[waylon]

That's another great point you bring up, color in composition - the way a more saturated, or a repeated color can focus the viewer's attention on a specific part of the picture.
http://www.artrenewal.org/images/artists/b/Bouguereau_William/large/Petites_maraudeuses.jpg (all hail artrenewal.org. )

Now, when you first look at this picture ("little thieves"), your eyes are probably drawn to the child's face. That doesn't have much to do with color, so we'll skip to the next focal point. After the child, your eyes are first drawn to the area around the lady's head. Granted, there are many compositional aspects that contribute to this (not the least of which is the child staring right at her), but I'd like to point out that her bright red kerchief, right next to the bright green patch of leaves, creates a very strong contrast, which immediatly catches the viewer's attention. Additionally, both the red and the green appear more vibrant due to their proximity to each other. (If you use your color picker, the leaves vary between 50-75% saturation).

The next thing that should catch your eye (assuming the entire image fits on your screen) is the basket in the bottom-left corner. Why is this? Repeated use of color. Since one of the main focal points is a bright red splotch, and there's not much red elsewhere, your eye separates out these two items and gives them more weight than much of the rest of the image. Granted, repetition is a well-founded compositional rule, but it works especially well with color.

[Flinthawk]

quote:

---

Originally posted by waylon:
(all hail artrenewal.org. )

---

no kidding, waylon. Exactly what I thought when I went through Jason's last post...*Add to Favorites*

[u2art]

red is the most attention getting color....and it is also the first color able to be seen by human children...interesting huh?

Art

[Jason Manley]

hahha...ALL HAIL ARTRENEWAL.ORG ....I completely agree....just found that site last week sometime and spent an entire day on there looking at master paintings that Ive never seen in my whole life...and they are good repros too.

man o man...good stuff. I must have added at least a hundred paintings to my favorite painting folder. yum


jason

excellent points all...keep em coming...Im learning too!

What else can you think of?

[waylon]

This has been touched on briefly, but here goes.

To create a rich, vibrant color, the most important thing you can do is place it in the context of colors that will amplify it's strength. One way is to place it adjacent to it's compliment (blue next to orange, for instance, as in the Monet House of Parliament picture Jason posted above.) You can also desaturate the colors near it, or even the entire rest of the scene. (Schindler's List, anyone?)

In any case, my point is that if you want a vibrant color in your scene, the best way to achieve the effect is not to turn up the saturation of that color, but to carefully plan the colors around it.

Does anyone with more experience want to extrapolate on this a bit?

[Jason Manley]

that was very well said.

take a look at how gerome pops his colors in the focal areas. notice that the same colors are mixed and mingled around the piece and that they are also pushed down in chroma in comparison to the crescendo of color brightness in the areas of importance.

link...check the bottom pic of the standard bearer for the mentioned theory
http://www.artrenewal.org/museum/g/Gerome_Jean-Leon/page4.html

another good example is the dual after the ball...
http://www.artrenewal.org/museum/g/Gerome_Jean-Leon/page5.html

and sentinel at the sultans tomb...also see the japanese imploring the idol(I think thats the title) and notice how the compliments are placed strategically in the focal area.
http://www.artrenewal.org/museum/g/Gerome_Jean-Leon/page7.html

jason

[waylon]

Well, this is more on the technical side of things. But while red is the most attention getting color, our eyes are most receptive to green, then red, then blue. What's interesting (try this at home, kids!) is that if you go into photoshop, make a green, red, and blue splotch on a black background, there'll be a noticable difference in brightness (as I mentioned above.) But if you make a crosshatch pattern of the three colors and fill the screen, then zoom way out, it looks gray.
http://fear.incarnate.net/~vlad/sijun/color.html

This makes sense when you think about it, but it's just really interesting that when your brain combines three colors that are apparently very different brightnesses, you get white. (we had an argument about this at work, once. One of my co-workers thought that the monitor had to bump up the brightness of the blue phosphors when it was displaying white.)

And as an aside... It's always fun when you first come across another artist's source of inspiration. Like when I first got to look at John Singer Sargent's work in detail, I saw where Spooge learned a lot of his technique. The Veb/Frazetta connection is a bit more obvious. And looking through the William Bouguereau gallery at artrenewal.org... well, has anyone played Icewind Dale?

[This message has been edited by waylon (edited March 16, 2001).]

[Jason Manley]

Great post Waylon.

