Re: Profoto
Re: Profoto
- Subject: Re: Profoto
- From: Roger Breton <email@hidden>
- Date: Sun, 2 Dec 2001 15:46:07 -0500
Steve,
I have always thought that Chromatic Adaptation is not used when the
colorimetric data is in D50 to begin with, such as when measured on a
Spectrolino or a 938 using D50/2 observer. I believe that all that
really goes on, in this instance, is behind the scene normalization to
the media white to make the colorimetry "media-relative". That is what I
understand gets encoded in the profile to make it "relative" -- the PCS
uses relative colorimetry.
If the data comes in at any other color temperature from the instrument,
then, on top of a normalization to the media white to make things
"relative" in the first place, there is also an extra chromatic
adaptation burden to bring the data into D50 which is the ICC Reference
White.
Does that make sense?
Roger Breton
Laval Qc
< clip >
The main issue that can occur with this is the chromatic adaptation
algorithm used. When a profile is originally built, the measurements
are - by definition - absolute. That is, the white point is the paper
white point. When the tables are calculated the paper-dependence is
removed from the numbers through chromatic adaptation calculations.
When the profile is used by the CMM, it needs to choose an adaptation
algorithm that matches the one used in the original calculation of
the profile if you are to get the best results. The current ICC spec
does not have a mechanism to note the method used though, so the CMM
is left guessing. The newer, not yet released version of the spec is
to include a tag which allows this to be recorded. Then the profile
can be used in a more appropriate fashion.
Regards,
Steve
wqeqwe