[Jaymo]

Well...yes. I do understand that (I think), but that does not seem to be the answer to my question...does it?

*sigh*

[aquamire]

If we had pure primary colors, IE: Yellow, Blue, and Red, Blue and Yellow would make a Black color, as well as Red and Blue. Thats because there's no such thing as a pure primary. Its now thought that Yellow, Blue, Red, Violet and Green are primaries, but even they arent pure. When you mix Blue and Yellow you get green, right? Well, actually Yellow and Blue cancel each other out as well as every other color (theres a smidgen of every color in every 'primary') except for GREEN, thusly Green is the only color left to bounce back and hit our eyes.

I may be wrong on this, but I think on computers since its additive mixing (were talking light) we are truly dealing with pure primaries, and thats why its so difficult to mix colors like you would on canvas, IE why when you mix blue and yellow you certainly dont get green. Im SURE theres a way to simulate how colors are mixed irl, but no ones done it yet it seems. I think it'd be a major relief for artists.

Ugh, I'm still confused about all this myself, I hope someone knows more than me.

Get the book Blue and Yellow Dont Make Green, by Michael Wilcox. Its a revelation in color mixing theory, however I dearly wish it translated to computer graphics, which it doesnt, eheh.

[Steven Stahlberg]

Yes, that explains it! There is a symmetry and logic there after all, additive mixing makes white, subtractive mixing makes black - except in the imperfect real world the colors aren't pure!

Edit: what I mean is, the RGB color wheel is additive, while the RYB wheel is subtractive. Light is additive, while paint is subtractive.

So again, I ask myself, wouldn't it be good to have access to a RYB wheel when mimicking real-world paint-mixing in a software?
Hm, I guess you could make your own such palette in Photoshop.

[ October 18, 2001: Message edited by: Steven Stahlberg ]

[Jaymo]

Steven: Well, that's a point, still the CMY(K)-wheel is the same as the RGB-wheel, at least distribution-wise, yet it's subtractive.
So what now? There still is no answer to the question. If the RYB-complements are the right set considering visual perception, then why are the opposing colors in the more "modern" color-systems RGB and CMYK not?
I'd really like to hear an explanation why yellow is not the complement of blue, but violet/purple. Anyone?

[Steven Stahlberg]

ARGH! You're right! The CMYK wheel..?!
I'm screwed, no sleep tonight!!

[pierre]

What is the confusion?

I think this is a very interesting discussion, color theories are many, for many reasons.

However sometimes I believe one has to think like a scientist as opposed to an artist when it comes to analasys such as these. That does not mean a rival towards the artist in you, but a temporarily scientific objective that you can later fuse into your artistic needs.

The artist in you may be reluctant to view it from a scientific view purely, but one can't escape the fact the cones and rods of your retina (unless you have a color anomalia - sometimes called "color blindness") is practically biologically the same as with the scientist.

I know that "seeing" and "seeing" can mean two different things, what I am referring to is the physiological way of seeing, not seeing with all your senses, emotinos, experiences etc. To be more extreme, adding the way a human perceive colors, not like a wasp or a goldfish for that matter. After all, our color theory would not work for the goldfish.


I don't think there should be too much of a confusion after reading all the articles in this:

understanding color vision and color theories

And when putting the knowledge into practice (especially with subtractive practice, i.e physicall blending paints), know that we do not have ideal/pure colors and will thus not produce the theoretic color, just the approximate such.

Pierre

[Steven Stahlberg]

Ok, I'm tripping, too much coffee I guess... so the RGB and CMYK wheels are called the Newton color circle, and they're all additive. That seems to make everything fit again for me. Thanks.

[Anthony]

It's been too long since we had these vastly educational discussions. Lots of people, lots of ideas. Thanks Jason for starting these threads!

[Jaymo]

Steven: No, the CMYK-colortheory is a subtractive one. Otherwise the mixing of colors would make them brighter. But in 4C-prints both subtractive and additive principles occur, because the remitted light is mixing, too. This is called autotypical color synthesis. (something to back this up)

Pierre: Thanks for the post, that's a very interesting link. Yet it does not seem to deal with the RYB-model and complementary colors. So my question still is:
Why is yellow not the complement of blue (like for example the Munsell system would suggest), but violet/purple?

Anthony: Didn't Wiked Ewok...?

[ October 20, 2001: Message edited by: Jaymo ]

[Steven Stahlberg]

Jaymo, yes you're right. Sorry, I'm just easily confused these days, busy with lots of other stuff... I used to know this. See this image from the link you provided:



It seems to show the 2 different color wheels we've been talking about right?

Except for one little 'shifting' of colors - what we normally call 'blue' is actaully cyan in the subtractive one on the right. So what we normally call 'purple' becomes 'blue' - problem solved. Just a matter of confusing labelling. The technical name for Blue in real life looks kind of 'purpley' to most people.

