Suggestion: Regarding ACES Input Transforms

Thank you for implementing ACES 2.0 support. I'm always impressed by how quickly new features are adopted. Thank you.

I would like to suggest that sRGB (Texture) and sRGB (Linear), which are currently located in the "Inverse Display → D65" folder, be moved back into the "Color Space Conversion" folder, where they were in previous versions. These two transforms are fundamentally different from the other Inverse Display options.

For reference, in Nuke, the default ACES input transforms are categorized as follows:

- ACES
- Output
- Input
- Utility

In comparison:
- The ACES folder corresponds to the "ACES Conversions" section in DRS20.
- The Output folder aligns with the IDTs in DRS20, such as Apple, ARRI, RED, Sony, etc.
- The Input folder matches the "Inverse Display" section in DRS20.
- However, there is currently no equivalent to the "Utility" category in DRS20.

The issue is that sRGB (Texture) and sRGB (Linear) are now placed under "Inverse Display," which is misleading.

sRGB (Texture) is not the inverse of an ACES ODT, unlike the other transforms in "Inverse Display."

sRGB (Linear) is a commonly used working color space in CG workflows and has no direct relationship with display transforms.

In DaVinci Resolve 19, these two transforms were placed in the "Color Space Conversion" folder, which functioned similarly to the "Utility" category. That categorization was much more intuitive and appropriate.

Therefore, I suggest moving sRGB (Texture) and sRGB (Linear) back into the "Color Space Conversion" folder to better reflect their intended purpose and to improve clarity for users.

The official ACES 2.0 OCIO config has now been released:

https://github.com/AcademySoftwareFound ... S/releases

File: studio-config-v3.0.0_aces-v2.0_ocio-v2.4.ocio

Based on the official OCIO color space structure, I have a few suggestions regarding the ACES Input Transform options in DaVinci Resolve.

No Input Transform
(Internally, this seems to be treated as ACES2065-1. This makes sense and should probably remain as-is.)

※ In the official config, the following camera-specific transforms are organized under the "Input" category.

Apple
• Apple Log
• Apple Log Rec.2020 -CSC
ARRI
• ARRI LogC3
• ARRI LogC4
• ARRI LogC EI800 AWG -CSC（Seems redundant with LogC3, possibly unnecessary）
• ARRI LogC4 -CSC（Seems redundant with LogC4, possibly unnecessary）

BlackmagicDesign
• Blackmagic Design Film
• Blackmagic Design 4K Film
• Blackmagic Design 4K Film Gen 3
• Blackmagic Design 4K Video Gen 3
• Blackmagic Design 4.6K Film
• Blackmagic Design 4.6K Film Gen 3
• Blackmagic Design 4.6K Video Gen 3
• Blackmagic Design Film Gen5
• Blackmagic Design Film Gen5 -CSC (Seems redundant, possibly unnecessary）
Canon
• (Omitted for brevity)
DJI
• DJI DLog DGamut -CSC
Fujifilm
• Fujifilm F-Log
• Fujifilm F-Log2
Panasonic
• Panasonic V35
• Panasonic Varicam VLog VGamut -CSC
RED
• (Omitted for brevity)
Sony
• Sony Raw
• SLog1
• SLog2
• SLog2 (Daylight)
• SLog2(Tungsten)
• SLog3
• SLog3 SGamut3
• SLog3 SGamutCine
• Sony Venice SLog3 SGamut3
• Sony Venice SLog3 SGamut3Cine
• SLog3 SGamut3 -CSC（Possibly redundant）
• SLog3 SGamutCine -CSC（Possibly redundant）
• Sony Venice SLog3 SGamut3 -CSC（Possibly redundant）
• Sony Venice SLog3 SGamut3Cine -CSC（Possibly redundant）

