[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math

View post   
File: 145 KB, 500x500, 1283751950225.jpg [View same] [iqdb] [saucenao] [google]
1943696 No.1943696 [Reply] [Original]

There's one thing I never understood about color.

If you look at an orange shirt, you see orange for one of 2 reasons. The first reason is that the shirt has been dyed with an orange pigment. The second reason is that the shirt has been dyed with comparable amounts of red and yellow dye, and the frequency of orange light is the average of the frequencies of red and yellow light.

If you look at an green shirt, you see green for one of 2 reasons. The first reason is that the shirt has been dyed with a green pigment. The second reason is that the shirt has been dyed with comparable amounts of blue and yellow dye, and the frequency of green light is the average of the frequencies of blue and yellow light.

However, if you look at purple shirt, a purple dye remains valid, but a mix of red and blue dyes does not seem valid, even though it works in real life. The frequency of purple light is not the average frequency of red and blue light. In fact, purple is on the other side of blue from red.

So why does red dye+blue dye=net purple dye?

>> No.1943702

Because this concept of averaging frequencies is wrong. See http://en.wikipedia.org/wiki/Color_perception for details.

>> No.1943721
File: 34 KB, 400x516, 1283646656453.jpg [View same] [iqdb] [saucenao] [google]
1943721

>>1943702
Ok...Red stimulates the short roads, blue stimulates the medium rods and the long rods a bit, so why does mixing them together stimulate the long rods without stimulating the other rods much?

>> No.1943734

>>1943721
It doesn't, but the brain doesn't care much about the excitation of the individual rods. The *ratios* between the excitation levels determine the perceived color, and two completely different spectra may produce the same excitation ratio.

>> No.1943737

>>1943721

>stimulate long rod

heh

>> No.1943739
File: 49 KB, 192x171, 1272446085992.jpg [View same] [iqdb] [saucenao] [google]
1943739

>>1943737

>> No.1943741

>>1943737
The usual term is "excitation" instead of "stimulation", allowing for even more direct alternative interpretations.

>> No.1943747
File: 34 KB, 500x389, 500px-HA_HA_HA,_OH_WOW.jpg [View same] [iqdb] [saucenao] [google]
1943747

>>1943734
Silly, silly me. Asking a complex question on /sci/ and expecting a simple answer.

Alright, to clarify further:
Purple dyes reflect a frequency of light that stimulates the long rods more than the other rods by a certain ratio.
Blue and red dyes, when thoroughly mixed in the right portions, reflect a combination of frequencies that stimulate the rods in the same ratio as seen in purple dye, even though Red would only stimulate the short rods on its own, and Blue would stimulate long and medium but mostly medium rods on its own. Why?

>> No.1943752

24 October, 2010. The day /sci/ ruined neuroscience for me.

>> No.1943762

>>1943747
I'm not sure about this, but I suspect the color purple is a color the brain made up to deal with the apparently contradictory situation in which the short and long rods are excited but the medium ones aren't. That means the color purple doesn't correspond to any specific wavelength - in other words, pure violet light wouldn't be interpreted as purple at all. http://en.wikipedia.org/wiki/Spectral_color#Non-spectral_colors seems to somewhat agree with me.

>> No.1943778
File: 178 KB, 571x570, 1285551245283.jpg [View same] [iqdb] [saucenao] [google]
1943778

>>1943762
So...then there's no such thing as purple dye, dye that reflects both red and blue?

Alright......now I get it

>> No.1943789

>>1943778
There is purple dye that reflects red and blue. What does not exist is purple dye that reflects the color frequency purple, because there is no such thing. According to my theory, anyway.

>> No.1943810
File: 49 KB, 571x556, 1284038850740.jpg [View same] [iqdb] [saucenao] [google]
1943810

>>1943789
that's what I was saying, but I forgot to put 'just' before 'dye that reflects both red and blue'

>> No.1944456

You must think I'm a total square, but you guys might find this interesting.
http://www.biotele.com/magenta.html