Skip Liepke told me once when I sat in a lecture of his that the best way to create your own look, style, technique and color palette is to take your four favorite artists and mimic them in the ways that you most admire them.

an example...if you like the color palette of william merrit chase and you like hidden the brushwork of bouguereau and you like the subject matter of Frazetta you can create something entirely new. Most masters did this...look at the painters on artrenewal.org and then go look at raphael. you will see rapael all over the place in their work.

the trick is to not learn from just one artist so you look like you are knocking them off...the trick is to combine as many things as you can from the art that you like into something that is your own. Influences are only that...influences. I learn a lot from doing things like that sometimes.


jason

[TekK]

i was asking about color theory resources on the web, and someone recommended this thread, so i thought i'd revive it, since it's such an awesome thread...

anyways, i had a question or two about shadows and highlights... up til now i've been working mainly in greyscales, but i want to get into colors, so i've been reading up what i van about color theory... anyways, the question i have concerns shadows and highlights... from what i gather, shadows should be done by using complementary colors from the base color... however, i'm not too sure about highlights... since i normally use greyscale and then overlay a color onto the "shadowmap," i've been using black and white to add shadows and highlights... in color would i use adjacent colors? or just a lighter shade of the complementary colors?

hope to keep this thread going!

cheers,
teck

[edit]oh yeah, and this is with regards to digital (RGB) art, as opposed to traditional media, just so there's no confusion... :P[/edit]

[ June 25, 2001: Message edited by: TekK ]

[Chapel]

Highlights will go towards your lighting color + the base color. Extreme white being the brightest highlight.

[Plouffe]

following Painted Melodie's thread about movies and color schemes.. Well has everyone notice the color scheme for Battlefield earth?? using 2 diferent schemes on earth tells to viewer in what environment they are in or what air their are breathing...
If its the air that the human breathe well the color looks more EARTHY and if its the aliens air its like green or blueish grey?? i cant really remember...

Anyways if anyone know what im talking about thats GREAT hehe... sorry if their is no relevance just thought i would bring it up after seeing Painted Melody talking about movies ...

[Steven Stahlberg]

I've learned in 3d cg that highlights tend towards different colors depending on the material. For instance, on metal the highlight goes toward a more saturated version of the base color, only reaching pure white as a last resort, in total over-exposure.

On plastics etc this effect is much less noticeable, and goes 'sooner' towards pure white (assuming the light is white).

Skin and other organic materials have the opposite effect to metals - the highlight is a LESS saturated version of the base color. In practice, in 3d, this means mixing some of the complimentary color into such a highlight.

And the color of skin towards the 'terminator' (shadow border) should, at least in strong light, usually be warmer than the rest (a hint of bloodred).

The scintillating 'wet' effect mentioned earlier, when small dots of saturated complimentary colors are placed next to each other, is probably due to the fact that the lens of the eye needs to flex different amounts to place different wavelength light in focus on the retina - the effect being most noticeable between the two extremes blue and red.
In PS, try putting fully saturated blue and red right next to each other, then try to look at both at the same time - you should feel an unpleasant sensation.

Steven

[unpleonastic]

just some thoughts:

I always try to remember that the eye works by edge-detection; it notices mainly the edges of objects (and expecially when the edge is moving!) There have been experiments on the nerves in the eye and to support this, I think.. anyway your brain takes this information, which has been spatially high-passed by the eye and reconstructs the scene for optimal perception. Yer eye has incredible dynamic range (as does yer ear.. at least 6 orders of magnitude, from 1 microwatt/m to 1 watt/m, but i digress)... you can detect as few as 500 photons/second of blue/green light (rods are most sensitive at this wavelength, they respond to about 1/5 of the photons that hit them[you know on a really dark light how the world looks grainy like a crappy photograph? those 'grains' are the actual photons that the rods detect(!) things look greenish/blue because the red cones are the least efficient at registering photons in their active range (%5), and so you stop seeing red first as the light dims]).. and yet you can look at the sun, which blasts something like 10^13 (or more, I don't remember exactly) and still have peripheral vision. As an example: you take a picture in bright sunlight with nice, deep shadows and either you see the detail in the brightly lit areas or the shadows...hardly ever both (or at least whenever I take a picture). ie. I used to take pictures of sunlit woods as I though the patterns of shadow and bright light were intriguing... but i never got it to look right. now i wait for a cloud to reduce the contrast of the scene and increase the fidelity of the picture. The typical contrast of a monitor is (at best) 1000:1, black text on white paper maybe 500:1 (don't know about the other media)- a very tiny fraction of the real world contrast(10^13:500) - making it very hard to portray something realistic on these media (hard for me, that is...). I sometimes find myself decreasing contrast of scenes in photoshop to get areas that are blacker than others, which sometimes works, and often screwa stuff up; I might argue that you eye has such incredible reange, it never sees complete shadow, it always detects some edges in there. It might be counter-intuitive, but try adding tiny details to shadows (?)

