Blue + yellow, theory and mixing

[WFMartin]

Originally Posted by sidbledsoe

There are thread observers like me who would prefer seeing examples of real world pigments that do exist and not all the photoshop examples, what filters do when they are combined. and laser light attributes. I know I will probably receive a lecture on why it is so pertinent, have at it.

Here ya' go--real-world pigments that do a pretty good job of demonstrating the fact that Blue and Yellow do not mix to produce Green, but are, as practical application, and "theory" both suggest, complements of each, other, and mix to produce neutral.. The Yellow paint on the left, mixed with the Blue paint in the center, created the Neutral paint on the right.


These are actual oil paints that I mixed, with a bit of white (I wanted a resulting "gray," rather than a jet "black". This is an un-retouched, un-color corrected photo of this resulting mix. The reasoning behind the result is quite understandable. There are only three colors of light, and they are Red, Green, and Blue. In this case, the Yellow absorbs, the Blue (because they are complements), the Blue absorbs both the Red and Green, leaving no color being reflected at all. Black (or neutral)_.

However, the presumption that for every tubed color there is another tubed color that just happens to be an exact complement (will mix to produce neutral) is rather wishful thinking, and does not usually work out that way. One of the colors usually has to be mixed, in order to be a true complement of the other.

[Einion]

Split off from this thread:
http://www.wetcanvas.com/forums/showthread.php?t=566534

Einion

[Einion]

Bill, you'll recall that you posted that swatch previously here; there's no need to upload an image a second time, you can use the same link. And while that near-neutral is a creditable result it is only the significant amounts of white that drop the chroma enough for this to approach grey; it would be great to see the result of a mixture of the base yellow and blue, sans white.

Originally Posted by WFMartin

However, the presumption that for every tubed color there is another tubed color that just happens to be an exact complement (will mix to produce neutral) is rather wishful thinking, and does not usually work out that way. One of the colors usually has to be mixed, in order to be a true complement of the other.

While I'm all for promoting the idea of often/usually having to mix to create neutrals - because that's what most palettes require - I have to point out that there are ample numbers of single-pigment mixing complements, as has been mentioned a number of times previously (including in the thread linked to above). With blues and earths one can work well with multiple others, in either direction, which is handy.

It is true that most people don't prioritise palette construction around this principle, however it is entirely possible to build one that consists largely, even entirely, of mixing pairs; one of the simplest is the six-colour secondary palette (CMY + RGB).

Originally Posted by WFMartin

There are only three colors of light, and they are Red, Green, and Blue.

*cough*

Too oversimplified!

Einion

[Patrick1]

Originally Posted by WFMartin

Here ya' go--real-world pigments that do a pretty good job of demonstrating the fact that Blue and Yellow do not mix to produce Green, but are, as practical application, and "theory" both suggest, complements of each, other, and mix to produce neutral.

Bill, why not just say that (unlike just about every other artist), you choose to define blue as the color/hue of a paint or paint mix which is a mixing complement to a given yellow pigment? This perennial dispute is not a matter of pigment mixing, it's a matter of arbitrary color name definition.

[gunzorro]

Patrick -- Thanks for pointing that out. The "blue" does look like a PB (purple-blue) bordering on a P (purple), so it would be a natural compliment of pale yellow.

Even if it isn't a true blue, the color make some terrific combinatons.

[Einion]

Originally Posted by gunzorro

The "blue" does look like a PB (purple-blue) bordering on a P (purple), so it would be a natural compliment of pale yellow.

Well in theory at least... in practice I'd put money on every version of French Ultramarine there is and any light yellow making a green at the midpoint between the two, especially since you get green even with some yellows that are way toward orange too.

Einion

[Patrick1]

Originally Posted by Einion

... in practice I'd put money on every version of French Ultramarine there is and any light yellow making a green at the midpoint between the two, especially since you get green even with some yellows that are way toward orange too.

Ah, but then it's because your blue is not really blue. Try it with a real blue.

[Einion]

Or with a pure blue, or a true blue... we've seen those too.

Regardless of the wording they amount to the same thing, a No True Scotsman argument.

Einion

[rltromble]

I think the problem with using this as example is that Bill didn't actually use just a yellow and blue per se. Its a mixture of Fr. Ultramarine and Permanent rose (Q. violet) and Transparent yellow. Which is a blue, reddish purple and yellow. So yes you are going to get either a brown, a green or a neutral.