Inconsistent and possibly incorrect `ACEStransformID` for ARRI LogC4 about aces-dev HOT 6 CLOSED

Thanks, let's close that one as resolved!

from aces-dev.

@doug-walker and I (and Carol) decided to change it on the OCIO side as it is easier and gets us moving.

from aces-dev.

Copying my response from the Colour-Science slack thread:

I had to go back and look at why we chose the present form. If I look at the LogC3 EI 800 IDT that is:
IDT.ARRI.Alexa-v3-logC-EI800.a1.v2

And the present LogC4 is:
IDT.ARRI.ARRI-LogC4.a1.v1

This is something like <Class>.<Vendor>.<ColorSpaceName>.<ACESVersion>.<Version>

So "ARRI LogC4" is the official name of the color space, with its component parts being "ARRI Wide Gamut 4" and "ARRI LogC4 Curve". This information is also mirrored in the white paper. So I would actually vote to have the aces-dev TransformID remain as-is. Despite the quasi-ugliness of having the vendor name appear twice.

I'm open to be convinced in another direction, but that is my reasoning.

As to why its different in the OpenColorIO Config, this was based on the observation that it already takes "creative" liberty in the ACES ID. For example between...
...
Which influenced the OCIO config discussion, in addition to the conversation with Doug in the PR: AcademySoftwareFoundation/OpenColorIO-Config-ACES#54

I can repeat this in the OCIO slack as well.

from aces-dev.

I take back what was said about the other discrepancy, it was referring to another CTL file which I didn't see on my first glance.

So to summarize this specific issue, it seems like it is different because it was different "on purpose" from feedback in the PR. I would still vote to adopt the name as-is in the aces-dev repo, since we're talking about an identification string and not a speaking name, I don't think the aesthetics of the string are of much concern. If there's another angle I'm not seeing please let me know.

from aces-dev.

Thanks for looking into this Sean! Thomas changed the ADESTransformID in the OCIO config to match what is in aces-dev, so they are both urn:ampas:aces:transformId:v1.5:IDT.ARRI.ARRI-LogC4.a1.v1. I confirmed that this is what is in the latest built-in OCIO configs.

I think this issue could be closed now.

from aces-dev.

(I will indicate that we changed the ACEStransformID in the config release notes.)

from aces-dev.