ACES 2.0 Output Rendering

Mine is a straw yellow: 0.9209, 0.7255, 0.2385 in sRGB, which C4D converts to 0.6754, 0.5034, 0.1106 in ACEScg.

I thought it would be interesting to render a version without the red and blue lights.

(All of these have the same highlight bloom applied as in the initial examples.)

1 Like

I think only one or two patches exceed BT.709 (quick check: one of them - patch 18/F3).
A while ago I made this exr with ACEScg samples from Pointer’s gamut:
mP_PointersGamut_MaxSaturaton_acescg.exr.txt (103.4 KB)
for edge cases in cgi i try to hoover around those values for acescg albedos.

Wow, that’s a very useful resource, thank you for sharing it!

I’ve learned so much on this thread, thanks to everyone who has participates in a spirit of sharing and curiosity.

@piotrus3333 I took your EXR and made an EV ramp from it, and of course put that through the various display transforms.

3 Likes

Something that I think these EV ramps illustrate for me is that, as we all know, ACES 1.0 had a tendency to “tilt” towards yellow. Some folk liked that, some folks hated it (I’m looking at you Troy!). You can see this in the EV ramp.

I’d say that ACES 2.0 now has a tenancy to tilt towards red. You can also see this in its EV ramp. This can make sunlight turn a depressing pink salmon color, rather than golden warm yellow color, as God in her wisdom intended. :wink:

OpenDRT and other transforms (RED IPP2 for example) I think do a better job at avoiding these tilts.

  • Disclaimer, “tilt” is not a technical term, so hopefully the nerdy among us will be able to get my meaning here.

Here’s a closeup of the yellow sphere with ACES2.0. Note how the red light at high exposure is producing a fleshy salmon result on the sphere.

If we lower the exposure -1.5 stops we can see how the red light overpowers the yellow sphere viewed in ACES 2.0.

Compare that to OpenDRT (Default) with -1.5 exposure. Here the red light is turning the yellow sphere orange, as one would expect.

Likewise, at the original high exposure OpenDRT retains the orange look.

What I conclude from this is that the issue with ACES2.0 is not so much that red goes to pink at high exposure. Of course it does. Rather the issue is how red has such a dominant overpowering effect on other colors. You might say there is too much weight given to red.