Open reinecke opened 3 years ago
There's a high-level decision I think we have to make about modeling the color pipeline through the timeline heirarchy where we select from two different approaches:
MediaReference
, Clip
, and Timeline
level and allow some system (like OpenColorIO) to discover how to go from one to the next via some out-of-scope to OTIO rulesetDo we want to introduce ACES AMF into an environment where we are strongly centralized around OCIO, and now OCIO/NanoColor? If we don't get it "for free" from OCIO, I'm very hesitant.
My recommendation:
Start by raising the topic at the ASWF Color Interop forum https://wiki.aswf.io/display/CIF/Color+Interop+Forum+Home and gather some opinions.
There should be an analysis of color transforms and how they fit in an editorial composition. See the nanocolor transformation charts for an example.
There should be an examination of where color space transformations actually exist in NLEs. I have a strong suspicion that the granularity proposed here is not accommodated in general.
In image pipelines it can be useful to know the intended colorspace of a timeline as well as the media composed into that timeline when trying to either re-create a "look" or find out what someone earlier in the pipeline may have been using in their work context (often to try and locate mismatches). OTIO should have a way to describe colorspaces used in certain contexts.