The physics/physiology of colour mixing

[brianvds]

Suppose I mix some bright yellow and bright blue paint. My eyes will usually perceive the mixture as green in colour.

Questions: If I analyze the light from the mixture with a spectrometer, does it show me that the mixture is emitting light in the green part of the spectrum, or does it show peaks in the blue and yellow? How come my eyes perceive it as green? (I think I can come up with a an answer for the second question, but not too sure about the first!)

[Rigil Kent]

The visible light spectrometer will "see" the mixture of pigments exactly as would your eye: less absorption in the green region, and more in the blue and red extremities. Pigments appear colourful because they subtract from the colour components of white light, so it's helpful to imagine starting off with pure white light made up of all the colours of the rainbow.

A bright lime yellow pigment will absorb poorly in the yellow and green region of white light. All the other colour components (red, orange, blue, violet) are properly absorbed and hence subtracted from the white light.

Similarly, the blue pigment may absorb poorly in, say, the green and blue region, while it subtracts red, orange, yellow, and violet.

When the two pigments are mixed, the mixture will absorb (subtract) red, orange, yellow, blue and violet from the full spectrum white light, while absorbing only poorly in the green region. Visually (and spectrophotometrically) we see a dull green.  



Some pure pigments (such as the phthalocyanine greens) are inherently, well, green. They can give brighter shades of green than a mixture of primaries.

Rigil

[BoogieMonster]

I'll admit this did once throw me for a loop in art class because I was used to coding RGB values for a computer. Which determines which colours of light to mix to come up with a certain hue.

IOW: When working with light the process is additive, but when working with pigment it's the other way around: Each pigment you add absorbs more of the spectrum and you see what's left.

Put differently once more: Mixing all the colours of light together results in white. Mixing all the pigments together results in black.

[brianvds]

In other words, the reason why a mixture of blue and yellow will give you green is because most physical pigments will in fact not absorb absolutely all the other colours apart from themselves. They will tend to leave some of their adjacent colours. Presumably, if you had absolutely "pure" blue and yellow pigments and mixed them, the result will be very dull and dark.

While painting myself, I have noticed that mixes of actual real paints as opposed to idealized ones can sometimes give surprising and counter-intuitive results. For example, in these two pictures no blue or green paint was used; the greens were obtained by mixing a bit of black paint into bright yellow:





And here is an example by Aaron Coberly, a classically trained artist. Once again no blue or green paints; the entire picture was painted using only four colours, namely black, white, red and yellow:

[cr1t]

And here is an example by Aaron Coberly, a classically trained artist. Once again no blue or green paints; the entire picture was painted using only four colours, namely black, white, red and yellow:

WITCHCRAFT!

[Mefiante]

Presumably, if you had absolutely "pure" blue and yellow pigments and mixed them, the result will be very dull and dark.

Correct, if by “pure” you mean “near-perfect,” provided the pigments’ colours are properly complementary in terms of their hue, saturation and intensity.  In fact, in that case you’d get something very close to black.  Interestingly, each of the subtractive primary colours (cyan, magenta and yellow) is got from a 1:1 addition of light sources of two of the three additive primary colours (red, green & blue): cyan = blue+green, magenta = blue+red, yellow = red+green.

'Luthon64

[Rigil Kent]

In other words, the reason why a mixture of blue and yellow will give you green is because most physical pigments will in fact not absorb absolutely all the other colours apart from themselves.

Yep. A nominally "yellow" or "blue" pigment does not have to restrict it's absorption accordingly. Yellows may have either a green or orange undertone. Blues may have violet or green undertones, and reds may have either magenta or scarlet undertones. If you want to mix up a batch of green by blending primary blue and yellow, your selection of pigments will have a big effect on the "brightness" of the resulting secondary. For instance, blending a orange-toned yellow with a violet-toned blue will result in a pretty yucky looking green.

Rigil

[brianvds]

And here is an example by Aaron Coberly, a classically trained artist. Once again no blue or green paints; the entire picture was painted using only four colours, namely black, white, red and yellow:

WITCHCRAFT!

That's exactly how I also felt. Whenever I see people who can paint like that, I feel I could cheerfully burn them at the stake. :-)

[Rigil Kent]

Whenever I see people who can paint like that, I feel I could cheerfully burn them at the stake. :-)

Or why not just embrace neoplasticism?

[brianvds]

Whenever I see people who can paint like that, I feel I could cheerfully burn them at the stake. :-)

Or why not just embrace neoplasticism?

Well, I always thought that if I just painted small enough, no one would be able to see all the mistakes anyway. But even in miniature art I don't need to look far before meeting my match:



Why didn't I take up something easy, like advanced quantum physics, as a hobby? :-)