
During the time line and render a freshly analyzed frame in the
Analysis Viewer so you can see the progress of the effect.
REFERENCE GUIDE
1084
F_Steadiness |

(Scripting) Value
Note: Disabling this control may speed up the

general analysis.
Clear Analysis clear N/A Click to delete all key frames from the corner pin
controls, allowing you to force a re-analysis if you feel
the need to.
Analysis range Source Clip This controls the range of frames any analysis runs over:
Range Range • Specified Range - reads the Analysis Start and
Analysis Stop fields for the range of frames to analyze.
• Source Clip Range - automatically determines the
range of frames to analyze from the length of the clip
attached to the Src input.
Analysis Start start 0 The first frame analyzed if AnalysisRange is set to

SpecifiedRange.
Analysis Stop stop 100 The last frame analyzed if AnalysisRange is set to
SpecifiedRange.
Mode mode Smooth This controls whether F_Steadiness is smoothing the

shot while keeping the overall camera motion, or
locking the shot to a single frame to completely remove
camera motion. It can be set to:
• Incremental Lock - calculates the pin that takes each
frame to the lock frame. This calculates the pin by
working from the lock frame out to each frame,
calculating the GME between each frame incrementally
and accumulating it to create the corner pin.
• Absolute Lock - this also calculates a pin that takes
each frame to the lock frame. However, it does so by
doing GME directly from the frame in question directly
to the lock frame.
• Smooth - in which case, the shot is smoothed for a
range of frames described by the Smoothing control.
Use Smooth to keep the overall camera motion, but to
REFERENCE GUIDE
1085
F_Steadiness |

(Scripting) Value
smooth out sharp bumps and kicks.
Scale scale disabled Indicates whether the calculated corner pin can include a
scaling factor.
Rotate rotate enabled Indicates whether the calculated corner pin can include
rotations.
Translate translate enabled Indicates whether the calculated corner pin can include
translations on the x and y axes.
Perspective perspective disabled Indicates whether the calculated corner pin can include
perspective transforms.
Analysis Region
Analysis regionBL N/A This is the region analyzed to calculate the four-corner
Region BL pin. This is especially useful when doing any form of
frame locking, in which case, go to the lock frame, look
at the reference clip and position the box over the area
Analysis regionTR N/A you want locked.
Region TR • Analysis Region BL - controls the position of the
bottom left corner of the analysis region.
• Analysis Region TR - controls the position of the top
right corner of the analysis region.
Advanced
Smoothing smooth 10 Controls the range of frames to average motion over

when Mode is set to Smooth.
Lock Frame lockFrame 0 Controls the frame locked to when Mode is set to either
REFERENCE GUIDE
1086
F_Steadiness |

(Scripting) Value
of the lock modes.
Note: Frame 0 for F_Steadiness is frame 1 in

NukeX. Therefore, if you are looking at frame 3
in the Viewer and want to use that frame as the
lock frame, you need to enter 2 as the
LockFrame value.
Accuracy accuracy 0.6 Controls the time/accuracy trade off. The higher this is,
the slower the analysis, but you have a better likelihood
of a good result.v
Filtering filtering Medium Sets the filtering quality.

• Medium - uses a bilinear filter. This gives good results
and is quicker to render than High filtering.
longer to process.
Invert invert disabled When enabled, the inverse of the calculated four-corner
pin is used during render. This works best with the lock
modes, and can be used to track static locked-off plates
back into a shot.
Auto Scale autoScale 1 Automatically fills in the black gaps at the edges of the
Src by scaling the output image up. A value of 1 scales
the image up until no black is visible, whereas a value of
0 disables scaling and leaves the black edges untouched.
Advanced > Four Corner Pin
Bottom Left pinBL N/A The bottom left corner pin calculated during the analysis
pass.
Bottom Right pinBR N/A The bottom right corner pin calculated during the
analysis pass.
REFERENCE GUIDE
1087
F_WireRemoval |

(Scripting) Value
Top Left pinTL N/A The top left corner pin calculated during the analysis
pass.
Top Right pinTR The top right corner pin calculated during the analysis
pass.

node.
F_WireRemoval
F_WireRemoval is particularly good at removing wires over heavily motion blurred backgrounds or wires
over smoke, dust, or clouds. It can be used to remove each wire in a sequence or to quickly create a clean
plate which can then be tracked into place.
F_WireRemoval also incorporates a tracker which automatically tracks a moving wire through a clip. This
tracker has its own control panel, which will float inside the Viewer if you have checked Show On Screen
Controls in the F_WireRemoval controls.
See also RotoPaint.
Inputs and Controls

