Is the color wheel wrong?

[graesea]

If you have time to read this post about color wheels, I'd really appreciate any feedback. The author claims that traditional color wheels are wrong and puts up some interesting supporting evidence.

http://blog.asmartbear.com/color-wheels.html

Thank you for looking.

[Horsa]

Very interesting reading. One point not mentioned in the article but shown in one of the illustrations is why "day-glow" yellow-green is so "bright". It is the wavelength of colour light that the human eye and brain are most sensitive to.

Some emergency vehicles are now being painted this colour instead of the traditional red, which is actually a less visibility colour.

[Tiasa]

There are many colors wheels, depending on what you use them for. There is a color wheel for light beams, a color wheel for transparent inks, a color wheel for vision , and a color wheel for opaque paint mixing. For paint mixing, I prefer the 12-color wheel where red is the complement of green; blue is the complement of orange; and yellow is the complement of purple. Even with a color wheel, paint mixtures don’t always turn out as predicted. The color wheel is still just an approximation.

[yellow_oxide]

From what I've seen there's two related problems with various color wheels-

• Human color vision is not symmetrical. It simply isn't. There's certain segments of the visible light spectrum that we see with greater ability to determine subtle hue shifts than others, specifically the greens. Yet most color wheels I've seen are obsessed with symmetry, giving equal portions of it to each color. It's a neat, clean, orderly view of the world of color, but the reality is that various segments of what we consider one color category or another within the visible light spectrum are of various sizes. Consider yellow, for example. Yellow is a very narrow color. You don't have to go much to the left before it's called green, or much to the right before it's called orange. You can't even go very dark with yellow either before no one would look at it and still say it's yellow. Green, on the other hand, is very wide and has a lot of varieties that are all still called by the name of green, even if it's so dark that it's nearly black or so light it's nearly white.

• Color wheels that I've seen are often based more in idea than in the actual realities of paint. I know there's a difference between paint (subtractive) and light (additive) since I do both paint and digital art, but I'm just talking about paint here. Paint based color wheels that have been supported and taught by artists themselves for generations just don't match the realities of what happens when two paints are mixed. To me, color theory is like a mountain of secrets. Small bits of it are revealed each time two pigments collide. Whenever the realities of paint differ from the expectations of color theory the paint is always right, and the theory is always wrong. The idea phase of color theory is actually very interesting and important, but for someone to stop there without testing the theory is like a scientist who comes up with a hypothesis and declares it true either without testing or regardless of contrary test results. To me, to learn color is to mix paint. While the theory was a guess, the paint reveals what's real.

I skimmed through the article and the color wheel recommended at the end is basically the Lab one that I use in digital art, although even it seems to have some occasional shortcomings (or at least photoshop's implementation of it does).

It's worth noting that Leonardo's primary colors were these, in this order- white, yellow, green, blue, red, black. [link]
It's kind of interesting that he seems to be noting a duality between white/black, yellow/blue (earth/air), and red/green (fire/water), which is basically the same as the Lab color space.

[llawrence]

Yeah, color wheels are abstractions, and you can't really call them "right" or "wrong." They're useful to one, or they're not. I've made a Zorn palette color wheel to help plan skin tone mixtures. Was that color wheel "wrong"?

However: the author's big point, at least at the beginning, is that red, yellow and blue aren't really the fundamental mixing primaries, and he's absolutely right about that. It's amazing how many artists left comments about how wrong he is. Huh!

[sidbledsoe]

There is no rule that says that colors must be equidistant in a circle from each other with regards to their physical attributes.
This illustrates one of the problems with over theorizing and trying to make things "fit" into a pattern that doesn't really exist.
It doesn't matter to me which colors are the proper primary colors.
If you want to mix the widest range of colors from the fewest three, then CYM does that best. But now put them equidistant from each other in a circle. With his CYM color wheel, opposite colors are not subtractive mixing complements. Put RYB in the same place and then the opposite colors are mixing complements. His first RYB wheel works good and thus is "right" as a practical complementary mixing reference but his CYM wheel does not work in this manner so in this respect, it is "wrong".

[Gigalot]

http://www.youtube.com/watch?v=09sM7Y0jZdI

[sidbledsoe]

Another reason CYM is good for injet printers is that there are very cheap and very strong and very transparent versions available of these colors making the range even more versatile.
There is no rule that says there has to be three primary colors. However, by the classic definition, Yellow is obviously a primary color, there is no mixing combo of any others that will make it. That is why it is common to both RYB and CYM. It is unique in this respect. However you can mix a R,B,C, or M color from others. Mixed colors may not be as bright and saturated as single pigment colors, but that is true of mixed oranges, greens, and violets also.

[sidbledsoe]

However the arrangement of CYM in an equidistant circular manner does appear to make the visual complements opposite from one another. I say appear because a visual complement is more of a subjective phenomenon. So in that case, the CYM wheel is right and the RYB wheel is wrong.

[yellow_oxide]

Originally Posted by sidbledsoe

However, by the classic definition, Yellow is obviously a primary color, there is no mixing combo of any others that will make it.

I've found this to be false, in a post on my blog about testing a palette made of just green (phthalo, blue shade), orange (indian yellow, a slightly yellowish orange), purple (dioxazine), and white (titanium) and seeing if I could get the traditional primaries of red, yellow, and blue out of them.

I'll post the image here so you won't have to go there, unless you really want to.



