[Awetopsy]

ok.. yesterday I heard somebody say that you could tell what a certain color is in its greyscale form.

So Ive been thinking about it and I did this little experiment...


but first I drew this out to see what I could guess were the colors.


now I originally guessed:
#1 = green
#2 = red
#3 = blue

obviously I was wrong.

i dont really know what the point of this is.. but it seemed usefull to me. maybe you guys already know this but at least I learned something.

Personally it helps me break down the colors that I truly see and helps me understand their values more.

hope this helps somebody else.

[Steven Stahlberg]

I don't know about the exercise, I haven't tried it, I'm sure it's useful if you felt it was. But that guy who said he could tell the colors in greyscale, that made react - it must be wrong. The greyscale will be completely different depending on the color of the light. So is he assuming a constant white color on everything? Not really realistic. Or maybe I just misunderstood what he meant?

Steven

[Frost]

In terms of RGB, the colors have different 'weights'. In a good color-to-grey color conversion algorythm, you'll have green having the most influence, and blue having the least. There was a thread posted about this, uhm... about a year ago { } ... tried looking for it but could not find it.

Maybe Waylon or Suma knows the specifics on this...

[Awetopsy]

well I was thinking about different lights would make a color have a different shade.... but like frost said a light coulndt hide the weight of the color.

maybe I dont undertand this properly...

[IEatLitleBabies]

i made this for craps and giggles, i dont know if it helps or not, but i wanted to see it anyways

sorry its big... i mean the picture that is...

[waylon]

Greetings! Yeah, I've gone into more depth on this before, in Wiked Ewok's color theroy thread (and elsewhere). But I think I can do a better job, and hey, this sure beats working.
http://www.sijun.com/cgi-bin/ultimatebb.cgi?ubb=get_topic&f=3&t=000583

Basically what it is, is that for any given "brightness" of red, green, and blue, our eyes will percieve the green as brightest, then red, then blue. That's been pretty much determined already in this thread, so I guess I didn't really need to repeat it. Oh well.

Anyway, the point I wanted to add is the part about the brightness of the colors. If you go into photoshop and pick a red [255,0,0] a green [0,255,0] and a blue [0,0,255], then paint splotches of color on a black background, the green will look brightest, etc. But if you make a crosshatch pattern of the three, it looks like pure gray. So when the colors are alone, their values seem to vary, but our eyes still combine them equally.
http://fear.incarnate.net/~vlad/sijun/color.html

One final word... I'm not really sure what this exercize is supposed to accomplish. Except for on the computer, there aren't many places where you can get such pure colors. In a given color set, a red might be just as bright, or brighter, than its associated green. And the color of the lighting plays a big part, too (if you have a blue light shining on red, green, and blue paint, the blue will obviously show up brightest.)

So it's a good thing to be able to judge values independantly of color, but I don't think it's horribly necessary to be able to guess what a color is based on its value. Though it is kind of fun to play around with.


[edit] ieatbabies: Woops, I was writing this when you posted your little pic. And I kind of forgot to mention - the reason yellow, for instance, is the brightest of all of those colors is because it's a combination of both full red and full green. Which is another reason why the aforementioned exercize might not work right - if your red is a little greenish, and your green is a little blueish, the red will be brighter than normal, and the green will be darker. So yeah. Don't make me get into the CYMK color model. [/edit]

[ June 21, 2001: Message edited by: waylon ]

[Frost]

Oh, by the way, re-reading the first post, I just wanted to say that it's impossible to determine the color of a surface from its value alone. How would could you know what's dark green or blue or medium red? Therefore I agree with Steven's first reply. =) (sorry, I guess I initially read wrong)

[Awetopsy]

Wow.. Waylon. thats a really interesting read. thanx.

Frost: I understand that you cant really tell what the color is but at least if you know your tones i Imagine you could make an educated guess...

Ieatlitlebabies: Its definately an interesting topic..