View Full Version : GCR ICC profile

12-15-2010, 07:16 PM
I am trying to generate a GCR ICC profile through ColorFlow, but I am having difficulty. Something I must be doing wrong.

First, I created a new CMYK Reference device.

Then filled in the information.

Then import IDEAlliance C3 dataset in Measurements.

Then, under the "Profile" little icon, the one that's just below the little "i" icon, I specify :

Origin = ColorFlow generated;

Total Ink Limit = 100;
Black Start = 0;
Max black = 100;

Black strength = Heavy (75%).

I click OK and export the resultant ICC profile.

Alas, it's not really a strong GCR profile.

What am I doing wrong?

Best / Roger

12-15-2010, 07:29 PM
How are you doing your analysis of the profile to come to the conclusion that it is not "strong GCR"?


12-15-2010, 09:07 PM
David, regardless of how I arrive at my conclusion (I sent you some docs offline), could you tell me whether the way I went about creating that GCR profile makes any sense?

Best / Roger

12-15-2010, 10:13 PM
Makes sense except for 'Total Ink Limit = 100;"

I'm assuming you meant 300, as that is how your profile is encoded... and well, that value makes more sense ;). In looking at the two profiles that you provided, "SWOP2006_C3 300 k100 G100.icc" is created from different measurement data compared to that of "SWOP2006_C3 GCR Heavy par ColorFlow (15 déc 010).icc". I extracted the measurement data from "SWOP2006_C3 300 k100 G100.icc" and created the profile with TAC=300, Black Start=0, Black Strength=75 in ColorFlow. I will send you this profile for your review.


12-16-2010, 02:28 PM
Sorry for the TIL type, 300 was the value I meant to write.

With regards to the "different" dataset, I can assure you that the "SWOP2006_C3 300 k100 G100.icc" was created from the same IDEAlliance dataset that yours was created from. I can extract the dataset from that profile here, as well, and will be taking a look at it, as a curiosity. It's possible that a slightly different A2B1 tag resulted because of some new, interpolation technique. But I doubt it. Last time I compare this encoding to ProfileWizardMio encoding, I got a difference of less than 1 deltaE with a max of 1.5 between the two profiles.

Best / Roger