[argyllcms] Re: Echanging CMF
- From: Roger Breton <graxx@xxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Wed, 17 May 2006 22:52:26 -0400
> A set of alternate CMF's is supported. I haven't gone as far as
> creating a file format (analogous to the illuminant .sp) to
> allow user supplied CMF's (where would they come from ?)
>
> CMF's supported internally are:
>
> CIE 1931 2 degree
> Judd & Voss 1978 2 degree (the "fixed" 1931 2 degree)
> Stiles & Burch 1955 2 degree, rotated to align with 1931 XYZ space, using Mark
> Shaw's matrix.
> Shaw & Fairchild 1997 2 degree observer, from Mark Shaw's Masters thesis:
> "Evaluating the 1931 CIE Color
> Matching Functions"
> CIE 1964 10 degree
> CIE 1964 10 degree - adjusted for better matching to the 2 degree values (this
> has problems)
>
> Not all of these may be available from the command line.
>
> It's worth reading Mark Shaw's thesis for some background on all these.
> See <http://www.cis.rit.edu/mcsl/research/mshaw/CMF_Thesis.pdf>
>
> Graeme Gill.
I just come back from the CIE Jubilee Meeting and I'd like to experiment
with Stiles 195x 2 degree CMF. Stockman presented a paper where he
transforms his recently measured LMS spectral sensitivities to Stiles 2
degrees CMF, saying that the '31 CMF were biased from the get go, because of
the insistance of incorporating V-lambda (and we know that V-Lamda is
crooked in the blue region, for historical reasons). I never understood why
my color match yielded *good* numerical matches but *poor* visual matches
but now, I'm starting to have my doubts. Hence, the need to experiment with
different CMFs. Apparently, all the data about Stiles 2 degree CMF is on the
CVRL site?
I'll have to take a closer look at the CMF argyll's already supported.
You're ahead of the game Grame. But maybe you're going to tell me that it
won't change the visual appearance of the matches from the 1931 2 degre CMF?
Regards,
Roger Breton | Laval, Canada | graxx@xxxxxxxxxxxx
http://pages.infinit.net/graxx
Other related posts:
