View Full Version : blue plus brown make purple

03-14-2002, 06:41 AM
With paints they don't, but with additive mixing they seem to. Up close you see squares of blue and brown. From a few feet back, it's purple. Very surprising considering what it's made from.


The only explanation I can think of is that this
pointillism works like additive mixing, and that
brown is just a dark orange (or a dark, desaturated orange...maybe the same thing here), and orange is close to the additive primary of middle/light red. So the brown here is behaving quite similarly to red. And with lights, red plus blue gives purple or violet.

A neat practical application of this I thought of is that if a painter didn't have any purple on his
or her palette, but did have say ultramarine or cobalt blue and burnt sienna, they could use a pointillist technique to make a purple. But so far in my one little experiment (printed the pattern out and held it under a bright light), it didn't work.

Could it be that it just isn't as bright and saturated as on the computer screen? I might try it with real paints later on...

03-14-2002, 06:52 AM
:) It does look purple. :)

I was reading in the April issue of The Artist's Magazine inside an article about Purple and Green making blue. Gosh, it is amazing what our eyes do and how we "READ" color.

In this article, it talked about using watercolors - Viridian and Winsor Viloet to make a blue. They did an exercise using only these two colors and the final product really did look blue.

I think more than anything, doing exercises like the one that you just put up, Domer, and the one in the Artist's Magazine, reading about how color is perceived is a learning tool for all of us to understand just what our colors can do for us.

Thanks for showing this, Domer.

03-15-2002, 03:06 AM
that's because you're warming up the blue.....{M}

03-15-2002, 02:04 PM
I mentioned a while back in another thread started by Patrick about Phthalo Green BS and Quinacridone Violet mixing to a very good dark. In masstone it's almost impossible to see that it's not in fact black but in washes or tints you can see clearly that it is a dark, slightly dull blue, something like Indanthrone Blue or Indigo.

This sort of oddball mixing result is a very good illustration of why traditional RYB colour theory is so poor in practice as it can't account for mixes of this kind, much less predict them!


03-17-2002, 01:31 PM
You have to remember that there is a big difference between additive mixing and substractive mixing.

Things accomplished in one mode cannot be directly duplicated in the other, without changing the colors involved in the mix.

In short, this is because with additive mixing, what you are actually mixing is "light" (i.e. the RGB model we use in the computer world). With subtractive mixing, you are mixing pigments, which contain light absorbing particles (i.e. the opposite of additive mixing).

The ultimate resource, IMHO, on color theory is http://www.handprint.com.


03-17-2002, 01:52 PM
Yellow, red, plus blue make brown. The three traditional primaries. Very little blue is used, and most browns lean toward red.

Traditionally, red plus blue make purple. Various blues mixed with various reds will get various types of purples. So, it doesn't surprise me at all that a visual effect of mixing brown among blue gets us a purple feeling to it.

The yellow in the brown mix formula...hmm..well, think of it as mixing with the red to form the orange side of the brown. Orange is the opposite of blue and thus mixed as it is here acts as a neutralizing agent. In other words...it either heightens or tames the sense of "blueness" depending on amount that is mixed together.

Squint your eyes at this block of color you have here, and you'll more than likely detect a low chroma of this purple block of color...which is the neutralizing of the oranges against the blue going on. Darker, yes..but somewhat greyed down.