Type Name
Input CleanPlate An optional input to allow you to supply a clean plate. This is used by
the CleanPlate repair mode which will warp the clean plate onto the
current frame and use the warped image to reconstruct the
background behind the wire.
Source The clip containing the wire to be removed.
REFERENCE GUIDE
1088
F_WireRemoval |

Value
F_WireRemoval Tab
setUserKeyFrame N/A Create user key frame - creates a user key frame.
deleteUserKeyFrame N/A Delete user key frame - deletes a user key frame.
snapToWire N/A Snap to wire - finds the edges of the wire and
snaps the edges of the region onto them.
trackBwd N/A Track backwards - plays backwards through the

sequence tracking from frame to frame.
stepBwd N/A Step backward - tracks backwards one frame.
stepFwd N/A Step forward - tracks forward one frame.
trackFwd N/A Track forwards - plays forwards through the

sequence tracking from frame to frame.
smartTrack N/A Smart track - tracks from beginning to end of

frame range in an intelligent order.
deleteTrackKeysBwd N/A Delete track key frames backwards - deletes track

key frames backwards through the sequence until
either a user key frame or the beginning of the
deleteTrackKeyStepBwd N/A Delete track key frame and step backward -

deletes a track key frame and steps backwards
one frame.
REFERENCE GUIDE
1089
F_WireRemoval |

Value
deleteTrackKey N/A Delete track key frame - delete the current track
key frame.
deleteTrackKeyStepFwd N/A Delete track key frame and step forwards -

deletes a track key frame and steps forwards one
frame.
deleteTrackKeysFwd N/A Delete track key frames forwards - deletes track

key frames forwards through the sequence until
either a user key frame or the end of the
deleteAllTrackKeys N/A Delete all track key frames - deletes all track key
frames from the sequence.
deleteAll N/A Delete all track and user key frames - deletes
both track key frames and user key frames.
Type wireType Three Points Controls the number of points on the on-screen
wire tool. Choose the number of points needed
to describe the wire you wish to remove.
• Two Points - choose this if your wire is
straight.
• Three Points - choose this if your wire is a
simple curve.
• Five Points - choose this if your wire has an s-
shaped curve.
On-Screen onScreenWire Show Sets the display mode for the on-screen wire
Wire tool.
• Show - shows both points and lines.
• Hide - hides both points and lines.
• Points only - only shows the points.
Show On showUI disabled Shows or hides the tracker panel in the Viewer.
Screen
REFERENCE GUIDE
1090
F_WireRemoval |

Value
Controls
Output output Source Sets the output mode for F_WireRemoval.

• Source - output the untouched source image.
Use this output mode to position the on-screen
wire tool over the wire you wish to remove.
• Repair - output the repaired source image,
with the wire removed from under the on-
screen tool.
• Wire Matte - renders a matte for the wire. This
may be useful if the wire has been tracked but
cannot be repaired using F_WireRemoval and
other techniques have to be used.
• Repair Matted - output the repaired source
image and a matte in the alpha channel. If you
want, you can manually adjust your image
further using the matte.
Track Range range Source Clip Sets the range of frames to track the wire over.
Range • Specified Range - use the Track Start and
Track End controls to specify the range over
which to track the wire.
• Source Clip Range - track the wire over the
entire range of the Source clip.
Track Start start 0 Specifies the start of the tracking range when
Track Range is set to SpecifiedRange.
Track End end 100 Specifies the end of the tracking range when
Track Range is set to SpecifiedRange.
Repair repairMethod Spatial Sets the algorithm used to remove the wire from
under the grain:
• Spatial - this method uses a slope dependent
filter that interpolates across the wire at the
most likely angle, given the image behind the
wire. It uses information from the current frame
REFERENCE GUIDE
1091
F_WireRemoval |

Value
only.
• Temporal With Static Scene - this method
uses LME to align frames from before and after
onto the current frame. This is useful for
sequences where the wire is moving and where
the motion in the rest of the scene is non-
uniform.
• Temporal With Moving Scene - also aligns
frames from before and after onto the current
frame, but uses GME. This is useful for
sequences where the wire is moving and the
motion in the rest of the scene is fairly uniform.
• Clean Plate - choose this method if you have a
clean plate you wish to use for the repair, or if
F_WireRemoval does not do a good job of
removing the wire from each frame.
Filter Size filterSize 5 If the wire you are trying to remove has details
within it (for example, a steel wire in which the
twisted threads are reflecting light), then the
algorithm may leave these alone, thinking that
they are grain. In this situation, you can decrease
the filter size.
Temporal tempOffset 1 Sets the time offset of the additional frames to

