Recap:
- @jedsmith merged his ‘kitchen sink’ version into the super-repo model dir for testing - please pull it down and give it a go: https://github.com/colour-science/aces-vwg-gamut-mapping-2020/tree/master/model
- @daniele - have we settled on AP1 as the target primaries? If we chose others, could we help the hue shift? Consensus is that AP1 makes the most sense in an ACES based system, but open to other feedback.
- This round of group testing - within Virtual Working Group Members - will be on all versions to hopefully narrow down to one set of curve + default settings for user testing
- @nick has a branch with curve overlays to assist - will update when it’s merged to the super repo
- Wider user testing framework: asking for feedback, please add your thoughts and other test operations to the Dropbox Paper
- We should add inversion to the testing framework for compositing - and recompression in grading
- Touched on invertibility. In regards to the archive, @daniele brought up that ourselves 20 years from now will be (hopefully) smarter than we are now, so we should archive what we are confident in and not what we think we’ll want later.
- Shadow rolloff: default value is 0. Should we deal with pixels with all three negative values?
- Threshold - start at 0 or 1? Asked for feedback - everyone has their own preferences. Interested in what different software packages do in their tools around these types of parameters. And what should it be called?
- Distance sliders - cyan magenta yellow? But threshold RGB?
Here’s the link to recording/transcript.