🚀 go-pugleaf

friedric@kirk.fmi.uni-passau.de (Carsten Friedrich) writes:

>
>now this is what i would like to know:
>
>- is it right that cmy and rgb are even in theory not complementary ?
>
>- if no, what is my mistake in the text above?
>
>- if yes, what is the correct way to convert from additive rgb to subtractive 
>cmy and vice versa ?
>

Subtractive colour mixing is not linear. To calculate the resulting colour
when you overlap two inkts, as is done by a printer, you have to know the
whole spectrum of the colour without overlap MULTIPLY this by the spectral
density of the overlapping inkt and calculate the corresponding RGB values
using the colour matching functions.

It is in principle possible to calculate the resulting RGB values from the
CMY(B) values if you know the RGB values of each resulting colour in an
N-colour system, so N=7 or 8 (no inkt) in your case. The equations are known
as the Neugebauer equations. Unfortunately these equations cannot be inverted
and secondly the CMY space is a subspace of the RGB space. For example a
saturated colour with a high luminance cannot be created on paper. I give the
two references that I have:

1. Die theoretische grundlagen des mehrfarbendrucks
   H.E.J. Neugebauer
   Zeitschrift der wissens. Photog., band 36, 4, pp.7-89, '37

2. Theory of masking for color correction. I masks drawn from the subject
   H.E.J. Neugebauer
   Jour. Opt. Soc. Am., Vol. 42, pp. 740-7, '52

There were some papers in a SPIE conference on this subject, but I do not have
the reference at this moment.

Hope this helps,

Rob Beuker
Philips Research Laboratories
Eindhoven
The Netherlands