Supported File and Camera Formats

Notes

When importing and exporting files, remember the following:

When you import images with a Read node (Image > Read), Nuke analyzes the contents of the file to determine the format. The file name extension is not used to determine file format, which allows flexibility with naming conventions in a production environment.

Regardless of format, Nuke converts all imported image sequences to its native 32-bit linear RGB colorspace.

When you render new images from Nuke (Image > Write), you can use a file name extension to specify format.

To import and export geometry objects from Alembic, FBX, or OBJ files, use the ReadGeo node (3D > Geometry > ReadGeo). To write them out again, use the WriteGeo node (3D > Geometry > WriteGeo). To import cameras or transforms from Alembic or FBX files, use the Camera node (3D > Camera). To import lights from an FBX file, use the Light node (3D > Lights > Light).

To import deep images (either in DTEX or scanline OpenEXR format), use a DeepRead node (Deep > DeepRead). To export deep images (in scanline OpenEXR format), use a DeepWrite node (Deep > DeepWrite).

Supported File Formats

The following table lists the supported file formats. The extensions listed under Extension let you specify the image format; use these as the actual file name extensions or the prefix to indicate output format for the image sequences.

Format

Bit Depths

Read/Write

Extension

Notes

AAF

n/a

read

aaf

Nuke Studio only

Alembic

n/a

read and write

abc

You can read meshes, point clouds, cameras, and transforms from Alembic files into a Nuke scene using the ReadGeo, Camera, and Axis nodes.

To write meshes and point clouds out again, use the WriteGeo node.

Apple ProRes

10, 12

read and write

mov

Support for the following:

Apple ProRes 4444 XQ 12-bit

Apple ProRes 4444 12-bit

Apple ProRes 422 HQ 10-bit

Apple ProRes 422 10-bit

Apple ProRes 422 LT 10-bit

Apple ProRes 422 Proxy 10-bit

ARRIRAW

Codex

12

read only

ari

arx

Alexa Mini LF

Codex HDE

Camera/Sensor

Nuke 11.2

Nuke 11.3

Nuke 12.0

Nuke 12.1

Nuke 12.2

Nuke 13.0

ARRIRAW SDK (.ari, .arx, and .mxf)

v5.4.3.5

v6.0

v6.0.0.4

v6.2.1.0

v6.2.1.0

v6.2.3.0

CUDA GPU Support

-

-

-

ALEXA Mini LF

-

-

-

ALEXA LF

ALEXA SXT(W)

ALEXA Mini

ALEXA 65

ALEXA XT

ALEXA Studio

ALEXA Classic

AMIRA

Note:  Nuke, Nuke Studio, and Hiero do not support .mxf from the AMIRA camera.

 

AVI

n/a

read and write

avi

On Windows, the AVI reader uses DirectShow multimedia architecture. When decoding .avi files, DirectShow tries to find the appropriate codec on the system. If the codec is not available, DirectShow and Nuke is unable to open the .avi file.

BRAW

12

read

braw

 

CIN

10 (log)

read and write

cin

 

DNG

8, 12

read

dng

Includes RAW 2.5K CinemaDNG

DPX

8, 10, 12, and 16

read and write

dpx

YCbCr encoded DPX files are not supported on the timeline.

DTEX

32

read only

dtex

To use DTEX files, you need Pixar’s PhotoRealistic RenderMan® Pro Server 20, or earlier, on your computer.

To read a DTEX file, use the DeepRead node.

EDL

n/a

read and write

edl

Nuke Studio only

FBX

 

read and write

fbx

You can read meshes, point clouds, cameras, lights, and transforms from FBX files into a Nuke scene using the ReadGeo, Camera, Light, and Axis nodes.

To write geometry out again, use the WriteGeo node.

GIF

8

read only

gif

 

Radiance

16

read and write

hdr, hdri

This format stores an 8-bit mantissa for each of r, g, and b and an additional 8-bit exponent that is shared by all three, which packs the floating point RGB triplet into 32 bits per pixel.

JPEG

8

read and write

jpg, jpeg

Adjust compression levels using the quality slider in the Write node’s properties panel.

