I've been changing some minor parts in my LittleCMS wrapper (the one using ctypes), and beating on the bush using my unit tests I just found out something strange.
All statements are based on LCMS 1.1.6. I have changed the default transform creation from using the cmsFLAGS_NOTPRECALC flag to using 0 instead. All of a sudden most of my unit tests choked. Mostly it was just some minor tweaking of +-1 for the 8bit unsigned values. But I've got one test where I run through the 24 colours of the GretagMacbeth ColourChecker, and there I've experienced some major differences between the XYZ -> Lab conversion using a transformation and the cmsXYZ2Lab() function (using built in profiles only). With the cmsFLAGS_NOTPRECALC all tolerances in Lab (for transformation as well as the function) the Delta E value (Euclidean) was < 0.1. Removing the cmsFLAGS_NOTPRECALC flag, smeltering the transformation into a device link, rocketed up the Delta E for a few fields to > 5. I'm aware of the "Annex B Apparent bug in XYZ -> sRGB transforms" in the tutorial, but this is an XYZ -> Lab transform, so the boundary conditions mentioned in that annex don't match up regarding the limiting gamut, etc. So what could be the reason for that anomaly? Are maybe the built in profiles somehow limited in their gamut to give me that? Curiously, Guy -- Guy K. Kloss Institute of Information and Mathematical Sciences Te Kura Pūtaiao o Mōhiohio me Pāngarau Room 2.63, Quad Block A Building Massey University, Auckland, Albany Private Bag 102 904, North Shore Mail Centre voice: +64 9 414-0800 ext. 9585 fax: +64 9 441-8181 eMail: [EMAIL PROTECTED] http://www.massey.ac.nz/~gkloss/
signature.asc
Description: This is a digitally signed message part.
------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________ Lcms-user mailing list Lcms-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lcms-user
