1. What about "ICC compliant" V2 monitor profiles which "correctly"
record D50 in the wtpt tag, though monitor white is != D50, but which
are so old that they don't contain a CHAD tag yet?
In this case, such profiles would have identical rel. colorimetric and
abs. colorimetric intents. But that was the expected behavior.
2. What about Graeme's printer profiles, if the printer was not
characterized for D50, but for a different illuminant? My understanding
v4 is very clear on that. A chromatic adaptation should be used
to mimic measurements under D50, and then the chromatic
adaptation tag should be added in order to specify the original
illuminant.
is, that Graeme's profiles generally (regardless of the deivce type)
record the absolute color of media white (under the actual viewing
illuminant) in the wtpt tag, and in order to establish the tables, he
transforms the measurements (as measured under the actual viewing
illuminant) with a Bradford CAT from absolute media white to D50.
This is the fatal flaw of v2. Media white point is used to store both the
illuminant and the media white point. These are separate things. One is
the illuminat, i.e. light source used to do the measurements. Other is
the media (paper) white, which should be, as any other patch, measured,
adapted to D50 using CAT and then recorded as media white tag.
But that's the v4 procedure, on v2 I think Graeme is right, and the
method he is using is the less destructive one.
In case (1) the profile does not contain enough information to compute
the "real" absolute colorimetric intent, since no CHAD is present.
Right. For that reason the abs. colorimetric you are obtaining is same as
rel. colorimetric.
In case (2) a "real" absolute colorimetric intent is possible (using the
same strategy as for the 1998 sRGB profile), but an illuminant relative
ICC-absolute intent is not possible, because no information is present
in the profile how the absolute color of media white can be decomposed
into the illuminant and the relative media color (i.e. relative to a
perfect diffuser).
Right. Because the media white is lost. But lcms will not do that because
this is a printer profile, and in this case lcms will do an abs.colorimetric by the
white point scaling method. I take this decision after examining a lot
of printer profiles. They expect the CMM to do so, I suspect mostly because
PhotoShop is doing so.
For both cases, the user would IMO need to supply additional information
(e.g. either the CHAD, or at least the illuminant (assuming Bradford)),
in order to make all kinds of transformations (media relative,
illuminant relative, "real" absolute) possible (with correct results).
Or just use v4 profiles. ;-)
Regards
--
Marti Maria
The littlecms project.
www.littlecms.com
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.323 / Virus Database: 267.9.2/55 - Release Date: 21/07/2005
-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
_______________________________________________
Lcms-user mailing list
Lcms-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lcms-user