Maya IFF

8, 16

read only

iff

 

MOV

n/a

read and write

mov

The writer supports the following codecs:

Animation

Apple ProRes 4444 XQ 12-bit

Apple ProRes 4444 12-bit

Apple ProRes 422 HQ 10-bit

Apple ProRes 422 10-bit

Apple ProRes 422 LT 10-bit

Apple ProRes 422 Proxy 10-bit

Avid DNxHD 444 10-bit 440Mbit

Avid DNxHD 422 10-bit 220Mbit

Avid DNxHD 422 8-bit 220Mbit

Avid DNxHD 422 8-bit 145Mbit

Avid DNxHD 422 8-bit 36Mbit

Note:  Interlaced writing is not supported. See Avid DNxHD Notes for more information.

Avid DNxHR:

8-bit 4:2:2

DNxHR LB, DNxHR SQ, and DNxHR HQ

12-bit

DNxHR HQX 422 and DNxHR 444

H.264 Main 4:2:0 8-bit

H.264 High 4:2:0 8-bit

Motion JPEG A

Motion JPEG B

MPEG-4

Photo - JPEG

PNG

Uncompressed

write only

mov

DNxHR Pattern: OP-1A, OP-Atom

Profiles:

4:4:4 12-bit

HQX 4:2:2 12-bit

HQ 4:2:2 8-bit

SQ 4:2:2 8-bit

LB 4:2:2 8-bit

MXF

 

8, 10, 12

 

read only

mxf

Supported codecs include:

Uncompressed 4:2:2 YCbCr 8-/10-bit

Uncompressed 4:4:4:4 RGBA 8-/10-bit

Uncompressed Avid 4:2:2 YCbCr 8-/10-bit

Uncompressed Avid 4:4:4:4 RGBA 8-/10-bit

JPEG2000

Avid DNxHD (1080p and 720p 1920x1080 and 1280x720, 4:4:4:4 and 4:2:2) 36, 115, 120, 145, 175, 185, 220, 220x
See Avid DNxHD Notes for more information.

Avid DNxHR:

8-bit 4:2:2

DNxHR LB, DNxHR SQ, and DNxHR HQ

12-bit

DNxHR HQX 422 and DNxHR 444

Sony Raw from the F65, F55, F5 and FS700 cameras. All formats that these cameras provide: 4K, 2K, 1K, 0.5K and 0.25K

Sony X-OCN from the VENICE, F55, and F5 cameras.

ARRIRAW from the Alexa Mini.

write only

mxf

DNxHR Pattern: OP-1A, OP-Atom

Profiles:

4:4:4 12-bit

HQX 4:2:2 12-bit

HQ 4:2:2 8-bit

SQ 4:2:2 8-bit

LB 4:2:2 8-bit

Camera/Sensor

Nuke 11.2

Nuke 11.3

Nuke 12.0

Nuke 12.1

Nuke 12.2

Nuke 13.0

SONY SDK (.mxf)

-

v3.1

v3.3

v3.3

v3.3

v3.3

SONY SMDK

-

-

v4.18.0

v4.18.0

v4.19.0

v4.19.0

GPU Support

-

-

VENICE

-

F55

F5

F65

FS700

Avid DNxHD Notes

The bit rates listed in the Codec Profile dropdown are the bit rates for 1080p at 29.97 fps EXCEPT for 36 fps (which is actually 45 Mbps @ 29.97fps). You should look at the codec format (422/444, 8/10-bit).

Note:  Nuke only supports 1080p and 720p. Non-HD resolutions are scaled to 1080p before writing.

This leads to a set of 1080p bit rates:

1080p/29.97 440x, 220x, 220, 145, 45

1080p/60 N/A, N/A, 440, 290, 90 (same at 59.94)

1080p/50 N/A, N/A, 367, 242, 75

1080p/25 365x, 185x, 185, 120, 36

1080p/24 350x, 175x, 175, 115, 36 (same at 23.976)

At 720p, the Codec Profile dropdown has a different interpretation. The bit rate is taken as the bit rate at 720p at 59.94fps. This leads to another set of bit rates:

