Martin, With respect, your blanket claim that it is " *impossible* to separate images with saturated shadows using Photoshop and GRACoL” sounds a little exaggerated. Can you elaborate on this Earth-shattering discovery? Your one example is insufficient without more facts. For example… • What GRACoL profile was used for that job - Adobe’s "Coated GRACoL 2006 (ISO 12647-2:2004)” or the one from IDEAlliance; “GRACoL2006 Coated1v2”? • What rendering intent was used? • How much editing had been applied after CMYK conversion? If you had to “rescue” bad separations, my bet is they were simply destroyed by the archaic process of retouching in CMYK, after conversion, rather than the infinitely better, faster and safer process of retouching in RGB prior to conversion. However to be sure of my facts, at your suggestion I just made some comparison tests of the above-mentioned GRACoL profiles from IDEAlliance and Adobe, compared to two “ISO Coated" profiles (one in LinoColor (ECI 2007, 340 TAC) and one in basICColor (2007, 300%). Incidentally, I’m told that beginning these profile names with “ISO” (an obvious attempt to suggest they are “ISO-standard”) was a breach of ISO’s copyright rules. Anyway, I converted my RGBXPLORER target (www.hutchcolor.com/Images_and_targets.html) from Adobe RGB (1998), Adobe ACE, Relative Intent, BPC. I then viewed the resulting CMYK images via their own embedded profiles in Adobe’s relative colorimetric soft proofing mode (View > Proof Setup > Custom… > Preserve CMYK Numbers, Simulate Black Ink). I also converted each CMYK file to Lab (Absolute) and measured very dark colored areas, near the black point. My tests showed that both visually and numerically, the IDEAlliance profile (GRACoL2006 Coated1v2) performed equally well, if not better, than the other three profiles. Apart from the inevitable loss of RGB saturation due to press gamut limitations, nothing “faded to grey” with the IDEAlliance profile that didn’t suffer equally with the other profiles. On the contrary, the IDEAlliance profile produced stronger shadow detail near the black point than the Adobe profile or either of the so-called “ISO Coated ...” profiles. Apparently this is because IDEAlliance profiles are made with a maximum black setting of 100%, while all the European profiles I’ve seen (and Adobe’s profiles) stop at about 90-93% K. FYI, in the days of enlarger separations, limiting the black separation to 90-95% was essential in order to keep black dot gain from killing shadow detail. It was also necessary on most scanners (especially Crosfield) due to their limited color computer algorithms. But in a modern ICC profile created from real press characterization data in good software, limiting K to anything less than 100% just limits maximum shadow density and detail. Anyway, if you’d like to defend your (somewhat incredible) claims about Photoshop and GRACoL, I’d be happy to participate with you in some clearly-defined off-line testing. After 47 years in color separation, I’m always anxious to learn new tricks. ........................................................ Don Hutcheson, President HutchColor, LLC Washington, NJ USA don@hutchcolor.com M: 908-500-0341 ........................................................ On Feb 28, 2015, at 13:03 , Martin Orpen wrote:
Europeans already have a default — ISO Coated v2 (300) — which is well established as being safe and reliable when you’re uncertain about the destination device.
I had to rescue a US ad campaign recently that was totally destroyed by your suggested Photoshop default because It is *impossible* to separate images with saturated shadows using Photoshop and GRACoL.
Give it a try. Make some rich dark blue, purple and red swatches in RGB and see them fade to grey in GRACoL.
When the clients complain, send them to my shop where Photoshop is never used for CMYK ;-)
-- Martin Orpen Idea Digital Imaging Ltd