[argyllcms] Re: dcraw outlined, annotated, and un-Dnged

  • From: Elle Stone <l.elle.stone@xxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Sat, 13 Feb 2010 16:16:19 -0500

On 2/13/10, Hening Bettermann <hein.be@xxxxxx> wrote:

> This looks like a great project to me.
Thanks! It's been a lot of work.

> "I use linux. This code, as is, won't compile under windows."
> Then it won't under MacOSX either, I assume?

Can you compile and run dcraw downloaded from
http://www.cybercom.net/~dcoffin/dcraw/, that is, the "dcraw.c" file
right under the note to linux distributors? If so, then my modified
code should also compile and my statement about "not compiling under
windows" needs to be corrected.

> Under 0.3.1 GOALS: you write
> "(1) Un-Dng dcraw, That is, replace the Dng-derived camera coefficients with
> camera coefficients derived from Argyllcms colprof matrix profiles"
>
> Is it restricted to *matrix* profiles? I get much lower dE values with LUT
> profiles.

Internally, dcraw uses matrix profiles. You can always use "-o 0" and
apply the LUT profile with cctiff or tifficc or most any editing
program (in which case my modified code is not necessary, as "-o 0"
bypasses the dcraw internal profile application process). However, if
you search the argyll mailing list, you'll find reasons why a lot of
people end up preferring matrix profiles for digital cameras.

> In any case, your text is already working on me with goal (2): "Gain a
> better understanding of what dcraw does when it renders a raw file."
Thanks! I'm working on cleaning up the comments - my writing style is
a bit wordy, while too often jumping right over important stuff.

Regards,
Elle

Other related posts:

  1. Home
  2. argyllcms
  3. Archive
  4. 02-2010