720p/59.94 N/A, 220x, 220, 145, N/A

720p/50 N/A, 175x, 175, 115, N/A

720p/29.97 N/A, 110x, 110, 75, N/A

720p/25 N/A, 90x, 90, 60, N/A

720p/23.976 N/A, 90x, 90, 60, N/A

Note:  Since the bit rates are for 1080p at 29.97 fps AND 720p at 59.94 fps (except for 36 Mbit which should read 45 Mbit). It is possible to calculate the bandwidth for all the other frame rates by:
BandWidth@1080p = fps/29.97 * NominalBandWidth, or
BandWidth@720p = fps/59094 * NominalBandWidth
where NominalBandWidth is the bandwidth listed in the codec profile knob OR 45 if the bandwidth listed is 36 Mbit. (Avid labels the codec profile names by the approximate bandwidth.)

OBJ

n/a

read and write

obj

 

OpenEXR

and

OpenEXR 2.4.2

16, 32

read and write

exr

OpenEXR handles 16- and 32-bit float. This 16 is also called "half float" and is different from the 16-bit integer that all the other formats that support 16-bit use.

Nuke supports multi-part OpenEXR files. For more information, see Notes on Importing OpenEXR Files and Notes on Rendering OpenEXR Files.

When working with deep data, Nuke supports scanline OpenEXR files. For more information, see Importing Scanline OpenEXR Files and Writing Deep Data.

EXR Compression

EXR file metadata contains a compression key/value pair detailing the compression type used to write the .exr file. The value is expressed as the name of the compression type or an integer referencing the compression used:

0 - no compression

1 - RLE compression, run length encoding

2 - Zip compression, one scan line at a time

3 - Zip compression, in blocks of 16 scan lines

4 - PIZ-based wavelet compression, in blocks of 32 scan lines

5 - PXR24 compression, lossy 24-bit float

6 - B44 compression, lossy 4-by-4 pixel block, fixed rate

7 - B44A compression, lossy 4-by-4 pixel block, flat fields are compressed more

8 - DWAA compression, lossy DCT based compression, in blocks of 32 scan lines

9 - DWAB compression, lossy DCT based compression, in blocks of 256 scan lines

PIC

SoftImage®

8

read and write

pic

 

PNG

8, 16

read and write

png (8-bit)

png16 (16-bit)

 

PSD

8, 16

read only

psd

While Nuke reads standard Photoshop® blend modes, it doesn't read Photoshop layer comps or recognize group blend modes. Photoshop layers are read into separate Nuke layers and anything that doesn't map into that is ignored.

R3D

16

read only

r3d

 

Camera/Sensor

Nuke 11.2

Nuke 11.3

Nuke 12.0

Nuke 12.1

Nuke 12.2

Nuke 13.0

R3D SDK (.r3d)

v7.0.6

v7.0.6

v7.1.0

v7.1.0

v7.2.1

v7.3.1

CUDA GPU Support

-

-

KOMODO

-

-

-

-

MONSTRO 8K VV

HELIUM 8K S35

SCARLET-W 5K

RED RAVEN 4.5K

WEAPON DRAGON 6K

EPIC/SCARLET DRAGON

RED ONE MYSTERIUM 4K

MYSTERIUM-X

MYSTERIUM-X MONOCHROME

RAW

n/a

read only

n/a

DSLR raw data files, such as Canon .CR2 files. These are only supported via the dcraw command line program, which you can download from the dcraw website. Bit depth and other specifications depend on the device. Some devices may not be supported.

RLA

Wavefront®

8

read only

rla

 

SGI

8, 16

read and write

sgi, rgb, rgba (8-bit sequences)

sgi16 (for 16-bit sequences)

 

TIFF

8, 16, and 32

read and write

tif, tiff (8-bit sequences)

tif16, tiff16 (16-bit sequences)

ftif, ftiff (32-bit sequences)

If utilized, the compression schema on imported TIFF sequences must be LZW®.

TARGA

Truevision®

8

read and write

tga, targa

 

USD

n/a

read

usd, usda

 

XML

n/a

read and write

xml

Nuke Studio only

XPM

8

read and write

xpm

This is the text-based format in which Nuke’s interface elements are stored.

YUV

8

read and write

yuv

This format does not specify resolution, so Nuke assumes a width of 720 pixels.