Florian Höch wrote: > Unless I'm misunderstanding what -iaw "Absolute Colorimetric with > scaling to fit white point" does (entirely possible), it seems to clip > output values. Hi, this seems to be a bug with dealing with a destination profile that is a matrix + gamma form. I've updated the V1.7 beta source with a fix (change is in link/collink.c) if that is any help. Graeme.