The problem with rainbows

I was just asked if I believe in the big bang, and it got me to thinking about red shifting and blue shifting… I generally trust that this is the case, though I’m not personally capable of testing it myself. Yellow shifting should be a stationary object, because this works on the primary colors. The problem with rainbows is that they distribute the primary colors every other and then mix for the middle stripes, except for purple at the end, because there is no last red stripe. And that is the problem with rainbows.

It’s not clear what you mean by ‘problem’. It sounds like you mean “There’s a problem in my understanding of rainbows / light / colour perception,” right? Like…the rainbow doesn’t have a problem. It is what it is, it just exists. The problem exists in your head, in your mis-understanding.

Or do you think the problem actually is with the rainbow, and not with your understanding?

After purple comes ultraviolet. The light you can see is only a very tiny portion of the light spectrum.

Purple is red and blue, using the patten of the rainbow, there should be a red stripe on the bottom.

Purple ‘is’ red and blue. ‘Is’. That’s a funny word to use there.

Some wavelengths of light we see as purple. Some red. Some blue.
Red has a lower frequency and thus higher wavelength than purple.
Incidentally, so does blue.

Which raises the question, how in the hell do we see purple when we mix red paint and blue paint?

Your ideas about how colors work are wrong. But that’s ok, because colors are confusing. Your eyes and brain are doing some very weird and complex things when they process colors. It’s tempting to think such simple things as ‘purple IS red and blue’. It’s much more interesting than that.

Ahh… ok, it was a random thought about patterns… I’ll do more research on it. I think in primitive terms… primary colors mix to make tertiary colors, and primary colors are why we see red and blue shifts, because they’re more fundamental (otherwise we’d see violet shifts)… the pattern of the rainbow suggests overlap from this principle of primary and tertiary colors, so it doesn’t make sense from that pattern who there’s no red stripe on the bottom to make it logically consistent.

Light isn’t a set of finger paint.

When you think really deeply about colors, you’ll blow your goddamn mind. Light, you see, comes in various wavelengths. Wavelengths can be long or short; it is a linear thing. One wavelength is either longer, shorter, or equal to another (although I guess being equal wouldn’t make it ‘another’). So…why does colour go in a circle? How can something that is completely linear…go in a circle? It’s nuts!

reddit.com/r/askscience/comm … ds_of_the/

It’s not a problem in any sense.
It’s a problem with conception, and perception.
The world is not adjusted to humans; humans are more or less adjusted to the universe through evolution. And their conceotion of it will never quite match up to reality.
When you get this in your thick skull all your problems will dissipate.

The universe has no need of ‘primary colours’; that is a human conceit.

Just as a reality check… Flannel Jesus’ post was helpful and not derogatory, I suppose that’s why he’s a mod and you are posters.

He also hasn’t just been trolled by you for 17 pages.
Everyone has a limit to their patience.

Actually he has. He participated in that thread, so if he thought I was trolling, he would have given me a warning.

I think you probably found the answer to that in the link you posted, but I may as well put a quick answer here for whoever may read the thread.

We have photoreceptor cells which peak at red, blue, and green, but these aren’t the only wavelengths that stimulate them, they are each sensitive to a certain range of frequencies.
When you look at purple, both you red cones and your blue cones get stimulated a bit.
So by creating a mix of pigments that stimulates the red cones and the blue cones up to a certain amount creates the visual experience of purple.

So while it is true that purple has smaller wavelength than both red and blue, seeing purple it isn’t really an arithmetic addition of the red and blue wavelengths, it is just red and blue cone cells receiving a certain amount of stimulation which is below their saturation point.

Light is definitely linear. Our perception of color is what goes in a circle.

To make things more complicated, pigments and light are sort of opposite. Primary light colors are pigment secondary colors, and the sum of all is white. Primary pigment colors are secondary light colors and the sum of all is black.

Teachers make us play with paint a lot in school. They really should make us play with light a lot too to avoid creating such a mind twist. The first time I saw the convergence of a red and a green light make yellow my mind was blown… I was totally wondering what weird sort of brown light was going to show.

Also the colors that you see aren’t always the colors that you are looking at. Many things reflect a multiple of wavelengths which give the impression of a color that wasn’t actually represented in the reflected spectrum. Dot matrix printers and LCD display monitors use that phenomenon to produce the perception of a wide range of colors (65,536) that aren’t actually there.

Just for clarity, the (incorrect) intuition isn’t that purple should be the result of arithmetic addition, it should be the result of an average. Kind of like how green can come from yellow and blue; green’s wavelength is between yellow and blue, so it fits the intuition.

And you both are of course right, James and pho, about why that intuition fails: light wavelengths are perfectly linear, but the way our cells in our retina react to them is decidedly not linear. The circularity is in our eyes and brain.

Yeah, I see what you mean… it actually is an average, sort of, but not an average of light wavelength, but an average of photoreceptor stimulation.

[quote=“Flannel Jesus”]
When you think really deeply about colors, you’ll blow your goddamn mind. Light, you see, comes in various wavelengths. Wavelengths can be long or short; it is a linear thing. One wavelength is either longer, shorter, or equal to another (although I guess being equal wouldn’t make it ‘another’). So…why does colour go in a circle? How can something that is completely linear…go in a circle? It’s nuts!

Isn’t that because red is roughly half the frequency of purple? So purple light will “contain” red light, and therefore stimulate the red receptors?