OpenColorIO 2 Configs for ACES

Discussions - Using ACES
21

@TooDee Golly that does look quite skewed in that video!

The difference between the two configs I believe is that the ACES-CG config is using a built-in transform for the sRGB display:


  - !<ColorSpace>
    name: Display - sRGB
    family: Display
    equalitygroup: ""
    bitdepth: 32f
    description: Convert CIE XYZ (D65 white) to sRGB (piecewise EOTF)
    isdata: false
    categories: [file-io]
    encoding: sdr-video
    allocation: uniform
    from_display_reference: !<BuiltinTransform> {style: DISPLAY - CIE-XYZ-D65_to_sRGB}

while the Maya one is not:


  - !<ColorSpace>
    name: sRGB
    family: Display
    description: |
      sRGB monitor (piecewise EOTF)
    isdata: false
    categories: [ file-io ]
    encoding: sdr-video
    from_display_reference: !<GroupTransform>
      children:
        - !<MatrixTransform> {matrix: [ 3.240969941905, -1.537383177570, -0.498610760293, 0, -0.969243636281, 1.875967501508, 0.041555057407, 0, 0.055630079697, -0.203976958889, 1.056971514243, 0, 0, 0, 0, 1 ]}
        - !<ExponentWithLinearTransform> {gamma: 2.4, offset: 0.055, direction: inverse}
        - !<RangeTransform> {min_in_value: 0., min_out_value: 0., max_in_value: 1., max_out_value: 1.}

You could confirm that by swaping in that groupTransform in place of the BuiltinTransform in the ACES-CG config.

While you’re at it, you may want to consider using the pure 2.2 display instead

 - !<ColorSpace>
    name: Gamma 2.2 / Rec.709
    family: Display
    description: |
      Gamma 2.2 monitor with Rec.709 or sRGB primaries
    isdata: false
    categories: [ file-io ]
    encoding: sdr-video
    from_display_reference: !<GroupTransform>
      children:
        - !<MatrixTransform> {matrix: [ 3.240969941905, -1.537383177570, -0.498610760293, 0, -0.969243636281, 1.875967501508, 0.041555057407, 0, 0.055630079697, -0.203976958889, 1.056971514243, 0, 0, 0, 0, 1 ]}
        - !<ExponentTransform> {value: 2.2, direction: inverse}
        - !<RangeTransform> {min_in_value: 0., min_out_value: 0., max_in_value: 1., max_out_value: 1.}
1 Like
22

Hello,

@carolalynn, @doug_walker, @Michael Dolan, @KevinJW, @zachlewis and all the other contributors are pleased to announce that we have a first version of the Studio Config: Release OpenColorIO-Config-ACES 0.3.0 · AcademySoftwareFoundation/OpenColorIO-Config-ACES · GitHub

Please give it a crack and let us know if you have any issues!

Summary of the Studio Config

Displays

  • sRGB
  • Rec.1886 / Rec.709 Video
  • Rec.1886 / Rec.2020 Video
  • Rec.2100-HLG
  • Rec.2100-PQ
  • ST2084-P3-D65
  • P3-D60
  • P3-D65
  • P3-DCI

Linear Colorspaces

  • ACES2065-1
  • ACEScg
  • Linear P3-D65
  • Linear Rec.2020
  • Linear Rec.709

Camera & Grading Colorspaces

ACES

  • ACEScc
  • ACEScct

Blackmagic Design

  • DaVinci Intermediate WideGamut
  • Linear DaVinci WideGamut
  • DaVinci Intermediate Log - Curve

Panasonic

  • Linear V-Gamut
  • V-Log V-Gamut
  • V-Log - Curve

RED

  • Linear REDWideGamutRGB
  • Log3G10 REDWideGamutRGB
  • Log3G10 - Curve

Sony

  • Linear S-Gamut3
  • Linear S-Gamut3.Cine
  • Linear Venice S-Gamut3
  • Linear Venice S-Gamut3.Cine
  • S-Log3 S-Gamut3
  • S-Log3 S-Gamut3.Cine
  • S-Log3 Venice S-Gamut3
  • S-Log3 Venice S-Gamut3.Cine
  • S-Log3 - Curve

ITU

  • Camera Rec.709

Texture Colorspaces

  • Gamma 1.8 Rec.709 - Texture
  • Gamma 2.2 AP1 - Texture
  • Gamma 2.2 Rec.709 - Texture
  • Gamma 2.4 Rec.709 - Texture
  • sRGB - Texture

Utility Colorspaces

  • Un-tone-mapped
  • CIE-XYZ-D65
  • Raw
  • Rec1886 - Curve
  • Rec.709 - Curve
  • sRGB - Curve
  • ST-2084 - Curve

Cheers,

Thomas

4 Likes
23

Congratulations to everyone involved!!! Incredible work!

