RE: relative colorimetric without black point compensation
RE: relative colorimetric without black point compensation
- Subject: RE: relative colorimetric without black point compensation
- From: email@hidden
- Date: Tue, 27 Jan 2009 16:38:36 -0500
On Sun, 25 Jan 2009 11:32:39 -0500
Colorsync-users Digest, Vol 6, Issue 10
Roger Breton wrote:
>...in coming up with "true" perceptual mappings to the PCS,
>any profiler has to make some kind of assumptions about the Source
This is precisely one of the major weakness of ICC V2 profiles which V4
profiles were intended to remedy. In V2 there was no defined Black Point
(BP), whereas in V4 the PCS has a defined BP as well as the Perceptual
Reference Media Gamut whose bottom is said PCS BP. Given a conforming V4
input profile, either RGB or CMYK, and a conforming V4 output profile, my
sense is that a V4 Perceptual transform has the potential to produce a more
pleasing result than RelCol transform with BPC.
Black Point Compensation, when going from a high dynamic range input space
to a lower dynamic range output space, is scaling the Y channel in the XYZ
colorspace. Its tantamount to shining D50 light on the source image so
that its darkest point matches that of the output; BP chromaticity doesn't
change (its still D50), white point is unaffected. As Graeme Gill mentions
in his email, BPC does not explicitly address off-axis colors. Off axis
colors are affected (a* b* channels of CIELAB colorspace not preserved),
some may become desaturated and some may indeed incur hue changes. This is
not to say that RelCol w/ BPC result is not pleasing - it is, but much of
that is derived from the CMM knowing both source & destination and thus
being able to do BP mapping. But its my gut feeling that a good V4
Perceptual transform can do as well if not better than BPC in maintaining
colorfulness. V4 Perceptual transforms can explicitly control what occurs
off axis as well as apply a more optimal tonal compression customized by
the profile builder.
Harold Boll
Adobe's paper on BPC: www.color.org/AdobeBPC.pdf
TOSHIBA AMERICA BUSINESS SOLUTIONS 2 Musick, Irvine, CA 92618
Harold Boll | 1.781.856.5174 | email@hidden
" The sunlights differ, but there is only one darkness. "
Ursula K. Leguin
This message (and any attached files) is secret, confidential and proprietary and is intended solely for specific addressee(s). If the reader of this message is not the intended recipient, please notify the sender immediately by reply email and delete it and all backup copies immediately.
_______________________________________________
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