Inverse Output Transform (Currently labeled as “Inverse Display” in Resolve20, which is clear, but calling it “Inverse Output Transform” or “Inverse Output” might align better with the terminology used in the official config, where these are categorized under “Output”.)
• Inverse Rec709.BT.1886
• Inverse sRGB Piecewise
• Inverse sRGB Gamma 2.2
• Inverse DisplayP3
• Inverse DisplayP3 Gamma 2.2
• Inverse Extended Display P3 (1000 nit)
• Inverse P3-D65
• Inverse P3-D65 ST2084 (108 nit)
• Inverse P3-D65 ST2084 (500 nit)
• Inverse P3-D65 ST2084 (1000 nit)
• Inverse P3-D65 ST2084 (2000 nit)
• Inverse P3-D65 ST2084 (4000 nit)
• Inverse P3-D65 (Rec.709 Limited)
• Inverse Rec.2100 HLG (1000nit P3-D65 Limited)
• Inverse Rec.2100 ST2084 (500nit)
• Inverse Rec.2100 ST2084 (1000nit)
• Inverse Rec.2100 ST2084 (2000nit)
• Inverse Rec.2100 ST2084 (4000nit)
• Inverse Rec.2100 ST2084 (100nit Rec.709 Limited)
• Inverse Rec.2100 ST2084 (500nit P3-D65 Limited)
• Inverse Rec.2100 ST2084 (1000nit P3-D65 Limited)
• Inverse Rec.2100 ST2084 (2000nit P3-D65 Limited)
• Inverse Rec.2100 ST2084 (4000nit P3-D65 Limited)
• Inverse DCDM (P3-D65 Limited)
• Inverse DCDM (300nit P3-D65 Limited)

Utility
(While these are currently listed under “Color Space Conversions” in Resolve versions prior to 19, it may be more intuitive to follow the official config and group them under “Utility”.)
• sRGB(Texture) - CSC
• sRGB(Linear) - CSC
• Rec.2020(Linear) （It would be helpful if you could prepare it.）

ACES Conversions
• ACES2065-1 (Even if there's some overlap with “No Input Transform,” it may still be useful to explicitly include these.)
• ACEScc
• ACEScct
• ACEScg
• ADX10
• ADX16

+1

No Input Transform
(Internally, this seems to be treated as ACES2065-1. This makes sense and should probably remain as-is.)

Despite being technically accurate I actually really dislike that this exists. When converting to ACES2065-1 it makes sense to actually define that as the output space rather than something called "no output". It's also super ambiguous because tagging files as "no transform" in an ACES managed project is actually suggesting you're treating it as an ACEScct input space ie. matching working space rather than ACES2065-1 file. Overall there is zero benefit to this idea over naming it ACES2065-1 and properly placing it in the ACES sub menu regarding ACES Input Color Space menu like you suggested.

Inverse Output Transform (Currently labeled as “Inverse Display” in Resolve20, which is clear, but calling it “Inverse Output Transform” or “Inverse Output” might align better with the terminology used in the official config, where these are categorized under “Output”.)

I think it's fine personally. The definition is accurate. The ACES OCIO config is not the reference per se regarding how the sub menus should be named or grouped.

What annoys the most is that the ACES Transform node drop down menu is still not categorized the same way the ACES Input Color Space right-click menu is. I have a gut feeling that this is sheer technical limitations of how that dropdown menu works but it would be lovely to have, also for DCTLs.
BMD at least sees the issue because they flipped the grouped alphabetical ordering to a more custom one with 'most used displays' at the top (thanks for that). But the list is so massive, it's really annoying having to scroll through it to find what you need.
The official ACES OCIO configs partially mitigate this by giving 3 sets of configs. One with D65 only, one with D60sim only and another with both.

Giving users the option to hide certain color spaces in ACES would help an insane amount. But having proper categorization would be incredibly welcome too.

It's also super ambiguous because tagging files as "no transform" in an ACES managed project is actually suggesting you're treating it as an ACEScct input space ie.

I agree with shebbe — it's definitely confusing that something labeled as "No Input Transform" is actually being converted from ACES2065-1 to ACEScct (the working color space) under the hood in the Edit and Color pages. And to add to the complexity, in the Fusion page, it's converted from ACES2065-1 to ACEScg (which is the fixed working space there).

Honestly, just renaming "No Input Transform" to ACES2065-1 might be a simple and effective improvement.

That said, I do think the current behavior — where media is interpreted as ACES2065-1 (currently labeled as "No Input Transform") by default — should probably stay as-is for the sake of project compatibility.
And yes, ACES2065-1 should definitely be available as an option in the ACES Input Color Space menu too.

Arri Log c3 is miising from ACES transform list