Additive colour theory and "traditional" colour mixing

[keroberous]

Hi folks. I am trying to reconcile the additive colour theory of light with more traditional grade school colour mixing.

For example, we all know that blue paint plus yellow paint makes green paint. How do we explain this from the perspective of light? So the yellow paint appears yellow because it reflects green and red light and blue paint appears blue because it reflects blue light and if we mix the yellow and blue paints together it should reflect red, blue, and green light so should appear white. But obviously it appears green so I'm clearly missing something but have not been able to find any info about what that something is.

Thanks!

 

[Drakkith]

So the yellow paint appears yellow because it reflects green and red light

Or it just reflects yellow light. That would mean that yellow + blue mixes yellow and blue light, which might add together to appear green to your eye.

 

[mjc123]

It is absorption, not reflection, that is additive. Blue paint absorbs strongly at the red end of the spectrum, but reflects a broad range of colours (not just a narrow "blue"), most strongly blue, but also including neighbouring colours like green and violet. Yellow paint absorbs strongly at the violet end, and reflects yellow most strongly, but also neighbouring colours like orange and green. When you mix blue and yellow paint, the yellow pigment absorbs blue light, and the blue pigment absorbs yellow light, but both reflect green, so the paint looks green. (If you mixed a range of pigments that between them absorbed all across the spectrum you would get black from the sum of the absorptions, not white from the sum of the reflections.)

 

[sophiecentaur]

Hi folks. I am trying to reconcile the additive colour theory of light with more traditional grade school colour mixing.

I remember both forms of colour mixing being described - except in 'creative' lessons where paints were discussed.

It is absorption, not reflection, that is additive.

Yes. In school, we were told about Additive and Subtractive mixing. When you mix pigments or filters, you let through less and less light until the result is nearly always a dark brownish grey. When you add random 'coloured' lights together, the result is brighter and brighter without limit and tends to a slightly coloured white.

 

[Yuriy Zayko]

A few years ago I was working on the problem of color formation. I presented my results in the article:
Journal of Applied Physical Science International, ISSN No. : 2395-5260 (Print), 2395-5279 (Online), 2015, Vol.: 2, Issue.: 4, pp. 137-144.
Below is an abstract
ABSTRACT
In this article issues related to the introduction of the concept of color temperature used in the colorimetric measurements are briefly discussed. An ambiguity of existing definition of color temperature is stressed. An unambiguous definition of color temperature and the technique of its implementation are suggested. Also a physical interpretation of the RGB color model is suggested which introduces the carriers of colors - complexes of photons. The thermodynamics of these complexes is studied. This will help to find new description of color formation which is oriented on human perception but is not connected with any observer.
I would be glad if these results will help in the discussion.

 

[sophiecentaur]

I would be glad if these results will help in the discussion.

Interesting stuff but I'm not sure how that Abstract is of use to this discussion. Illuminant is, of course, an important factor in colorimetry but the level of this thread is not deep enough to include in a simple additive / subtractive mixing discussion.
Is your paper available on line (for free)?

 

[Merlin3189]

mjc has pretty well answered your question.

What I wanted to add, is that you shouldn't put too much faith in these popular notions about colour (colour triangle and mixing), even when discussed by experts. Colour is a very complicated subject, which is much more of a psychology topic than a physics one.

Some issues which often cause confusion:
Two lights which appear the same "colour" may be physically different, being made of different combinations of several (many) spectral colours.
Paints, pigments, coloured objects appearing the same "colour" may similarly be reflecting different combinations of spectral colours. They may look the same under one illumination but different under another.

The "colour" we ascribe to something we see is very much a psychological question rather than a physical one. Apart from the extreme case of physical anomalies, which affect about 8% of males in the UK, everyone's colour perception is affected by factors other than the specific physical composition of the light they are seeing. Context is a big factor: colours can appear different when adjacent to different colours or in scenes with different colour balance.

 

[tech99]

What I wanted to add, is that you shouldn't put too much faith in these popular notions about colour (colour triangle and mixing), even when discussed by experts. Colour is a very complicated subject, which is much more of a psychology topic than a physics one.

I think there is frequent confusion about the entirely psycho-perceptual effect of additive colour mixing. However, the colour triangle, which was invented by Maxwell, was successfully used to design colour TV systems, so must be reasonable?.
I have often thought of making a teaching tool using some RF bandpass filters to simulate subtractive colour mixing using a spectrum analyser.

 

[sophiecentaur]

What I wanted to add, is that you shouldn't put too much faith in these popular notions about colour (colour triangle and mixing), even when discussed by experts.

Rumours of the death of colourimetry are a bit exaggerated, I think. We have a very successful TV and colour reproduction technology which is not based on anything as simple as a "colour triangle". Our vision evolved to deal with many random inputs - varying light levels and illuminants and also surrounding colours. It seems that the standard technology of colour TV can provide 'enough' information with three display primaries to allow the majority of viewers to agree about colour matches over the gamut of colours limited by the primaries. So the tristimulus system can be said to 'work' pretty well (well enough for billions of people to be happy to watch TV). Actually, that's a bit of a cheeky statement because millions of people were happy to watch glubby monochrome TV too!

Of course, we have to remember that the Science of Colours is based on Psychology as much as Physics but, in a thread with the word "colour" in the title, we have to acknowledge the psychovisual content from the start. Additive and subtractive mixing can be made to work very well, probably in spite of their shortcomings.

We exist in a popular culture where colour is still synonymous with wavelength. Things have a long way to go.