PDA

View Full Version : Help with color theory


meta87
08-24-2005, 06:02 AM
Well I'm just getting my feet wet in this fun stuff. I have started to learn how to paint using Painter 9 and also traditional oil painting. I have been reading some color theory articles on the web and am confused. I have read that a color's shadow has some of it's complimentary color in it. Well I have been painting a portrait and am confused as to whether or not I should be adding some green into the darker areas of the face since the skin is mostly pinkish?

Any tips on this kinda thing would be great, thanks! :)

Zoontjens
08-24-2005, 01:17 PM
I struggle with this too, a lot. i had no formal study but try it on my own (for fun). so i might not be totally right. it's just what i read/heard.

Before doing skintones do basic exercises! cubes in different environments etc. start with basics. Skintones is immensely complex, because more parameters are involved (translucency, skin thickness, skin quality, age, veins beneath the skin etc etc).

shadows are often less saturated (but with translucent materials like skin this is often the opposite! due to sub-surface scattering etc. but does not matter now...).

With additive color mixing (paint) you can make a less saturated red by adding its complementary, green. But be careful, and add the right type of green (hue, saturation, value) then! Anyway, it's a lifetime study i guess. some painters calculate color exactly etc. like a 3d renderer.

in general, when something is lighted from one side, the other side receives light from the environment (pretty obvious). This is why shadows outside during daytime are blueish (sky).

for skintones, check (of course) Linda Bergkvist's work and tutorials. furiae.com...

most important, just try things out and see what works and what doesn't, define your style!
cheers!

Rebeccak
08-24-2005, 01:46 PM
meta87,

Take a look at the photos posted for the Now OPEN!!! Open Figure Drawing Workshop with Hong Ly and Rebecca Kimmel 002 (http://forums.cgsociety.org/showthread.php?t=269454&page=1)~ one thing I was noticing was that the color scheme for these photos is really red and green ~ well, the figure is a very light tint of red (pink) and the bkgrd is a very dark, desaturated green (not black as you might initially think). Behold, complementary colors! The main mistake which artists make when they are first starting out with painting is to think that shadows are composed of grays ~ when in fact, they are composed of color. If you look at a portrait by any good artist, you will note that shadows do indeed contain the complementary color of the light source. I'll try to find some sort of example...but in the meantime check out the photos and I hope you see what I mean by their being color in supposedly 'black / gray' areas. :)

A good example is a portrait of Isabella Brandt (sp?) by Rubens. Note the green in the shadows.

http://www.artchive.com/artchive/r/rubens/rubens_isabella_brant.jpg

http://www.artchive.com/artchive/r/rubens/rubens_isabella_brant.jpg


~Rebeccak

dbclemons
08-24-2005, 02:03 PM
It's a general rule, but you don't need to follow it precisely. Using the complement in the shading helps to neutralize the shadow on the chroma base color, but there can often be other hues at play as well, depending on the lighting situation, and the look you're after.

Check out this reproduction of an Eakins portrait
http://www.artchive.com/artchive/e/eakins/eakins_cook.jpg

Most likely the subtle cooler mid-tones are created by the underpainting than an actual application of green. It's more an issue of how the paints themselves mix.

Whereas a colorist like Gauguin gets all crazy with color
http://www.artchive.com/artchive/g/gauguin/gauguin_cezanne.jpg

-DBC
or what Zoontjens said :}

ekah
08-24-2005, 04:17 PM
I don't have much of an attention span to get deep into the scientific studies of behaviors of light, but I believe it helps to know some basics of how light behaves in order to understand colors of light and shadow, and how they appear on surfaces when painting. Other things like bounced lights (radiosity/global illumination), reflectivity, transluscency are also something to consider when painting, especially for skin. Although my background is in traditional arts, having worked in 3D programs with Materials/Shaders have helped me to understand behaviors and appearance of light and surfaces better. Although some of the impressionist painters seem as though their colors are unrealistic, but to me, many of them actually painted colors they saw in a more obvious or exaggerated manner.

