What do you get when you mix red and blue paint?

December 23, 2013 2:00 AM

David Briggs' The Dimensions of Colour, a comprehensive online explanation of traditional (what you've probably been taught) and modern colour theory, and its applications to visual art. Invaluable for artists and non-artists alike. (The answer: probably some kind of brown. Yes, your kindergarten teachers fed you lies.)
posted by Quilford (28 comments total)

Well, the last time I mixed red and blue paint, I was out on the ocean, and we all ended up marooned.
posted by GenjiandProust at 2:02 AM on December 23, 2013 [10 favorites]

More seriously, I can't wait to dig into this. I'm moderately interested in color and the difference in perceptions between projected and reflected light, but I don't know very much solid about it. So, yay!
posted by GenjiandProust at 2:04 AM on December 23, 2013

Ah, I just realised the post might imply that mixing red and blue paint gets you David Briggs' The Dimensions of Colour. I accept no responsibility for any loss or damage, whether direct or indirect and however caused (including through negligence) that you may suffer as a result of this misunderstanding.

Would be pretty cool though.
posted by Quilford at 2:08 AM on December 23, 2013 [1 favorite]

I always got some sort of weird dark puce color.


Then I got upset.
posted by louche mustachio at 2:30 AM on December 23, 2013 [4 favorites]

And there it is, that exact weird dark puce.


It's good for shadows on pale skin tones sometimes.
posted by louche mustachio at 2:32 AM on December 23, 2013 [1 favorite]

Thanks for this, it justifies the vague sense of uneasiness I always had that what they taught about colour in art class didn't fit with what they taught in science.
posted by Proofs and Refutations at 3:56 AM on December 23, 2013

Handprint has always been my goto site for color theory.
posted by euphorb at 6:06 AM on December 23, 2013 [4 favorites]

I've been looking for something similar to this. Thanks for posting.
posted by uraniumwilly at 6:31 AM on December 23, 2013

Not only is this site pretty much the ONLY exhaustive color theory science resource for painters, but David Briggs is an active contributor on conceptart.org and will answer most any questions you have about the content.
posted by overeducated_alligator at 7:11 AM on December 23, 2013

Many thanks for this resource. I spent three enjoyable hours this morning combing through The Dimensions of Colour. David offers a wonderfully comprehensive overview of colour theory.
posted by Schadenfreude at 7:25 AM on December 23, 2013

This explains why my "purple" eggs looked so funny this past year when I just used food coloring rather than buying the Easter egg dye kit.
posted by rabbitrabbit at 9:08 AM on December 23, 2013

thanks for this very interesting post, i was hooked from the first sentence and can't wait to explore.
posted by OHenryPacey at 9:14 AM on December 23, 2013

Oh now this is awesome. It starts with the Split Primaries pallette, I was taught this in painting class and I didn't know it had an actual name. I use this whenever I mix pigments, although I do tend to use far too many secondaries like green, and oh how I wish Cobalt Violet wasn't so expensive.

Pigment is a very strange thing. I remember when I went back to art school and had to take Painting 101 all over again, after having taken it almost 20 years prior. The teacher's first assignment was to paint only in mixed yellow and black. I had been painting a long time and had a box full of extra-fine artists-grade pigments, so I started painting away. This looks weird, blackish yellows and yellowy blacks, I suppose she had a reason for telling us to do this. So I'm painting and the teacher walks by and says hey what the hell is this?!? How are you doing that? I'm like, what? It's what you asked us to do. No, come look at everyone else's canvases. So we walked around and everyone else is mixing yellow-black and getting an astonishing range of weird greens. WTF? Yeah, this is the whole point of the assignment. So we went back to my palette and tried mixing up some greens. I could not do it. Then the teacher tried it. She couldn't do it either. She looked at my paint tubes and asked me where the hell I got them. Oh, just my regular art store in LA about 4 blocks from my house, lots of famous painters shopped there, you could get really good stuff I've never seen anywhere else, and cheap. After considerable experimentation (and wasted expensive paint) we concluded that the sickly greens could only be produced with cheap student-grade pigments.
posted by charlie don't surf at 10:30 AM on December 23, 2013 [4 favorites]

I worked closely with the color chemists when I was working as a database programmer for a big house paint company and pigments formulas are really counterintuitive. The chemists all seemed to be half mad scientists and half wizard.
posted by octothorpe at 10:56 AM on December 23, 2013 [1 favorite]

Not only is this site pretty much the ONLY exhaustive color theory science resource for painters,

Handprint.com isn't exhaustive?
posted by sebastienbailard at 11:17 AM on December 23, 2013 [1 favorite]

This is academic nonsense. I spent most of my life mixing colors. Brown is the result of mixing that so-called red and blue in the Grumbacher "double primaries palette" because that "red" is orange. And the yellow component of that orange is the complement (opposite) of purple, and therefore kills the purple. There is nothing there resembling a primary red, and you won't get it by mixing any of those other colors, so they are not true primary colors. A true subtractive primary red and primary blue will give you effortless purple. If you don't get that result, you haven't got primaries. Add primary yellow to your palette (and white for lightening the colors), and you can mix virtually any color you want. Primary red, yellow, and blue in the right proportion will also give you a black darker than black construction paper. But don't tell painters this if you want to sell paint.
posted by weapons-grade pandemonium at 11:36 AM on December 23, 2013

A true subtractive primary red and primary blue will give you effortless purple.

