[argyllcms] Re: Different approach to profiling?

  • From: Graeme Gill <graeme@xxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Wed, 05 Sep 2007 10:25:57 +1000

Lars Tore Gustavsen wrote:

With this method, do you have to do the tiffgamut step for every image?

Yes, that's the idea.

icclink: Warning - Image gamut is wrong colorspace for link (Lab != Jab)

Because separate utilities are being used together, there is a little
manual coordination needed. You need to make sure that the colorspace
using for the gamut matches that used in profile for the gamut mapping.
For a lot of intents this is CIECAM02 appearance space ("Jab"),
while tiffgamut output defaults to D50 L*a*b* ("Lab").

So you need specify Jab for perceptual intent, and the same viewing
conditions as you will be using for the source colorspace of the link,
something like:

   tiffgamut -v -ij -cmd ProPhoto.icm Flattened.tif

and the error should go away. (I will change tiffgamut to
automatically switch to Jab space if a viewing condition is
specified, which should lead to less surprise).

Graeme Gill.

Other related posts: