Timeline Gamut- Node Gamut Interaction (Beta 3)

Not sure if this is a bug or a feature (I don't fully understand).

Use YRGB (unmanaged) color science.

Select timeline gamut Rec.709, create a node with a L*a*b colorspace override. Everything works fine, the RGB channels are correctly mapped to L*a*b and DCTLs use 0.5f as the a and b color centers.

Now set the timeline gamut to CEI L*a*b which is useful to watch the CIE 1931 color space diagram in the scopes.

But now the RGB channels are no longer correctly mapped, selecting or deselecting L*a*b for the node colorspace override makes no difference.

What does the manual say?

I have found funky stuff happening in this black box also, and as much as I'd like to find an explanation, I don't. What is missing for Resolve is a technical reference manual. Currently there is only a storybook that does not answer a lot of questions. Technical reference is necessary because it gives an answer to whether something is a feature or a bug in addition to just telling plain and clear what is going on.