Please tell us where to purchase these true primaries.

This is academic nonsense.

And the pot asserted that the kettle had a Chroma and Value of 0 in the Munsell color system.

I am currently using some CMYK watercolors in a printmaking process. Theoretically you could mix "any color" from those pigments. But nobody paints in CMYK.
posted by charlie don't surf at 12:46 PM on December 23, 2013 [3 favorites]

Chuck Close did...
posted by newmoistness at 1:09 PM on December 23, 2013

Violets never solves anything.
posted by Chitownfats at 1:18 PM on December 23, 2013 [3 favorites]

it justifies the vague sense of uneasiness I always had that what they taught about colour in art class didn't fit with what they taught in science

Unless they were teaching you about mixing paint in science class, I think the difference would be better explained by the fact that mixing colored light and colored pigment is different. If I mix all those colors of paint I will not get white, for example.
posted by Hoopo at 2:13 PM on December 23, 2013

I am currently using some CMYK watercolors in a printmaking process. Theoretically you could mix "any color" from those pigment


I'm only familiar with CMYK from printers, but doesn't CMYK work rely on the illusion of those colors rather than actually creating new colors? Like, if you look really closely at a number of the colors produced by CMYK there's small dots of one color on top of other colors?
posted by Hoopo at 2:21 PM on December 23, 2013

but doesn't CMYK work rely on the illusion of those colors rather than actually creating new colors?

My understanding is that color is what you perceive, and spectrum is something that photons do. So any successful illusion is in fact color. For example if you see orange on your computer screen, the color is orange but the spectrum is in fact red & green.

That said, CMYK, RGB, YBR, and even any single spectra have their own unique blindspots. One of these hours I'll read the linked article and be sure.
posted by tychotesla at 2:42 PM on December 23, 2013

That's true; I suppose what I meant was that with CMYK printing you can actually see the dots with the naked eye, whereas when you mix paints to create a color it's pretty much uniformly that color (except maybe under a microscope or something, I'm not sure)

Also the computer screen thing is colored lights...different ballgame
posted by Hoopo at 2:51 PM on December 23, 2013

I think the difference would be better explained by the fact that mixing colored light and colored pigment is different. If I mix all those colors of paint I will not get white, for example. Exactly this! As a projectionist I wish I could get back the hours I spent explaining to corporate clients that I can NOT project black! In projection, black is an absence of light, and even though the screen we are projecting on is white, trust me when I focus light on it in certain areas and not in other areas the "not in other areas" will appear to you as black, if the areas I am focusing light on are bright enough. Which BTW is the real reason movie theaters are dark.
posted by HappyHippo at 8:56 PM on December 23, 2013

Brown is the result of mixing that so-called red and blue in the Grumbacher "double primaries palette" because that "red" is orange.

The point is that those reds and blues are psychologically pure reds and blues. Mixing those psychologically pure colours gets you brown when in fact you'd expect to get purple.
posted by Quilford at 1:58 PM on December 24, 2013

Brown is dark orange. That you don't think of it as such comes from the fact that it's along the skin tone axis, which we tend to parse differently.
posted by effugas at 3:10 PM on December 24, 2013

Brown is not dark orange. I make brown out of Cadmium Yellow and Cobalt Violet. But you can make browns out of many mixes, you can also make a reddish brown or a greenish brown (that one is a bit tricky) and many other variations. Most art students get nothing but browns, until they learn how to mix paint. It's really easy to get crappy browns from mixing 3 primaries. We used to call those browns "painting 101 brown" or "calf shit brown." There are umbers and ochres and ambers and sepias, transparent browns, opaque browns, even brownish blacks. Brown is about the most damn complex color on the palette. Many Old Masters paintings are executed almost entirely in brown, with only smaller areas of bright colors or white. Colors were expensive, browns were cheap as dirt. Browns were made from dirt. It is easy to make a color pop by simultaneous contrast when you've got a relatively monochromatic color scheme. Some schools of academic painting use ébauche or imprimature techniques to lay down a transparent layer of earth tones like umber or sepia and the entire canvas is executed in earth tones until the final colors are laid on top. Other methods add umbers and ochres to nearly all colors, to give them a common basis.

And do not get me started on flesh tones. OK too late now. I read one esoteric book that recommended mixing flesh tone from white, to which you added red, yellow, blue, and black, to represent "the 5 races." What balderdash. But I tried it on a whim, and found it works surprisingly well, if you are careful. But ultimately there is no single flesh tone. Look at a work by a master portraitist like John Singer-Sargent. Now look really close. Look at those colors. On her right cheekbone, purples, and green above and below her lips making the pale red lips pop. There is a slight blue tint under her left eyebrow, and on the left side of the neck. That is how you do flesh tones. And this painting is huge, about life size. Oh how I wish I could paint portraits like John Singer Sargent.
posted by charlie don't surf at 5:32 PM on December 24, 2013 [5 favorites]

Oops, that "really close" link does not seem to work at high rez. Let me try that again. If a mod would care to change the link and delete this msg, that would be nice. A longer edit window would be nice.
posted by charlie don't surf at 5:41 PM on December 24, 2013