Re: The CIECAM challenge in ColorSync
Re: The CIECAM challenge in ColorSync
- Subject: Re: The CIECAM challenge in ColorSync
- From: Graeme Gill <email@hidden>
- Date: Tue, 24 Jan 2006 10:55:40 +1100
Marco Ugolini wrote:
In a message dated 1/23/06 2:31 AM, Graeme Gill wrote:
CIECAM02 may well be a component in a smart CMM, but to be recommending
CIECAM02 when there's nothing to apply it to, is rather putting the cart
before the horse, IMHO ....
What do you mean by "nothing to apply it to"? CIELAB is being applied now.
Why wouldn't CIECAM perform a similar function?
Within a conventional CMM, CIELAB is being used purely as a color
representation and interpolation space. The viewing conditions
are completely fixed (D50, ICC viewing conditions), and there is no
translation whatsoever between different viewing conditions, or gamut
mapping going on, or gamut clipping going on. All that stuff has been
pre-computed, and stored in the profile tables at profile creation
time. Hence the expression "Dumb" CMM. All the "smarts" were done
elsewhere. Even the interpolation is on such a fine scale (within each
cell of a CLUT), that the difference between CIELAB and any other
perceptual space would be very minor (and would be contrary to
the correct interpolation of the ICC profile in any case).
Even Adobe's black point compensation is done in XYZ space
(which is the wrong space to do it, but that's an issue for
another day...)
The ICC has been steadily working to eliminate the guesswork of
what maths the CMM is meant to do, and the maths doesn't involve
anything where CIECAM02 could be applied.
Graeme Gill.
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden