Color - Notes and Advice, Mostly for Oil Painters

[Horsa]

Originally Posted by Keene

[i][
Physical Properties of Color

Subtractive primaries: magenta, yellow and cyan; yellow+magenta = red, magenta +cyan = dark blue or violet, yellow + cyan = green
Additive primaries: primaries = red, green and deep blue; yellow = red + green, magenta = red + deep blue (violet), cyan blue = green + deep blue (violet).

The above may be true of wavelengths of light, but I have yet to see a set of "magic" paints or pigments where yellow=red+green.

"Colour" is not a physical reality. It is a social construct. The are cultures that do not make a different ion between what we would call "blue" and "green" for example. In these cultures they are the same "colour". What we call colour is the brain's interpretation of neural impulses from photosensitive cells in our retinas. Colour does not exist until the brain interprets the impulses.

The seven colour rainbow spectrum that we all learn is an artificial division of the wavelengths of visible light, crafted to fit a "magic number". Newton added the division of "indigo" between blue and violet because the spectrum needed to have seven colours. Look at a spectrum from a prism and tell me how many distinct bands of colour you "actually" see.

Colour theory is a great example of the dangers of conflating theory with reality. Try to achieve some of the effects that colour theory tells us we should obtain from pigments. Then try the same with colored lights. Does mixing the here primaries give black or white? Can you actually add red and green to produce yellow?

Why does a tree painted grey, and elephant painted brown, or a river painted brown look "wrong"? These are the actual most common local colours of these objects.

White is hotter than red, but blue is warmer than straw yellow. At least to a metal smith. Blue flame is considerably hotter (warmer) than yellow or red flame, just ask your local welder.

As for "wonky, easily disprovable colour theories that produce great art" is the theory really wrong if it's application leads to the desired result?

[Keene]

Just for clarification for anyone who doesn’t know, additive color refers to light. Pigment mixing is subtractive. The mixing properties of light are quite different from those of pigments. In light red + green = yellow and in pigment these two colors are generally considered complimentary and each dulls (subtracts from the intensity) of the other.


The book Art and Vision: the Biology of Seeing, by Margaret Livingston, PhD, has a good discussion of current (2008) theories of the perception of color. I also like the website Handprint (http://handprint.com/HP/WCL/wcolor.html) and of course, WetCanvas


Keene

[AllisonR]

I absolutely love your outline. So clear, so concise. I work this way as well, suffling through all the fluff and making a list of what is important.

I think it is very helpful and useful information. As others posted, this is a bouncing board, not rules written in blood. Take what you like, disregard the rest. FWIW, I am an acrylic painter. The information is useful for that medium as well. Especially if you paint in an "oils" style - value underpainting, hue afterwards....

[Keene]

Allison,

Thanks for your kind comments. I’m sure the notes are equally applicable to acrylic, but much is omitted that would be important when working in watercolor. On my own website, I call the notes posted here simply “Color”. I kept notes on watercolor separate from those on oil, but haven’t posted them.

Keene

[Gigalot]

I`ll try translate this post to other language for some painters who don`t speak English because , I think, it is useful for all of them.

[Patrick1]

Originally Posted by Horsa

As for "wonky, easily disprovable colour theories that produce great art" is the theory really wrong if it's application leads to the desired result?

It depends if the idea is presented as a subjective viewpoint about seeing & using color, or if it`s presented as a fact.

Examples of subjective color seeing & useage: the right color to use for cast shadows is ___, the correct color to use to darken yellow objects is ___, the coolest yellow is ___, the best palette to use in a bright midday plein air painting session, etc. If having a specific viewpoint like this helps you to achieve the color results you want, then it`s not wrong ...as long as it`s not presented as fact.

But I have seen good artists express factually incorrect viewpoints. As an example, one highly regarded artist and painting materials authority (i.e. allegedly an expert) on this site stated that Zinc White does not impart a hue shift when mixed with other colors. You can easily verify or disprove this. Another one you might`ve heard: that lemon yellow is more greenish (and thus mixes cleaner greens) than middle and orange-yellows because it contains more blue. (it`s actually mainly because of a big difference in green reflectance, and much less of a difference in blue reflectance). Lots more examples like this you see & hear all the time by artists that do great work. Yes, it`s possible to do great art holding belief(s) that actually are wrong.

[Patrick1]

Originally Posted by Horsa

The above may be true of wavelengths of light, but I have yet to see a set of "magic" paints or pigments where yellow=red+green.

...

Colour theory is a great example of the dangers of conflating theory with reality. Try to achieve some of the effects that colour theory tells us we should obtain from pigments. Then try the same with colored lights. Does mixing the here primaries give black or white? Can you actually add red and green to produce yellow?

Hmmm... not quite sure what you`re getting at here. Unless it`s so obvious I`m missing it. Modern color theory can give an approximate guess as to the results of mixing two pigments or lights of known spectral characteristics. But it`s still just an educated guess because (with paints) there are so many physical variables that it can`t account for. So the only way to know for sure in the end is to mix them. But in most cases I doubt a well-educated guess will be wildly off.

[Gigalot]

The rule Red+Green mixture must become much duller then Red or Green itself always works. Even if you have an ultra-saturated fluorescent pigments - the result will be an ocher looking thing.

[Keene]

For those who are interested, I have posted on Pinterest 20 paintings by various artists emphasizing color. I add and delete paintings as I see fit, but maintain the number at 20. I also have posted 20 by contemporary artists, 20 compositions that I like, 20 by masters and 20 of mine.


Keene