1 Like
24

Fantastic work all! Any particular reason AWG3/4 / LogC3/4 are missing?

1 Like
25

Hi @Shebbe ,

This is a great question and it was asked on Slack: Yes, there is a good reason, we haven’t finished the work with ARRI (@scoopxyz) to finish integrating the ARRI transforms: DRAFT: add ARRI contributions to OCIO-ACES CLFs by scoopxyz · Pull Request #54 · AcademySoftwareFoundation/OpenColorIO-Config-ACES · GitHub and we wanted to have a release for Siggraph!

They will come soon, stay tuned!

Cheers,

Thomas

2 Likes
26

Precision landing!
Thanks for all your great work!

Peter

1 Like
27

The same is true for the currently missing Canon input transforms too, yes?

28

Correct! Sorry for the delay but yes, the Canon IT’s will be in the next release as well.

2 Likes
29

OpenColorIO Configuration for ACES - Release Candidate 1

Hello,

The OCIO Configs Working Group is pleased to let you know that the OpenColorIO Configuration for ACES - Release Candidate 1 configs are available:

Cheers,

Thomas

4 Likes
30

Hi,

The OCIO Configs Working Group is pleased to announce that the OpenColorIO Configuration for ACES - Release Candidate 2 configs are available. This is hopefully the latest release before the final in a few days: Release OpenColorIO-Config-ACES 1.0.0-rc2 · AcademySoftwareFoundation/OpenColorIO-Config-ACES · GitHub

Cheers,

Thomas

4 Likes
31

Hello,

The OCIO Configs Working Group is pleased to announce the stable release of the new OpenColorIO Configuration(s) for ACES. We sincerely thank all the people from the past and the present that contributed to this community effort.

Cheers,

Thomas

3 Likes
32

@Thomas_Mansencal could you say a bit about how to properly use these texture color spaces? Thanks!

33

Hi @Derek,

Those are intended to be used for texture workflows, the way to read it is as follows:

%OETF/EOTF-1/CCTF% %Gamut% - Texture

So for example:

  • Gamma 1.8 Rec.709 - Texture decodes textures that have been encoded using Rec.709/BT.709/sRGB Gamut and the Gamma 1.8 OETF/CCTF.
  • sRGB - Texture decodes textures that have been encoded using Rec.709/BT.709/sRGB Gamut and the sRGB EOTF-1/CCTF.
  • sRGB Encoded AP1 - Texture decodes textures that have been encoded using ACEScg/AP1 Gamut and the sRGB EOTF-1/CCTF.

Cheers,

Thomas

34

Thanks so much. So a couple follow-up questions:

  1. Would that mean that if a texture was made under an ACES (sRGB) Output Transform it would be read in with sRGB -Texture, and if it was made under an ACES (pure gamma 2.2) Output Transform it would be read in with Gamma 2.2 Rec.709 -Texture?

  2. What would be the situation where a texture is encoded with AP1 gamut rather than Rec.709 gamut? Would that be done by setting the color picking space in the 3D paint program to ACEScg (for AP1) rather than to sRGB - Texture? Along those lines, why is there a Gamma 2.2 AP1 - Texture and not a sRGB AP1 -Texture? Or is that what sRGB Encoded AP1 - Texture means?

35

My guess is that AP1 texture spaces are introduced to start supporting wider gamut textures (albedo) for materials. Scans or creations don’t need to be limited by sRGB primaries anymore and can be more physically accurate.

Pointer’s gamut vs sRGB primaries
image

Pointer’s Gamut vs Rec. 2020 primaries
image

36

The Linear Rec.709 (sRGB) color space has:

    from_scene_reference: !<MatrixTransform> {matrix: [2.52168618674388, -1.13413098823972, -0.387555198504164, 0, -0.276479914229922, 1.37271908766826, -0.096239173438334, 0, -0.0153780649660342, -0.152975335867399, 1.16835340083343, 0, 0, 0, 0, 1]}

Previously it was:

    to_reference: !<MatrixTransform> {matrix: [ 0.439632981919, 0.382988698152, 0.177378319929, 0, 0.089776442959, 0.813439428749, 0.096784128292, 0, 0.017541170383, 0.111546553302, 0.870912276314, 0, 0, 0, 0, 1 ]}

Curious what the reason was for the change? They appear to do the same thing.

37

The direction has changed! :slight_smile:

1 Like
38

We do use that in Unreal Engine all the time, our working space is ACEScg and UE has buitlin sRGB encoding/decoding transfer functions.

39

Thanks @Thomas_Mansencal
so just to make sure I’m correctly understanding: This would be used for a texture map saved as EXR (in ACEScg AP1) that needs to be read into Unreal Engine?

40

Not EXR files, but 8-bit texture files that use sRGB EOTF-1 and AP1, e.g. PNG.