Could you post your source PSD with the two spot channels you're merging?
Jeffrey, did you get the file and being able to reproduce? Do you have any clue about it?
Hi Jeffrey, are you working on this bug?
I'm wondering if anything is going on about this. It's still a problem for us! Does the file I sent is showing the problem clear enough? Here is a new one for you. Juste look at the name of the channels (it's the CMYK value of the spot color). When merging in Photoshop 2023, it become something else with very different values... in Photoshop 2022 it's keeping the original values. Please tell me if you can see the problem and if anybody is workin on it. We are expecting an issue soon
I can reproduce the difference between 22 and 23 (with identical Color Settings) with »source en channels low res.psd«.
It seems to affect the first blue on its own right away, not the first pink, though.
Disabling GPU doesn’t seem to make a difference, either.
But the difference does not seem to arise in RGB.
Thank you c.pfaffenbichler. I'm glad to know that the problem is quite clear. What is the next step now? As working in RGB is not an option for us and we are working with documents many years after their conception, they have to repeat the same using the "merge" action. Is anybody from Adobe will work to solve it?
Is anybody from Adobe will work to solve it?
Unfortunately only an Adobe employee could definitely tell you, I guess.
@PECourtejoie already pinged Jeffrey Tranberry, so we can hope that he will look into it again.
But even if he does there are no guarantees about a timeline for a possible bug-fix even if it happens.
Bugs that affect only a small number of Photoshop users … I expect you can guess the rest.
Here is a basic file (source solid colors) without gradients so it will be easier to talk about cmyk values variations. I've put also a screenshoot of the channels value and comparative visuals of every CMYK channels (v22 is upper and v23 is under). The v22 version is keeping the same CMYK values after merging the channels and v23 don't.
Copy link to clipboard
Can moderator convert this discussion into a bug post?
Copy link to clipboard
Hey everybody, wanted to share some insight into this issue that hopefully will answer your questions.
It is correct that you can see differences in the merged results between v22 and v23. However, the merged results in v22 (and earlier versions of Ps) were incorrect. The merged results now produced with Ps v23 (and later) are actually the right results, but unfortunately they are different.
The differences you are seeing are real, but not a bug. It is the result of fixing a previous bug with how spot channels were merged in CMYK based docs.
Sharing a file that shows this... notice how far off the colors are from the bottom-middle result. All of these images should be reasonably close to the same and only limited by their respective color spaces/profile gamuts.
Thanks everyone, hope this helps.