Re: Is this a scum dot?
Re: Is this a scum dot?
- Subject: Re: Is this a scum dot?
- From: bruce fraser <email@hidden>
- Date: Sun, 17 Nov 2002 16:40:29 -0800
At 3:58 PM +1100 11/14/02, Graeme Gill wrote:
bruce fraser wrote:
If the scanner profile translates RGB 255,255,255, relcol, to
something less than L*100, it is by definition broken.
I'd expect it to typically translate it into something
more than L*100 (assuming it wasn't clipped in the
mechanics of the ICC profile) - the paper color of a test chart
is not usually 100% reflective, and the scanners would
normally have some margin to allow for reflectance up
to or beyond 100%.
In the research lab, that's quite true. In the real world, though,
you immediately run into a couple of problems.
As you noted, few ICC implementations, and none of the mainstream
ones found in applications that we use every day to produce real
work, support overrange encoding. LAB is bounded in L* to 0-100 and
in a* and b* to -128 to 128.
As a consequence, when we scan profiling targets, we generally take
considerable pains to avoid scanning the white of the target at 255
or the black of the target at 0 (particularly if you're scanning
transmissive rather than reflective).
But the key point that color scientists tend to overlook is that the
goal in image capture, whether from a scan or from a direct digital
capture, is rarely if ever to reproduce the original (if you're going
to print, you'd have to violate the laws of physics to do so). Images
just about always need to be interpreted by humans to reconcile the
dynamic range and gamut of the original with the almost invariably
smaller dynamic range and gamut of the output. Given that, it's
generally useful to have capture RGB translate to L* 100 so that you
can actually get paper white without wasting bits on overrange
encoding that the PCS doesn't support. If a lower value translates to
L*100, you waste bits. If Capture RGB 255 translates to something
other than L*100, you get a scum dot.
Absolute colorimetric capture generally produces hideous results.
When we look at prints, we adapt to paper white. Translating white in
the capture to anything other than paper white clobbers the
already-limited dynamic range and introduces the appearance of a
color cast.
We don't get paid for producing colorimetrically-correct results. We
get paid for producing color that looks right.
Bruce
--
email@hidden
_______________________________________________
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.