Check out some of the links below. You can skip the technical stuff, but it should give you some general ideas.

color in the world (http://www.handprint.com/HP/WCL/color10.html) - Go to "surface & shadow color (http://www.handprint.com/HP/WCL/color10.html#surfshadow)" section of the page
color in context (http://www.handprint.com/HP/WCL/color4.html)

meta87
08-24-2005, 07:36 PM
Thanks alot everyone! I'm overwhelmed with info now. Checking out everything everyone posted! :)

Jack Youngblood
08-24-2005, 07:59 PM
Not being a natural colorist, I also needed help in sorting through the contradictory morass of advice and theory in dealing with colour. After a chance encounter with a disney colorist at a party, this is what I did:

First throw away all secondary colors (Green, Purple, and Orange). They are unnecessary as they can be created from mixing.

Then invest in a one cold and one warm version of all your primaries (in oil painting this would be Yellow: lemon and cadmium, Red: Alizarine and cadmium, Blue: Cobalt and Prussian). Also black and white.
When you place these round a color wheel you will see how keen they are to mix. E.G the cold Lemon yellow and the cold Prussian Blue will make a natural green).

Introduce into your vocabulary the polarities: saturated/unsaturated, warm/cold, light/dark, complimentary.

You can see these issues applied in the Rubens painting:
He has contrasted the green figure with a complimentary red background.
The hands are more saturated than the cuffs (i.e. in a B & W photo the difference between them would be minimal).
Light tone is kept in the middle whilst the edges are kept dark (a favorite old master trick).
And (yes) he added complimentary to his warm yellow skin tone to take it down in saturation value as well as making it darker (if he had added black it would be darker but also 'deader').

These (as you can tell) are really oil painting tips but work as well in the digital. Photoshop gives us the added ability to digitally isolate all these components. PS 1 had the ability to convert to HSL format from where we could see the separate saturation values., Rubens would have KILLED for something like that! Even de-satuaring an image can give you useful feedback.

I briefly covered these points in a small tutorial thing I wrote: http://www.jackyoungblood.co.uk/chest_2.html

ekah
08-24-2005, 08:12 PM
I briefly covered these points in a small tutorial thing I wrote: http://www.jackyoungblood.co.uk/chest_2.html

Hey, how about a little warning for the squimish? :D

I saw your work in Expose. http://www.jackyoungblood.co.uk/pic01.htm Very impressive work. :)

Jack Youngblood
08-24-2005, 09:03 PM
Hey, how about a little warning for the squimish? :D

I saw your work in Expose. http://www.jackyoungblood.co.uk/pic01.htm Very impressive work. :)


Woops. Sorry. And thanks.

jmBoekestein
08-24-2005, 09:35 PM
:surprised

I- *... I just wanted to say that in physics complementary colours bring eachother into existence through wavelength interactions and quantum stuff with molecular things... wow that's gruesome... uh, you can use any set of complementary colours as long as you approach the rela environment we live in. Oh yeah, ambient light bouncing around gives shadows a less saturated effect sometimes due to negation of different wavelengths, or sometimes obvious colour bleed happens. It's good to carefully observe everything.

But I'm telling you getting into colour's a pain in the ass if you don't know about value in the image, I stepped back from colour after complicating it all too much for myself.

Chris_Beatrice
08-26-2005, 03:56 PM
It is very difficult to sort through all the contradictory information out there. Most of the reason we see all this contradiction is that many artists describe the way they CHOOSE to handle color (or the way one of the "old masters" chooses to handle color) as the RIGHT way to handle color. It's really amazing. If the person talking is an accomplished artist, or the artist you are using for inspiration is accomplished, you can follow his/ her "color theory" and you will likely get good results. But that is still only a tiny fragment of the whole story.

When you hear rules like the one about an object's shadow (or shaded side) having some of its complimentary color in it, or whatever, you should always ask, "why?" Is there a logical explanation for this? You should also definitely look for empirical proof of this. What you will find is a couple of artists or paintings that absolutely exhibit this, but you will also find a lot of great art that does not.

A lot of beginning artists think that "realistic" painting is simply about understanding some inherent rules about light and color, and so there is a right answer for how to use color. There is, absolutely, if you want to make pictures that look like photos. But don't be misled into thinking that those are the same "rules" followed by the "old masters" or any other great artists. Even though their work, and handling of color may look "realistic", it's not scientific in the way a photo or 3d rendering software works (thankfully!).

