Re: Rendering Intent and Converting to Lab
Re: Rendering Intent and Converting to Lab
- Subject: Re: Rendering Intent and Converting to Lab
- From: Thomas Knoll <email@hidden>
- Date: Thu, 14 Feb 2002 06:03:03 -0800
At 10:49 PM +0000 2/13/02, neilB wrote:
If the accepted definition of Perceptual is to be believed at all, we
understand that it maps the whole source space to the destination space
- perhaps not as elegantly as we once thought - but that's the Gist -
right?
The accepted definition of Perceptual is wrong, and does not match at
all what any profile I have ever seen is really doing.
They do *not* map the entire source space (which the destination
profile cannot even know). They profile creation software just makes
a wild guess about what a typical image's gamut is, and uses some
combination of compression and clipping to make that typical image
look "good".
So, in moving the data to the PCS does it, then, attempt to map the
source space out to the size of the PCS?
No, I have not seen any profiles that expand the gamut during the
source->PCS half of the transform.
In other words, if we coiuld look at the data in the PCS, is it
different data depending on the intent chosen?? Must be - eh?
The Adobe profiles use a different black point for perceptual and
saturation vs. relative colorimetric. So yes, the data is often
different. Using black point compensation in Photoshop tends to mask
this difference, however.
Thomas Knoll
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.