[Jaymo]

Okay, i found some words about the complementaries on the link given by pierre:

• Newton Color Circle: "R,G,B are thought of as the additive primary colors, and their complementary colors are placed across from them on the circle. The colors then fall on the circle in the order of the wavelengths of the corresponding spectral colors."
  
  
• Munsell System: "The simplest and most widely used subjective color system. Complementary colors are not on opposite side of value axis."
  

Since the opposing colors in Newton Circle and Munsell System are the same this means there are two sets of complementary colors or it is just wrong.

I think there are two sets, one in the mixing sense of printcolors (Newton Circle), one addressing the perceptual complementary colors. My reason to believe this is that there are colorsystems in existence that distribute colors by even amounts of wavelength (i.e. Munsell) and some that distribute by the subjectively even distribution of colors (i.e. CIELab).

So if the perceived distance between colors is different from the distance in wavelength, this could be a reason why the complementary colors related to perception are not placed opposite their respective primaries in the wavelength oriented models.

Does this make sense?


Steven: Yeah, that's an interesting point, too.

[ October 20, 2001: Message edited by: Jaymo ]

[Jaymo]

silence=approval?

[pierre]

First of all, you don't get magenta by even distribution deriving from the RGB wheel, similarly, you don't have magenta in one slice of the Newton Spectrum. You get magenta by connecting the two ends of the spectrum.

We have mentioned this before and it could be repeated that we do not have ideal colors and thus, for e.g, in the case of magenta, there is a lot of yellow which must be rivaled by adding blue to the pigments (especially if the purpose is for CMYK printing). The absence of purity in the pigments is also the cause to shifting of saturation when blended, optically or physically. Either to a higher percentage of black or the opposite, in both cases, saturation decreases.

yellow + green = more blackness
red + blue = more blackness
red + yellow = more whiteness
blue + green = more whiteness

This shifting is part of the reason why it is good to have at least e.g.:

two yellows (warm(reddish)cold(greenish/lemon)) on your palette
two blues (for e.g utramarin(reddish/warm) cobolt(greenish/cold))
two reds (for e.g Crimson/Alizarin(bluish/cold) Cadmium(yellowish/warm))

+ maybe other colors

This makes for a lesser shifting in the blending.

Better color prints can for instance be achieved by using more "primary" colors to beging with, e.g. addition of green along CMYK.

The cheaper and better printing process today uses a process similar to the Munsell and NCS color system (derived from Munsell): Achievment of any color by one slice of a primary or secondary color + addtion of white or black (i.e the amount of black on the paper). Therefor you get brown by yellow + magenta + black, not yellow + magenta + cyan.

The reason for this mumbo jumbo is to execute the thought of one color system as the only right one. We can't just pick colors from a linear axis system in a color wheel from a certain color system and expect it to work in the same way every time, because we simply don't have ideal colors at our hand. What about in theory only then? Alright...

I think we all agree with the additive and subtractive factor. In the former you add light, while in the latter you subtract it. The monitor for instance uses the additive method, RGB, right? R=yellowish red, G=yellowish green, B=reddish blue. So the answer is, why exactly those colors, isn't it possible that we could have achieved the same effect by other primaries, the answer is yes, we could. Any combination of extracts from the spectrum that in a blending system produces the same colors could have been used, but the RGB seems to be the most efficient, because they very closely match the cones in our retina, but is nonetheless not the only possible combination.

That must lead us to a clue on the definition of Complementary colors. Additive blending of the TV R+G produces Yellow, why, because in combination, they stimulate the two rods in our retina that in turn gives the information to our brain that this is yellow. NOTE, that they do stimulate two parts/lines of our the spectrum, that is both cones separately (in practice not totally separately since especially the R and G cone overlap quite a bit, but just for the sake of simplicity we assume they get independent stimulation) . We could have achieved yellow by stimulus of ONLY ONE part of the visible spectrum too and produce the stimulus of yellow. That one part would have stimulated both the R and G cone and produced the same yellow. Some colors, you can't get by one wavelength of the spectrum (only one line of the spectrum) you must have two, for e.g some turqoise colors must be combination of wavelengths.

Any wavelength or combination of wavelength that produce white in an additive blending complements the other.

Any wavelengths or combinations of wavelngths that produce black in a subtractive blending complements the other.

The trick is to find the colors/wavelengths (one will not succed fully in physical blending, and the TV is another matter because you can't really apply all color theories on it since the light energy must be measured by other methods. Black on a monitor is still many times brighter than black on an oil from Bougerau (he didn't use purely black areas by the way).)
that will produce the comlementary result, and those cannot be found simply by a non logarithmic axis system so widely thought.

I know much more can be said about this, but I am getting sleepy.

Pierre

[Jaymo]

Pierre: Thanks a lot for that great answer, I'm getting a lot closer to the reasons of my confusion and that feeling we are discussing different matters. In theory I pretty much understand, but I feel I really should get some painting colors and do some mixing for full comprehension of the subject.

If you feel like going on with this after some sleep, be assured I'll be listening. This is very interesting.