Ahh that was not related to color theory, but I will try to tie it in. ok: say you have a bright red object next to the deepest black. right at the edge on the (imaged) red side, the red cones and rods will be going crazy active; their neighbors on the black side will be completely repressed. howevever, the green/blue rods will not have this sort of edge detection across the imaged red/black gap - they will merely report "no green, no blue". So: right at the edge in the black region the red and luminance (from the rods) will be strongly repressed, while the green/blue is slightly repressed: thus yer eye reports a slight blue-green at the black side of the image boundary - a cool color. Or, you could think of it as your brain turning up the gain of any of the color channels when it is not getting much signal from them... a bright red object will saturate the red cones, but the blue & green cones will go on looking for data, and they will find it in the shadows, where it is not drowned ot by the stoplight or whatever. Of course, this is all so subtle (your brain does an excellent job of reconstrucing the real world)...


perhaps this was usefull? perhaps it is wrong?

[Collosimo]

I made some interesting observations in relation to Waylon's previous comment.

When I went to the link, and viewed the images, I understood what it was about. I also made some other observations.... When viewed up close I found that the green dot grew outwards and looked bigger and brighter. This will be because of its value and its temperature. The interesting thing I observed next, happened when I was experimenting with the idea of the scale.. the cross hatched optical mixing, as opposed to the big dots... Well basically I decided to walk as far back from the monitor as I could go. I wanted to see if the larger dots then neutralised to grey like the hatching... because of my change in position and therefore scale. Of course as I already geussed it didnt really have that affect, probably because it is not a pattern, and I probably wasn't far enough away. What I did notice was that the red dot was now growing outwards and was cooming forward off the 'canvas'. Red was now the more prominent colour. Somebody mentioned this about red earlier... This effect is apparent in the real world when you look over the other side of a stadium, and you mainly see people wearing red clothes!

Something else I noticed... apart from the theory that when the hatching goes grey because of each colour canceling each other out... I concluded that scale was another attribute (due to my previous experiment).. I think that this strongly relates to our understanding of aerial perspective...

So what is aerial perspective... well, It is commonly stated as the effect of air volume, and little bits of dust amd smog that after a large distance, tend to make distant things look hazy... desaturated, and closer to the midtones.

I now think that apart from the density of air, and the little particles creating the aerial perspective effect, I think that scale (perspective).. is now obviously to me a very important factor. Waylons cross hatched green, red, blue... highlights the theroy here, becuase what happens to the colours when they are organised as cross hatching, and at a very small scale.. well the lose saturation, and the value comes closer to the midtones aswell! sound familiar well its the same idea as air density in aerial perspective.. except now I realise scale has just a much to do with it.


(Wow that was long.. I hope you guys can understand that.. my grammar and descriptions were terrible. I couldn;t help it though, this stupid thing is doing OVR (Overtype) like in MS Word.. So I couldn;t go back and edit.)

[Ben Barker]

quote:

---

Originally posted by Jason Manley:
always interesting color posts

---

You should write a book man. Your many posts on color theory contain more info than all of the "paint by numbers" stuff that clogs most of the art shelf now.

[Zaphod]

unpleonastic: I read your post and instantly rememberd some of my black and white photo lessons. We learned that you should measure the amount of light in your palm when it's in the shadow of the object scene you're photographing. After that you measured the amount of light in you palm in the sun. When you take the photo you use the exposuretime based on the light in the shadow. Then it involved a lot of tweaking when you developed the film. You use it to get a good range of values. So that you don't loose the details in the shadows or in the light. Needles to say you have to use up the whole film in the same light (because of the developing of the film later on).

But I gather that you're not developing your film and copying you photos yourself. So you could try to measure the light in the shadow of your cupped palm atleast. The camera assumes that the overall value in your scene is a neutral gray, so and when you have a strong lightsource everything gets underexposed. To be able to get around this your camera requiers you to be able to set the exposuretime (compact cameras often don't have these settings. Systemcameras do). So you can't use it on any camera. Also, when working in colour you should always overexpose to get ritcher colours. Or you could try to measure the light in your palm both in the sun and in the shadow. Then use a mid value when you take your photos. This *should* enable you can take pictures in different light on the same film and still get decent pictures.

