Re: scanback profiling question (rendering intents)
Re: scanback profiling question (rendering intents)
- Subject: Re: scanback profiling question (rendering intents)
- From: Graeme Gill <email@hidden>
- Date: Mon, 12 May 2003 17:37:54 +1000
Terry Wyse wrote:
>
More to the point, why would there even BE the need for some sort of
>
rendering intent required on the Device->PCS given that all colors in the
>
Device can be accurately mapped to the PCS (Lab) space? What on the device
>
side could be considered "out of gamut" that a rendering intent needs even
>
to be considered?
Because the ICC format defines it this way, not because
it makes any sense (it doesn't).
Conceivably it's possible to work around the fact that
the ICC gamut mapping doesn't work, by defining a standard
PCS gamut extent, and configuring the profile to expand/compress
the devices gamut to the standard in the Device->PCS conversion,
and then compress/expand the standard PCS gamut in the PCS->Device
conversion. Since the ICC format doesn't define a standard PCS
gamut, this clearly wasn't the intention, and even if it had
been, it's hard to think it would work very well without
also standardizing the gamut mapping algorithm as well.
(in order to ensure that the Device->PCS->Device conversion where
the Device is the same is a null conversion.)
The presence of different AtoB tables allows vendors to
add "secret sauce" (perhaps along the lines of adopting
a de facto PCS gamut and manipulating Device->PCS conversions ?).
Such things aren't likely to help profile interchangeability though.
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.