- This topic has 52 replies, 4 voices, and was last updated 4 years, 8 months ago by
.
-
Posts
-
Lightroom has been mentioned as glitching too.
I don’t have Catalina.
At the moment it looks like the discussion at Colorsync list describes similar symptoms, but the convo is unfocused.
You can browse the archives here, Wire:
Thx for archive link
After looking into the problem with TRC discontinuity on colorsync-users I think the answer to your original question as to whether color on Mac is just messed up is and that’s the current state of affairs is: Yes
I describe my experience over there if you wanna look.
The good news is the failure modes are so arbitrary and nuanced that it largely appears to work——except when it doesn’t
I amazed that 25 years on from the origins of the ICC and Colorsync the promise of consistent color across devices has utterly failed to be achieved. And at this point I would say that outside of each individuals own silo of imaging experience, wherein the more diligent he is the more he tends to think is a keeper of the light truth, there is no hope of this ever being achieved.
On my Mac I can not get any color managed app to reliably agree with any other re the most basic numerical evaluations, much less say anything with confidence about how it will play with world at large.
This is an amazing epistemological discovery! Using purely deterministic mechanisms it’s pretty likely the color on my screen doesn’t agree with anyone else’s using these same mechanisms! Yet it also pretty much works.
I’m becoming suspicious that we can’t never get these mechanisms perfectly aligned because our sensory faculties are so variable. Even if the numbers are right everywhere it will be thwarted by my seeing something weird.
I should like to think that Apple noted this years ago and just quietly gave up on Colorsync as a waste of time 🙂 It works approximately among those who constantly cross-check w/each other. Which is good for business
We’ll know the situation is understood and resolved when people can stop checking it.
FWIW, the folks at the list did say the bug got introduced some time after 10.14.3, so it must be somewhat recent. Not the first time Apple screwed up, neither the last. Luckily we still have everything before Mojave that works, as I said it did work fine in El Capitan for me, but then I upgraded straight to 10.14.5 and so started witnessing that misbehavior straight away. It’s not that bad, but it will be bad if Apple totally ignores the issue. Should we file a bug using https://feedbackassistant.apple.com/welcome ? Maybe that’s the proper way forward, doubt they monitor the lists anyway.
Sure.
I’m still channeling Colorsync’s skrogging DisplayCal XYZ LUT profiles.
One of the posts on colorsync-users “discontinuity” thread got me looking at the ICC profile spec:
If I grok that doc properly, The whole format has been settled business, spec-wise, since the 90s. And the spec reads it’s totally appropriate for a display profile to harbor both TRC and cLUT tags. so presumably if the profile format allows this, an implementations ought not to just barf up random output.
If Apple doesn’t want to support it, that’s another matter. And we’re talking the masters of ICC here. A lot of this is Apple’s baby.
I am hoping that whatever is defect results in the cLUT-style display profile rendering glitch, some boundary condition living deep in Colorsync for eons, is now being exposed by classic pure TRC profiles in 10.14.6.
If it has finally surfaced for garden variety config, it will be fixed—notwithstanding that I have no technical basis for this hope. Just a hunch. 🙂
Maybe someone at colorsync-users may connect these dots and at least try to llook for a connection.
But the stone of the posts over on colorsync-users suggests no one at Apple has participated in the living memory of the list.
It would be pretty funny if in next release the MacOS displays color panel just greys out most of the selections with no explanation, but it wouldn’t surprise me.