Hope it might be of some help. I'm not a good photographer by any means (personal opinion atleast). So it might not be the most accurate advice around. It might not translate too well if you're taking photos i colour. But it could be worth a try.

<EDIT>Why you use the palm your hand is because it's supposed to have a neutral gray value. You can also buy a (expensive) sheet of paper, from kodak or whatever, which is guaranteed to be neutral gray.</EDIT>

/Zaphod

[ July 15, 2001: Message edited by: Zaphod ]

[Jason Manley]

I can agree a bit with the last statement though I dont always see that as the case...there is always color variation in any one value of light...look for it..you will see it. highlights are almost never pure white...unless you are making a black and white drawing...just as shadows are never pure black...unless you are in a vacuum or a black hole.

highlights contain bits of color from the main lightsource in a subtle degree. it is important to put bits of color in your highlight as that shows the viewer (whether they know it or not) what color the light is.

jason manley

[Jaymo]

Jason: Rereading the whole thread has left me with one question unanswered:
What is the "right" set of complementary colors or are there two sets?

Judging from this article there are complementaries for mixing "real" colors and complementaries concerning perception. Is that true? What is your knowledge on this subject?

These are the two sets I am referring to:

• colorwheel 1
  
  
• colorwheel 2

[ October 17, 2001: Message edited by: Jaymo ]

[Steven Stahlberg]

I'm sorry, as far as which color is opposite to which, I don't see much difference between these two wheels?

I know there's one theory that has blue, yellow, red as primaries, and another that has RGB - red green blue. For RGB the opposites would be Cyan, Magenta, Yellow. But they both seem to have this... at least roughly?

[Jaymo]

Well, the main difference to me is yellow being the complement to either purple or blue, which makes a difference i guess. The other combinations are shifted, too.

[edit]
Argh! But you are right, I chose the wrong examples for the sake of having found such nice big graphics. Damn. :�

Here is a better example to compare with, which illustrates my point:

colorwheel 3

[ October 18, 2001: Message edited by: Jaymo ]

[Kaiju]

Well, that is basically one person's theory and it is a bad one in my opinion. I think while she is correct that the color compliments she chooses do make the colors more vibrant.. neither color pulls one forward or back. The reason you use complementary colors is to compliment. When you do choose to use complementary colors keep in mind that it does not necessarily mean you have to use the exact opposite tone and at the same value as the foreground color. In essence she is using different tones of the complementary colors at the same value. If you ask me that lemon picture is hideous. It hurts the eyes because everything is at the same value. I don't think she really understands color theory herself.

*Edit: Jagermeister hangovers are bad.

[ October 18, 2001: Message edited by: Kaiju ]

[Jaymo]

Kaiju: You may be right with most of your comment, but i think you are mistaken concerning the meaning of the term. It's called complementary colors, AFAIK, therefore has little to do with complimenting. It's the complement of the other color.

[Steven Stahlberg]

Yes, I see now what you mean. Yes that third color wheel is what my art teacher taught me as a kid...

I'm no expert at color theory, just a layman. But I'm under the impression that mixing any 2 'opposites' in the Red-Yellow-Blue wheel always produces brown, not gray, right? And mixing the opposites in the RGB wheel is supposed to produce gray - if the colors are balanced right? So that would mean that the older Red-Yellow-Blue wheel isn't really putting the true 'opposites' opposite each other?

Confusing.

[Kaiju]

quote:

---

Originally posted by Jaymo:
Kaiju: You may be right with most of your comment, but i think you are mistaken concerning the meaning of the term. It's called complementary colors, AFAIK, therefore has little to do with complimenting. It's the complement of the other color.

---

Sorry I fixed it. I wasn't paying attention or something. I was still trying to say the same thing.

[Jason Manley]

wow..some confusion here...

