ChromIQ profile - soft proofs fine but prints are off

[mjhn]

Using v. 4.3.2 Basically, create the profile by printing out the target (1 page) thru Adobe Print Utility as well as thru ChromIQ to ensure no color management's applied. Created profile "soft proofs" in Ps just fine - same visual as when using Canson's Photographique Rag II (trying to profile Epson Legacy Fiber for use on Canon Pro-1000, which Canson's profile has been pretty accurate). However when printing the file using the profile, it's considerably lighter and color shifts in the greens and purples, overall lighter and less saturated with a higher black point. So soft proof looks to be accurate, but something happens when that info is fed to the print driver. My printing workflow/set-up is correct for Ps

i1Pro with UV cut
Mac
Ps 2026

 

[itsab1989]

Can you tell me which operating system you are using?
A few days ago I visited a friend of mine. He had bought an Epson ET8550. I wanted to help him set it up and install some of my own profiles for the device on his system. Although I made the ChromIQ I am still using some older profiles from before this. They are working fine on my Mac but on his Windows machine the prints turned out pretty bad.
They were desaturated as you describe but in this case the images looked kind of sepia / brown. I played around with the driver settings on his system but neither Photoshop nor Epson Print Layout produced good prints although the softproof was looking nice.
I had a chat with @pharmacist about this and his description about the Windows settings matched what I did. So I am not really sure what the issue might be.
I will try to do some research and maybe someone here can share some thoughts as well.

 

[mjhn]

I'm running Ventura OS. Another thing I can't figure out is printing the target thru ChromIQ, which says they disable any kind of color management, greens, blues and purples when printed thru Adobe Print Utility, (which is another way of printing it out without any color management since they took that ability away in Ps) are different shades though the saturation and overall values are similar. Chat and Claude's help were of no help with a lot of running around in circles.

 

[itsab1989]

It is really hard to disbale colormanagement correctly on macOS. I tried so many different things and the driver grey it out and locks it for me but sometimes it is not respected for the real print. Especially when you click cancel on any button in the driver setting this tends to reset things. But I don’t see what I can do about this.

 

[mjhn]

I also tried Canon's software for no-CM, which actually matched the prints out of Adobe, so I'm not sure what the differences are between those two and ChromIQ. Back before that option was removed from Ps, it seemed like a better process printing out the target thru Ps drivers with no-CM to create the test target, as that created profile was now going to be put back into the same printing/working flow. Still, my question is why does the profile softproof as one would expect (printing on a mat paper), yet prints completely different, when using Canson's Photo Rag II profile the softproofs and prints pretty much align. Weird.

 

[itsab1989]

The no colormanagement thing is something I am going to look into again at some point (although I have done it many times before without getting a definite result).
But the softproofing issue is almost certainly not related to ChromIQ itself because when it builds the profile it only passes the data to Argylls colprof and the profiles are no different then anything that commandline colprof would build for you. So there must be something wrong in the printing pipeline.
Or it does indeed have to do with the measured chart not having the correct colormanagement setting applied for the print. Or for whatever reason a different setting than what the print for your photos is setting. In this case the ICC profile would show you a good softproof because it does it based on its measurement but during printing the profile it applies corrections for conditions that are not the same as the ones that were used to print the test chart.
But take this with a grain of salt. Not sure whether this is actually true.