Gotta take a nap, too.
Cheers...

[Jason Manley]

excellent conversation....the contrast from what we are seeing (how that is useful to us) along with what we are feeling (based on what we are seeing) and scientifically how we are seeing is very well rounded.

great work.

jason

[ October 22, 2001: Message edited by: Jason Manley ]

[lookitsfrank]

wonderful thread. I just wanted to bump this to the top again for a fresh read. I think we should have more conversations like this on thing such as composition and other topics. would make for a wonderful learning process just to read through them.

[Frost]

Imho, I think this whole RGB/CMYK stuff should be considered with a grain of salt. I'm sure it was convenient to produce these systems which allowed for a quantizeable representation of color as we percieve them. They are just rules based on simple mathematics to make our job easier. How long has CMYK/RBG systems been with us? Probably not very long... I'd say since the beginning of color print and CRTs (TVs, etc.). Both these systems are convenient, and are the good/evil twin brothers of themselves.

Color is caused by a stream of light frequencies measured in speed. There is absolutely no concept in nature of "wheels" and how color need to wrap around to form a circle... (perhaps some old master made this up on his palette so it was easier to generate all the colors he'd need?). Mixtures of frequencies... is all it is. No simple rule can be made about them. If you want to know your stuff, go read scientific manuals... artists are no good at this color stuff. =)

[BooMSticK]

yeah Frost, but its a damn interesting subject.
,Boom

[Idlewild]

Got this link on a different forum:
http://www.webwhirlers.com/colors

This guy seems to deal mostly with colors
on the web, but much of it probably applies to all forms of color theory.

[Tarandon]

Ok, I didn't read the whole thread, but I just wanted to explain why shadows appear cool under warm light and visa versa.

First off, we have 3 types of cones in our eyes, red gren and blue. These cones are receptive to only certain wavelengths of light and their combinations create our great color spectrum. However, just like a muscle gets tired so can these cones. If the only light sources are a red and yellow light bulb, the red cones in the eye start getting tired and begin to lose their ability to transmit. Since your eye is now less sensitive to red light, your partial shadow will be percieved to have less red light. Or more blue and green. Blue and green are much cooler and the shadows therefore appear cooler. The shadow itself is not cooler, we just percieve it as such. Of course if only one light source is present no matter what quality of light, the object will block out all light from the source. Blah blah blah.

I hope that made sense, If you finished reading it


Later!

[LoTekK]

crikey, tarandon!
now i finally understand the whole thing behind light/shadow -> cool/warm... thanks, man, now it makes sense (as opposed to being just sort of good info... )

[Basement bound]

This is an exallent thread. I am learning a bit slowly. I want some feed back about if I am catching on. Theroy is great, but some idea of application has always helped me better. Here is my thread, notice that I have two images of rocks at the end of it. The second rocks image was done after reading the thread. Do y'all think I am getting it, or perhaps have some more pointers.

JA
http://www.sijun.com/cgi-bin/ultimatebb.cgi?ubb=get_topic&f=4&t=002737

[ December 31, 2001: Message edited by: Basement bound ]

[jr]

jason, i think he wants to be called malcohm now...
this is a very interesting thread.
if you guys want to see great use of color, check out the work of sabastian kruger, not sure if i spelled it right, but his paintings are incredible.

[Akab]

Oh my gosh, this is an awesome thread! Simply incredible, like a revelation. (of course, it won't mean much unless I put it into practice...must...solve laziness)

A humble thanks to all who participated in this thread. m( )m

[don perkins]

wow, great thread. I am red-green color deficient. So I have to rely on theory and guesswork, and stealing color reference from photos I take using careful lighting.
This thread has been extremely helpful, and occasionally confusing. Thanks very much, I am learning a lot- I hope. Don

[Ben Barker]

Man, I work at this computer store, and one of the salesman there is named Don Perkins. Your posts always creep me out for a second.

[don perkins]

oh, hi ben...I've been meaning to say something in the store; I wasn't sure if that was you or not....Sorry, had ta do that.

[LoTekK]

quote

Quote:

oh, hi ben...I've been meaning to say something in the store; I wasn't sure if that was you or not....

lol, one can almost imagine an Awkward Moment(TM) coming up...

[Ben Barker]

Nah, I already asked him

Heh, sorry to polute your thread with the off topic stuff, Jason. Apparently I can't delete my posts on this forum. Perhaps Sumaleth will oblige me.

[ January 06, 2002: Message edited by: Ben Barker ]

[Jason Manley]

nonsense is a different thread altogether...please stick to the topic.

can any of you find some examples of master paintings with the below color palettes?

what is the definition of "tertiary" colors in regards to the color wheel?

what is the definition of the term "split compliments"?

what is the definition of the term "analogous colors"?

here are the color schemes for the first question...what other schemes can you find??

please provide links.

1 dominant earths (greens and siennas) with orange accents.

2 dominant cools (multiple grays) with warm accents.

3 dominant greens with red accents


thanks...lets see what you know.


jason