Hey,
at some point in the OT VWG (maybe like 2 years ago), I remember switching my CG tests to a “linear-bt.709” rendering space for the sake of simplicity.
This scene was indeed lookdev using TCAMv2 back in the day. But I do not think it matters for the point I am trying to make.
You may observe the same issue by:
- having your working space set to ACEScg
- setting the light to (0,0,1) in “linear-bt.709” (with a constant in Nuke for instance)
I think it is a valid choice to keep working/rendering in “linear-bt.709” but using the ACES 2.0 Output Transforms for display.
Here are more images:
Working space is ACEScg but the light is a “linear-bt.709” blue primary:
Working space is ACEScg but the light is an “ACEScg” blue primary:
The breakup is easy to reproduce with “lin-bt.709” ramp displayed with the ACES 2.0 Output Transforms:
Hope that clarifies things,
Chris