Offset use for the Temporal With Static Scene or
Temporal With Moving Scene methods. This
determines which two frames before and after
the current frame are used to fill in the
background behind the wire.
Luminance lumCorrect disabled Enable this where there are global luminance
Correct shifts between one frame of the sequence and
the next, or between a frame of the sequence
and a clean plate you are using for the repair.
REFERENCE GUIDE
1092
F_WireRemoval |

Value
Note: The Spatial repair mode does

not benefit from Luminance
Correction.
Lum Block lumBlockSize 31.12 Altering the Lum Block Size could produce a
Size better result if Luminance Correction is not
performing as expected.
Points
Point 1 point1 N/A The position of the start point on the wire.
Point 2 point2 N/A The position of the point on the wire between
the start point and the mid point.
This is only active if Type is set to Five Points.
Point 3 point3 N/A The position of the mid point on the wire.
Point 4 point4 N/A The position of the point on the wire between
the mid point and the end point.
This is only active if Type is set to Five Points.
Point 5 point5 N/A The position of the end point on the wire.
Start Width startWidth 15.56 The width of the wire at Point 1 of the on-screen
wire tool.
End Width endWidth 15.56 The width of the wire at Point 5 of the on-screen
wire tool. This allows you to make your repair
region wider at one end than the other, for
example, where there is motion blur on the wire.
Overall overallWidth 15.56 Alter the width of the repair region along its
Width entire length, and for all key frames.
About about N/A Displays a dialog containing information about

this node.
REFERENCE GUIDE
1093

Nuke12.2v3 ReferenceGuide PDF

Uploaded by

Copyright:

Available Formats

Nuke12.2v3 ReferenceGuide PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Nuke12.2v3 ReferenceGuide PDF

Uploaded by

Copyright:

Available Formats

REFERENCE GUIDE

The EULA is available here: https://www.foundry.com/eula

Nuke™ is a trademark of The Foundry Visionmongers Ltd.

Digital Domain ® is a registered trademark of Digital Domain, Inc.

Primatte™ is a trademark of IMAGICA Corp.

Primatte™ patent is held by IMAGICA Corp.

Linux ® is a registered trademark of Linus Torvalds.

Windows ® is the registered trademark of Microsoft Corporation.

Houdini ® is a registered trademark of Side Effects Software, Inc.

Boujou is a trademark of 2d3 Ltd.

3D-Equalizer is a trademark of Science.D.Visions.

RenderMan ® is a registered trademark of Pixar.

Cineon™ is a trademark of Eastman Kodak Company.

The Foundry Visionmongers Ltd.

Rev: Wednesday, September 16, 2020

Draw Nodes 101

Time Nodes 200

Add 3:2 pulldown 200

Remove 3:2 pulldown 224

Channel Nodes 250

Color Nodes 264

Filter Nodes 378

Erode (blur) 418

Erode (fast) 419

Keyer Nodes 474

Merge Nodes 521

Transform Nodes 551

Particles Nodes 897

Deep Nodes 991

Views Nodes 1023

Split and Join 1030

Metadata Nodes 1032

ToolSets Nodes 1039

Other Nodes 1045

All plugins 1071

FurnaceCore Nodes 1072

Viewing the Context-Sensitive Online Help

Selecting the Help Source

To select the help source:

1. Press Shift+S on the Node Graph to open the Preferences dialog.

1. Directories contained in the NUKE_PATH environment variable.

• custom - Use your own custom help system.

5. If you set documentation source to foundry:

Using the Help Menu

Contacting Customer Support

Inputs and Controls

size boxsize 64 Sets the size of the checkerboard boxes in pixels.

Control Knob Default Value Function

• color 2 - the box at 1,1 and then every alternate box

color 3 color3 N/A

Inputs and Controls

Control (UI) Knob Default Function

Inputs and Controls

If you set this to something other than all or none, you

Control Knob Default Value Function

rotate rotate 0 Sets the amount of rotation to apply to color position in

When disabled, use the area controls to specify the color

Note: Click wh to toggle between xyrt and

Inputs and Controls