The cadmium orange I didn't use, and the manganese "mineral" violet I was just trying out. As you can see in the bottom right corner I did successfully produce a bright middle yellow mixing phthalo green with indian yellow. Because the green was so overpowering I had to keep mixing more indian yellow into it and went a little too far at the end, but the second to last mix is what I'd call a middle yellow. It wasn't as high chroma as a cadmium lemon, for example, but was still fairly bright.

I would say there's no mixing combo that will reproduce any especially high chroma pigment with results of equal chroma, whether it's a traditional primary or not, but in terms of the ability to mix a full range of hues you can choose as your own primaries pretty much any three (or more) sufficiently spaced pigments. The more the better, as long as it doesn't get unmanageable and as long as you don't have something like 10 really low chroma pigments and 1 especially high one, which might be harder to harmonize (but not impossible).

I don't do the whole having one palette that gets used on every painting thing like a lot of people seem to do. Instead, my primaries are every pigment on my palette, and these change from one painting to the next.

[birdhs]

color wheels are teaching tools.

any mechanic will tell you that sometimes more than one tool may do one job, it all depends on many factors.

The type of thinking that leads you to believe color wheels are wrong or right will leave you as you continue to paint over the years.

try to stay open-minded and not believe everything you read.

greg

[sidbledsoe]

Originally Posted by yellow_oxide

As you can see in the bottom right corner I did successfully produce a bright middle yellow mixing phthalo green with indian yellow.

whoa hoss, phthalo green? try it again without white and see what you get, it should be very green.

[sidbledsoe]

also, if you start with an intense indian yellow color, yes, you can easily neutralize the red orange component in it with a blue green and be left with the predominate yellow color only grayed somewhat and tinted out to a large degree by the white. I would try it again and start without a yellow color to begin with.
But you know, when you mix a red from magenta and a blue, you are really doing the same thing, neutralizing the contaminants and leaving the red component that was already there to begin with. Maybe red, yellow, and blue really are primary colors, I don't know.
When I look at colors I can see the other colors in them, that is, if they are there. For instance, when I look at magenta, I can see the red clearly, but I can detect that violet in there too. When you look at cyan, it kinda looks like a blend of greenness and blueness combined. But look at yellow, red, blue, and green.
These colors have a unique identity to them. They don't look like a blend of anything else in existence. Yes green is in that group of unique looking colors, even though it is a blend of blue and yellow, but it looks like neither one at all. You can't say that about violet or orange, they both have the look somewhat of the two colors it would take to mix them.
I am not the first to notice this phenomenon by the way, I in fact read it somewhere back there somewhere, and I agree that I can see it.

[yellow_oxide]

I tried briefly, but couldn't get it to work with the very few other oranges I have in oil paint. Just cadmium orange, quin gold GS (PO49), and quin burnt orange (PO48). I really need to get a few more, but I don't use it as often as other color categories. The reason I used that "indian yellow" is because it's nearly as orange as my cadmium orange, but is a single pigment instead of a PO20+PR108 mix.

What makes yellow unique, in my understanding, besides being one of the four basic colors that our vision is based on, is that it's able to be relatively high chroma at high values. Like I said before, it may not be possible to reproduce the chroma of a high chroma pigment by mixing others. In the case of yellow, the problem is reaching that level of chroma in the upper reaches of value.

Red, green, yellow, and blue are special as far as our vision is concerned. In terms of the Lab color space we see colors as an answer to three questions-
How dark/light is it?
How green/red is it?
How blue/yellow is it?
Value is on a scale of 0-100, while green and blue are both negative numbers and red and yellow are positive. So something like this-
L=80
a=-40
b=20
would be a light yellowish green.

[DeBro]

Everytime I have come across a discussion about the color wheel, it is always centered around which color wheel is best for mixing this or that as though this is the sole purpose of using it in art. Some people even outright dismiss the color wheel when it comes to painting with pigments.
My primary use of the color wheel is to select and use the individual hues to build color chords for use as color schemes in my pictorial designs (abstract or representational). For a time, I used the Munsell wheel, but has since switch to the CMY wheel because in my opinion it is the most accurate in representing the hues of the color spectrum in a circle.
I'm currently constucting a 24 hue CMY color wheel using only single pigment paints, excluding the Cadmiums, available today in the acrylic medium. First, I use the computer program DrawPlus to create the 24 hue CMY color wheel in the CMYK color space that it provides. I also bougth The Color Company's CMY color wheel from Dick Blick. The hues of both the virtual and physical wheels matched up fairly well. So, using these two wheels as references, I began constructing my CMY color wheel with pigments. Any mixing that was taking place was to use two single pigment paints to precisely match hues of the reference wheels and in some cases tinting with white to achieve spectrum value. For example, the range of four analogous hues after primary yellow and ending at green, I used mixes in varying percentages of Hansa Yellow Light (PY3) and Phthalocyanine Green - Yellow Shade (PG36) which gave me relatively high saturation hues. I have documented all these mixes with the estimated percentages of the pigments used and I'll post them in this forum along with a photo of the color wheel when I'm satisfied that it's done.

I recently posted an image of an abstract painting in the Acrylic forum which I copied and pasted below featuring what I've achieved so far in my construction of the CMY color wheel. The color scheme is based on six analogous hues ranging from Cyan to Blue Violet. Each hue is presented at varying values relative to an eleven step value scale. This shows that the color wheel has more than one use. Size 46'' x 46''