Color Wheel vs Color Spectrum

[ifixpix]

Hi Folks- I am a newbie to Retouchpro-(Not to Photoshop, however)

There seem to be a few math (and maybe science) geeks out here so I figured I'd toss this out to you.

The color spectrum seems to be a linear model with ever decreasing wavelengths as you climb up the spectrum.

Red-Yellow-Green-Blue-Violet

However, we can predictively add colors with a "wheel" model.
It seems intuitive that Yellow and green would produce Blue-the average of the 2 wavelengths of energy.


But.....how come mixing Red and Blue does not produce some average of those wavelengths? (a yellowish-green)

Instead, it produces a Magenta (Violet-like?) color.

It's as if our eyes and brains are wired to see a linear physical reality (the spectrum) as if it were joined at the ends so that it forms a circle.

Does this seem strange to anyone else or should I just "let it go?"
Cheers
Bob

[Ed_L]

I understand where you're coming from, Bob. But I don't have an answer for you, so I'll just say "Welcome to RP".

Ed

[bart_hickman]

The color wheel works because the human eye RGB cones aren't quite R,G, and B. Very roughly-speaking, the blue cones see from about violet to green (with a peak in the middle of blue.) The green cones see from about blue to red (with a peak in green). The red cone sees from green to red (with a peak in the yellow/orange range), but it also has a passband up at violet. This latter property is key because it means the red cone essentially completes the circle by sensing the shortest wavelength violets in addition to reds thus overlapping with the short-wavelength side of blue.

So violet light stimulates both the red and blue cones and you interpret that as violet. For slightly longer (bluer) wavelengths, the red is no longer stimulated and you see blue. If you alternatively shine red + blue light, it also stimulates the red and blue cones and you see violet. This is why you (the human eye) can't tell the difference between true spectral violet and a mixture of red and blue. This is also why digital cameras will see true violet as blue. Ie they can tell the difference between pure violet and red/blue, but they can't tell the difference between pure violet and pure blue.

I'm sure we've all taken photos of purple flowers or (more rarely) clothing only to have them turn out blue. This is because the flower was spectrally pure violet.

Using your digicam, you can tell if something is pure violet rather than blue + red. If you saw the item as purple and it came out blue, then it was spectral violet. If the purple item came out looking purple, then you know it's really a mixture of red and blue. I've seen flowers ranging from absolutely pure violet to some combination of violet, blue, and red. In the latter case, the flower comes out looking about right.

I assume camera manufacturers don't put a violet passband on the red color filter in digicams for reasons of manufacturing difficulty. It's possible that the highest-end DSLR's might incorporate this, but as far as I know, they don't.

Bart

[byRo]

...did somebody here call a Geek?

OK, so let's say it like an electron. Can be thought of as a wave or as a particle.
Nah, just kidding....

But, really, the answer is something like that. Have a look at this diagram.
http://www.efg2.com/Lab/Graphics/Col...romaticity.htm

The numbers around the outline are wavelengths. So you can see that it is both linear, and circular at the same time.

The Red-to-Blue "workaround" can be seen as the straight line joining the two end points.



Rô

[ifixpix]

Thanks Folks-Great explanation Bart!
Now if only we were wired so that we could perceive the other 8 dimensions...nah..... too confusing...

I wonder what evolutionary purpose this pass filter serves.. nature is fond of circles

[ifixpix]

...did somebody here call a Geek?

OK, so let's say it like an electron. Can be thought of as a wave or as a particle.
Nah, just kidding....


Rô

Maybe it's a bunch of particles that form themselves into a wave pattern.

[byRo]

a bunch of particles that form themselves into a wave pattern.......

. . .
. . . .
. . .

exactly!

Rô

[bart_hickman]

Groan!!!

a bunch of particles that form themselves into a wave pattern.......

. . .
. . . .
. . .

exactly!

Rô

Uuuuugggh!

Bart

[Peter S]

You asked for that!