The definition of complimentary colors is any two colors that are maximum distance away from each other in the color spectrum...and that means that any two colors which have the absolute least in common would be compliments.

here is something to think about....

if I have a cool blue and a warm blue and want to find the compliments to both..which simple colors would I look for?


basic compliments are below

red/green

blue/orange

yellow/violet

you can break it down from there.


jason

[Jason Manley]

There are many things I do not agree with in the article above. for example take a look at her little swatches of color in the squares...they appear different in her example because they are different...if you dont believe me do a print screen and test em im photoshop...hahaha..

the problem seems to be that she reads too many books and does not do enough practicing herself (her work tells me plenty of that)

mixing compliments does not make colors darker in my experience...there may be a mixture that seems darker..but it is not..it is simply less intense..less saturated. The value does not change (unless the compliments mixed are of different value in the first place)...just the intensity changes.

mixing compliments makes an entirely good variety of colors. it is these colors that hold your images together if you use them right...the impressionists used them A LOT..they are called transition grays. They can be orangish..bluish..greenish..reddish..any color you choose. all browns..and grays will seem like a color when it is used in conjunction with other colors.

all colors are RELATIVE to those that surround it.

IF YOU DO NOT KNOW WHAT COLORS ARE CREATED BY MIXING COLORS THAN YOU NEED TO DO THE COLOR MIXING CHARTS PASTED INTO THE NEXT POST....that is if you want to be a good painter..or good texture mapper or good lighter...or..you name it.

if you want to be an artist then you will need good color knowledge.

do the excersize.


jason

[ October 18, 2001: Message edited by: Jason Manley ]

[Jason Manley]

Here is the excersize I mentioned. It can be used for any palette you chose and not just the limited palette below. It should be done with all colors you use in any given palette in order to increase your range of knowledge about what colors are at your disposal.

doing this is sort of like doing scales if you are a piano player..it lets you know your full range of notes that you have and are capable of using.

------------------------------------------

IF YOU KNOW WHAT THE POSSIBLE COLOR MIXTURES ARE FROM A LIMITED PALETTE THEN YOU CAN USE THEM DIGITALLY.

if you do need to memorize what colors to use for a limited palette(or any palette) follow this traditional excersize.

take these four colors (indian red, burnt sienna, ivory black, and yellow ochre) and do mixture charts. You will also need titanium white or flake white.

1 mix each color with every other color to a visual 50/50 mixture. Make a three or four step value scale down to white with each mixture. make a note of what colors are used for the mixtures. (yes I know this is a lot of work..but it will pay off..trust me)

2 mix each mixture to a 75/25 percent split and a 25/75 percent split and do the value scale to white for each.

DO THIS FOR EVERY COLOR ON THAT FIVE COLOR LIST.

what mixtures make nice greens (relatively)

what mixtures make nice warms?

what mixtures make nice cools?

What is your coolest mixture?

what is your darkest mixture?

what is your most intense mixture?

what is your most....

best of luck

jason

ps..yes I did this for many hours in the beginning. I had two different mentors make me do it...yikes! boring...but relaxing..and its good for your color mixing skills too.

Remember...rembrandt, velasquez and many other artists only used the above colors and their stuff looks like it is full color in ways....it is all about where you place your colors and knowing how to mix your colors.

jason

[Steven Stahlberg]

quote

Quote:

for example take a look at her little swatches of color in the squares...they appear different in her example because they are different...if you dont believe me do a print screen and test em im photoshop...

Yes, and the purple is obviously not as saturated as the blue.

quote

Quote:

The definition of complimentary colors is any two colors that are maximum distance away from each other in the color spectrum...

Which spectrum? On the RGB color wheel that means yellow is opposite blue. I know, blue and yellow paint mixed produces green, I learned that almost 40 years ago, the confusing thing is blue and yellow light mixed produces... white. Or grey. Can somebody explain?

Would it be a good idea for paint programs to actually change their color wheel into the other one, the Red-Yellow-Blue one? (Whatever it's proper name)

[Jaymo]

Jason: Thanks a million for the elaborate answer. I'll have to get some colors and try out the exercise.
Yet I'm still confused in the colorwheel matter. You seem to agree only with the complements in the RYB-Wheel. Could you explain why the opposite colors in the RGB-colorwheel are not the complements of the primaries? Or are the complementary colors not defined that precise and it's more a matter of color temperature? Since you disagreed with that paintress I don't think you'd agree to that.
Ah, BTW, don't you think it's called complementary colors (since complement means opposition, polarity, antipode...)?

Steven: Hehe, looks like i've sparked your interest for the problem, too.

[Jason Manley]

the list of colors I mentioned above is the compliment base that I speak of.

red green

blue orange

yellow violet

red orange/ green blue

warm orange/cool blue

cool yellow/warm violet

etc...

jason