> all colors can be specified with light wavelength measures isn't that true? 
> can't it be that instead of RGB color you say
> light color wavelength instead?

Not at all. There are lots of coloursthat are not equivalent to that
of visible light of some single wavelength. Just think of purple.

The term "colour" as used here means "colour as perceived by a human
with normal colour vision" . Without referring to some animal's
perception (or technical sensor's), the term "colour" is meaningless,
and what actually exists in the physical sense is a spectrum of
(visible) light wavelengths.

Only so-called spectral colours (which are a very limited subset of
the colours we can perceive) correspond to a specific wavelength of
visible light. Read up on colour perception. For instance, start with
http://en.wikipedia.org/wiki/Color ,
http://en.wikipedia.org/wiki/Spectral_color ,
http://en.wikipedia.org/wiki/Color_vision ,
http://en.wikipedia.org/wiki/CIE_1931_color_space and
http://en.wikipedia.org/wiki/Purple .

Ideally and simplified, for single largish fields of uniform colour,
"idealistic" colour models as RGB or CMYK are as far as I know
equivalent. But that is not what usually is meant when talking about
"CMYK support" in software like GIMP. CMYK is used to describe colours
in images as printed by actual physical processes on paper. In that
context there are lots of very arcane and small-scale additional
details that affects how the image end up looking. Think of issues as
how well different inks can cover each other, how inks spreads onto
the paper, what the colour of the actual paper itself is, how much ink
can be printed before the paper cannot absorb any more and the ink
starts to smear or whatever, etc. I am really no expert in this, but I
do know enough that I understand this is not anything trivial;)

Gimp-developer mailing list

Reply via email to