-------- Original-Nachricht -------- > Datum: Thu, 20 Dec 2007 14:34:57 +0100 > Von: Thomas Lunzer <[EMAIL PROTECTED]>
> What was actually > surprising me, is: > Why would the argyl cmm produce such a different result from the Adobe > cmm, allthough > I used perceptual intent in both cases.(Note: I experimentally omitted > the f and k flag, to be > shure this is not affecting the result) > Do you have an explanation for that behaviour? The results of which transformations exactly did you actually compare? And how was your printer profile created? Note, if you convert e.g. with Photoshop RGB to CMYK then the CMM basically just applies the tables for the selected intent in the given profiles. The origin of these tables is not the CMM, but the profilers, who created the profiles. But Argyll's icclink, as it is used here (with the -G option), creates a new gamut mapping and also CMYK separations _from the scratch_ (i.e. only using the A2B1 table of the given printer profile). It does not make use of the perceptual B2A0 table in the supplied profile. So it is not surprising, if the resulting device link differs from the transformation which you get by simply concatenating the A2B0 transformation of the source profile with the B2A0 transformation of the destination profile. Also note that perceptual intent is an issue of art. You can't really judge numerically whether the result is right or wrong, but you can rather only judge subjectively, whether you find the result pleasing or not. Expect to get different result from different profilers (and icclink in gamut mapping mode acts rather like a profiler (or "smart CMM"), and not as "dumb" CMM which just applies precomputed transformations recorded in the given profiles). Regards, Gerhard -- Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Lcms-user mailing list Lcms-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lcms-user
