[argyllcms] Re: negative primaries

  • From: Elle Stone <l.elle.stone@xxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Fri, 11 Jun 2010 05:50:04 -0400

On 6/10/10, Graeme Gill <graeme@xxxxxxxxxxxxx> wrote:

> Given a bunch of real cameras, I wonder how common it is
> for the cameras spectrum locus to fall outside the standard
> observer spectrum locus, when the matrix has been created by
> minimizing the (say) sum of squared errors of a representatively
> weighted basket of real world color spectra.

The "standard observer spectrum locus" means the horse-shoe-shaped
outer boundaries of the CIE 1931 color space chromaticity diagram? And
any negative primaries in the camera matrix are definitely outside the
standard observer spectrum locus?

The 360+ camera matrices used by dcraw, mostly but not all supplied by
Adobe in the form of "camera coefficients", were presumably created
from target shots by "minimizing the sum of squared errors" (there is
code in dcraw for doing such from a ColorChecker chart, but Dave also
takes it8 target shots for new cameras).

I turned the the dcraw camera coefficients into camera matrices. They
all have at least 2 negative primaries, a few have 4 negative
primaries. All of them have negative green Z primaries (average
-0.32588). Most of them have negative blue Y primaries (average

If anyone is curious, you can download my dcraw camera matrix
spreadsheet from
These matrices are relative to D65, not D50, but having done the
chromatic adaptation for a few matrices, the value change is not that
great - the negative numbers are still going to be negative.

Elle Stone

Other related posts: