Sam Berry wrote:
On 27 September 2010 14:18, Graeme Gill <graeme@xxxxxxxxxxxxx> wrote:Good point. I will fix dispcal and spotread so that the color temperature correctly takes into account a non-standard observer (I overlooked this in adding an alternate observer).I can't immediately think of any easy way to do it. I've had to calculate D65 XYZ from scratch using the D65 spectrum. To generalise it would seem to be an awful lot of work. Perhaps a warning that CCTs and VCTs are no longer valid would suffice?
It was actually trivial to fix, because rather than use one of the approximation formula for computing CCT, I do a numerical search for the temperature at which the locus spectrum has minimum delta E to the measured XYZ. (This is the definition of a CCT.) So all I had to do was change the code so use the selected selected observer, rather than 1931 2 degree for computing the XYZ of locus spectrum. Graeme Gill.