![]() Color Forums / Color Talk / Advice & Critiques
Not sure about the right colors to use in a project you're working on? Ask your fellow lovers for advice. No clean orange in RGB?
![]() ![]() maikev maikev wrote: 1 Month Ago
It sounds impossible, but I always had problems to mix a really good orange, and I missed the "buttons" for that!
Now I started to think and to read about colour models and was a bit surprised: In the book "Farbe im Design" (German) a prisma is printed. The text said, that the colours of the rainbow are: Red, orange, yellow, green, blue, indigo an violett. (Don't run away - I knew this before, too!) In the picture the rainbow of the prisma has: Red, yellow, green... There is no orange, only a smooth change from red to yellow. I wondered if I'm a bit colour-blind, but I normally *can* see orange! I looked around in the room and there is a lot of orange. I went outside and there was a lot of orange, too. In the RGB-colours yellow ist mixed from red and green. To get a saturated yellow we need 100% (#255) of red and 100% (#255) of green. Orange is mixed of yellow and red. How can I put red in the yellow? That was always my problem! And the "workarounds" always looked a bit dirty... In the book I found a few pages along the text that the reason for the usage of this "primary" colours is that we can mix all colours of the visible spectrum accurate from it. What's going on? Have they never tried it? Am I to stupid? I know that mixing orange from yellow and red was thought in the subtractive colour-model we learned at school and not in additive colour-model. But why are there no hints about how to mix red and yellow in additiv colour-modell (RGB)? ![]()
![]() ![]() ![]() rotten rotten wrote: 1 Month Ago
i suggest you to practice hexadecimals (base 16), where 0 to 9 for representing values 0 to 9, and A, B, C, D, E, F for representing values 10 to 15.
in rgb; where full yellow is FFFF00 and full red is FF0000, the middle-most colour is FF8000, a full orange. e.g. r100 g100 b100 is 6&4 6&4 6&4, #646464.
r50 g50 b50 is 3&2 3&2 3&2, #323232
r25 g25 b25 is #191919; to translate a rgb value into hex value, take the value of red (25 in this case) and divide it by 16. the result is 1.5625. the integer, 1, will be the first number in the hex formula. the remainder (0.5625) should be multiplied by 16, which results in the number 9. so, 25 translates into 19 in hex.
here's r16, g16, b16 which is #101010:
if you have integers or remainders that translate into from 10 or 15, then you should translate the 2-digit number into a single letter where 10 is A, 11 is B, 12 is C, 13 is D, 14 is E, and 15=F. r255 g255 b255 is #FFFFFF 255 / 16 = 15.9357 0.9375 × 16 = 15 15&15 = FF
![]() ![]() ![]() lesaint lesaint wrote: 1 Month Ago
Orange is my favorite colour, and I have spent time myself ruminating on how it seems to be the least assertive and most difficult to pin down of all RGB colours (pigment too, I think, at least when I'M painting).
I had once calculated FF80000 to be RGB orange (see my comment on Hue 30)... and yes, this is "pure" RGB orange if your definition of true orange is the midway point between red and yellow. Why does the little orange guy have to be so squirrely? I love you so! Come play with us! AND in answer to your question about "mixing" orange in RBG, since RGB yellow is born of full red and full green, the only way to "add" more red to the yellow is to increase the proportion of red by decreasing the green present. In other words, you can't add more red to RGB yellow, but you can take away some of the green from RGB yellow (taking it down from FF to a lower value, in rotten's example of pure orange here, to 80), thereby giving the red a louder voice. AND... for the record... just because this annoys the crap out of me and I too used to feel like I was stupid or going insane: in the subtractive (pigment) colour model, magenta (not red) is actually the primary color, because magenta + yellow = red. In my mind, this makes red a secondary color, and orange actually a tertiary color, but that's a whole other can of worms. Keep up the orange love :) ![]() ![]() maikev maikev wrote: 1 Month Ago
Thank you for your answers!
@ rotten: I also figured out that I get orange when subtracting green from the yellow. In the GIMP and other grafic programs there are controllers for rgb. If the controller for r is set to the highest value (255) and the controller for blue to 0, the controller for green shows a orange "spectrum" from red to yellow. But those orange tones look not as bright as the orange tones in nature are looking! And adding some "white light" to it (increasing brightness) makes it not better. No wonder... @ lesaint: Yes, you have fully understand the problem! Orange is my favorite colour, too (but I like a lot of colours). But not the orange tones I get when reducing the green in yellow in the rgb-modell. If a clear, bright orange is so difficult to make, how can all those marigold-flowers do it? And if it's only difficult in mixing light-colours, where all the colours together where white light, why do we have such a luminous orange in sunsets? Sorry, but I never will get those colours by decreasing green from yellow. They need a "full" yellow (and perhaps a bit more!! ;-) ) and then put *red* in it (not magenta! I tried the cmyk-modell and the results are the same). :-((( I have considered if my monitor must be adjusted, but the red looks okay and the yellow looks okay, so it can't be so bad. All the primary colours are looking okay. in the subtractive (pigment) colour model, magenta (not red) is actually the primary color, because magenta + yellow = red. Yes. Because of that you can's mix yellow and red in cmyk-modell, too! Because you need full yellow for the yellow and yellow+magenta for the red. I'm not shure what "analog" painting artists will say if somebody makes it impossible to mix primary yellow with primary red in their oil paintings! ![]() ![]() ![]() _stefan _stefan wrote: 1 Month Ago
Actually there is no system to reproduce all colors from the whole color space. The International Commission on Illumination (CIE) has made studies about all possible colors in 1931 and defined their space. On this image you can see, which system (RGB, CMYK, ...) supports which parts it.
The real big problem seem to be blueish green btw :) ![]() ![]() maikev maikev wrote: 1 Month Ago
Thank you, _stefan!
The real big problem seem to be the blueish greens btw :) I just tried it only a few seconds... There really seems to be a problem, although blue and green are booth primary colours in RGB. But when you mix them, you geht cyan. In the RGB-spectrum cyan is between green and blue. Cyan is a very bright colour, and if I want a saturated green with a primary blue in it I can't even see the point in the spectrum where this colour "should" be! I looked a bit in the Internet (thank you for the links!) and found something interesting: A virtual colour-museum! Here: http://www.colorsystem.com/ *Uff*, it's in English *and* German... Btw.: Sorry for my English. Using a dictionary while writing is no good for concentration on the sentences... ![]()
![]() ![]() maikev maikev wrote: 1 Month Ago
*rotlol*
Yes, in this color space you can reach aaaaall colours. You only have to add a bit more grey if it seems to be difficult. *gg* There are a lot photos of sunsets with very bright orange in it. I wonder how this is possible and take one of them and made a palette out of some colours I found in it: " />
The colours aren't as nice as in the photo... But in patterns they look better. Perhaps the combination of colours make them more luminous. Another consideration is that in pixel-images it might be possible to mix red and yellow in a raster (or even blue and green). I tried it today but instead of a rastered orange a small "reptile skin" pattern came into being. ![]() ![]() |
Advertise on COLOURlovers via The Deck
![]() ![]() ![]() |
![]() |












![922~[mat]_sable](http://www.colourlovers.com/images/badges/c/8/8573_922~[mat]_sable.png)





" />







