Re: colorsync-users digest, Vol 3 #411 - 8 msgs
Re: colorsync-users digest, Vol 3 #411 - 8 msgs
- Subject: Re: colorsync-users digest, Vol 3 #411 - 8 msgs
- From: Graeme Gill <email@hidden>
- Date: Mon, 23 Sep 2002 16:58:26 +1000
Chris Murphy wrote:
>
It's really a CMM problem as one can't expect ICC profiles, which
>
aren't programs but rather just data, to really do anything. They
>
aren't responsible for conversions today. It's the CMM that does it.
The distinction between data and program is a matter of degree,
not kind. It would seem possible to create a profile format
that more easily facilitates better gamut mapping, either relying
on the CMM doing more work, or even encoding it in table form,
much as the current ICC format does. One approach might be to
standardize a "Profile Connection Gamut" for instance.
>
Imation's CMM actually did do dynamic gamut compression based on the
>
source and destination spaces. So it wasn't to the level of
>
sophistication where all pixels were analyzed first, let alone in
>
context. But nevertheless it did allow for dynamic gamut compression
>
using conventional ICC profiles.
This is what I meant by a "non-conventional" use of ICC's. Rather
than letting the choice of pre-determined tables provide intents,
the CMM does its own gamut mapping, and uses the tables purely
for device characterization data.
Graeme GIll.
_______________________________________________
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.