CMYK Custom Settings - Adobe Color Management

Are you talking about things like, for example, making a version of US Web Coated (SWOP) v2 with higher black generation? And/or profile editing functions within Photoshop?

Many professionals are clamoring for the inclusion of those features. I doubt that Adobe will offer those capabilities in Photoshop any time soon, though. But I've been know to be wrong once or twice before... ;-)

Marco,

I think it's more of a confusion in that so many people think that they already can edit the Adobe v2 profiles simply by clicking over to Custom CMYK. It would be nice if Adobe would add a feature to do that, but I don't see that happening anytime soon. I'd love to be wrong though.

As I'm sure you already know, for those already owning ProfileMaker, it's pretty easy to download ColorLab and use that to reverse engineer a data file from the Adobe profiles, drop that into ProfileMaker and build a new profile with whatever black generation or ink limits you desire. I've done this, and they seem to work just fine.

And/or profile editing functions within Photoshop?




Yes. Photoshop is, after all, the main weapon in our arsenal in preparing files for press. Or at least it's mine. Adobe has added many features in the last two releases that made photographers and designers happy. How about throwing a crumb to the people on whose backs the app was built?

Yes, I have done that too. Actually, in MeasureTool you don't even need to use ColorLab to prep anything: just open ProfileMaker, select a testchart in the Reference Data section of the dialog box (say "ECI2002CMYK i1.txt"), then, in the Measurement Data section, choose "Import ICC Profile" and select the profile you wish to create a modified version of (be it with higher GCR, or no black, or lower black start, or different black width, or lower TAC, etc.).

I am not sure about the "reverse-engineering" part. I am pretty sure that one cannot extract from a profile the measurement file that was used to create it (certainly not the spectral data), if it is not already inside the profile itself in one of those discretionary tags that the ICC allows for, like the ones that Gretag uses. But profiles like US Web Coated (SWOP) v2 offer nothing of the sort.

Marco,

That works with any Gretag profile, as the measurements are stored in the profile itself, but it won't work on the Adobe profiles. That's why you need to use ColorLab to convert the test chart to Lab using AbsCol. That then gives you the Lab data that the profile produces from a known test chart. It's not 100% accurate, but in my tests and others like Terry Wyse, it's more than good enough, with perhaps a one percent change here or there.

With Gretag profiles, you don't even have to open Measure Tool. Open ProfileMaker itself and drag the profile you want to rebuild into the bottom window and it automatically loads the correct reference data.

So if I do the "import ICC profile" and pull in the SWOPv2 one Photoshop is using I can then set the separation settings in Profilemaker and create a new profile with those settings that would be based on the SWOPv2? How does it work if the spectral data isn't in the SWOPv2 profile for it to compare to the ECI2002CMYKi1.txt? Does it just use the color data that's already in the SWOP profile and only changes the black, TIL, etc?

Peter,

My message posted before I saw your last reply. Any further, more precise instructions on how to do what you're describing with ColorLab?

CopyCraft out of Lubbock TX is one of those printers that still instruct their customers to use CustomCMYK settings of SWOP coated, 15% dot gain, GCR, Light, 100, 300, 0 to convert their RGB files to and adjust their monitor to match their prints. They have a sheetfed Komori Lithrone L-628 press, color balanced using a closed loop scanning spectro densitometer. Woof!

Their catalog is printed on glossy aqueous coated text stock and includes a print of the untagged Ole No Moire CMYK file included on Photoshop install CD's. They burn the CMYK numbers contained in this file unaltered and unconverted straight to plates so it acts as a by the numbers test target.

Under a loupe the dot percentages on the print are very close to the numbers in the CD test file and the print matches to screen assigning the v2 US Sheetfed coated profile to the CD test file. If I assign v2 US Web the CD test file looks too light compared to the print. Roughly what this indicates is their color balanced press acts as a CMYK-"What Numbers Are There Is What Prints" constant similar to RGB minilabs that maintain a stable linearized sRGB appearance in their prints excluding slight drift inherent in all printers.

Anyway, I scanned the CopyCraft print in AdobeRGB and got a very close visual match to print opened in PS viewed side by side with the original CD test file. I did two conversions on the scan-the CustomCMYK as instructed by CopyCraft and the v2 Sheetfed Coated. Guess which one gave nearly identical numbers to the CD test file-the v2 Sheetfed Coated-right down to black generation readouts.

Converting the scan to CopyCraft's CustomCMYK turned CD test file numbers like 5C,28M,32Y,0K-(the forehead area on the model) to 8C,42M,42Y,0K and black readouts nearly doubled in the gray canvas backdrop area in the image. Why would they want me to convert my files this way when it's obvious it would produce dark muddy seps?

Here's how close CopyCraft's 300lpi print <http://www.pixentral.com/show.php?pi...YUgX873FWkpGE1 > was to the original CD test target.

Sorry, I meant ProfileMaker, not MeasureTool.