This is the only source I have encountered on the web or in print that expounds light and color principles, how the eye works, etc., and also applies them to art-making in a comprehensive, well-informed and unbiased way (sometimes the site seems really slow to load, unfortunately):


http://www.handprint.com/HP/WCL/wcolor.html

Though the focus is on watercolor painting, the in-depth section on color perception is very much worth reading. It will NOT tell you what YOU should do, or may want to do with color in your art, but it WILL debunk many of the myths out there, and even explain how they arose. It will also, I think, convince you that color is the most relative aspect of art, which is at the root of much of the "confusion."

Enjoy,

Chris

ekah
08-26-2005, 07:25 PM
When you hear rules like the one about an object's shadow (or shaded side) having some of its complimentary color in it, or whatever, you should always ask, "why?" Is there a logical explanation for this? You should also definitely look for empirical proof of this. What you will find is a couple of artists or paintings that absolutely exhibit this, but you will also find a lot of great art that does not.

A lot of beginning artists think that "realistic" painting is simply about understanding some inherent rules about light and color, and so there is a right answer for how to use color. There is, absolutely, if you want to make pictures that look like photos. But don't be misled into thinking that those are the same "rules" followed by the "old masters" or any other great artists. Even though their work, and handling of color may look "realistic", it's not scientific in the way a photo or 3d rendering software works (thankfully!).

This is the only source I have encountered on the web or in print that expounds light and color principles, how the eye works, etc., and also applies them to art-making in a comprehensive, well-informed and unbiased way (sometimes the site seems really slow to load, unfortunately):


http://www.handprint.com/HP/WCL/wcolor.html



The site you refer to is the same site I posted in my message posted above, and it illustrates how light and color behaves in the real world, not in any person's painting. Whether you consider that empirical evidence is up to you. What one does with that information and how one applies or interpretes that information to one's painting should be left up to the artist. That's what makes the artist different from any computer software. The "handling of the color" by "3D rendering software works" are only "scientific" if the artist chooses to make it so. The same goes for 2D painting programs.

I don't think that anyone is misled into thinking that these are the same rules used by the "old masters". The comment about the "old masters" not having used those "rules" or "scientific" info is moot since a lot of the information we have today did not exist then. However, there is no doubt in my mind that the "old masters" were doing their best to paint the world they see as realistic as possible based on their observation of the real world and the available technologies of their time, but that's just my opinion based on what I've read about their process. Linear perspective was a technological advancement and a tool, or rather one of the "rules" used by artists during the Renaissance period. The difference is obvious when you compare them to Byzantine art. I don't think that DaVinci would have been opposed to learning about the science behind light and color had it been available to him.

A lot of beginner artists break the rules before even learning them. It's their choice, but I think it's their loss in my opinion.

Btw, I love your donkey image. :)

Chris_Beatrice
08-26-2005, 09:43 PM
I'm not sure where you're coming from with all that, but in case there was a misunderstanding maybe this will clarify what I meant. Science has come a long way in understanding how light works in the physical world, and for the most part 3d rendering software mimics this. There are many layers to this type of technology, of course, and fully mimicking reality is more expensive from a processing standpoint, obviously. 2d artists have used a combination of real science, pseudo science, as well as simply looking at their pictures and using their own eyes, to produce lots of great art, and also lots of so-called "color theory." Most anything you encounter called "color theory" especially with the tagline "for artists", is going to be extremely, extremely limited. But these are rarely billed that way, and I think that causes beginning artists to waste a lot of time. It's very difficult to know how much to believe, but also test, as well as that in the end the point for most artists is to find their OWN "voice."

How light and color work in a 2d image is somewhat different from in 3d. Moreover, most of the most well-known and admired artists of the past certainly do not simply mimic (3d) reality. Whether one wants to mimic reality or not is of no concern to me - I'm not making value judgments about that at all. One person may be a photo-realist, the other an Impressionist. All I am saying in a nutshell is many beginning artists simply don't see the difference between, say, a Rembrandt and a photo. The Rembrandt looks "real" (and more importantly, very powerful), so they think that must come exclusively from a precise understanding of the "science." Or, if Rembrandt says, "here's what I do when rendering shadows..." many beginning artists will believe that is what all "good artists" must do, because "that's how shadows work", because they don't yet appreciate the subjective and relative nature of artmaking, even "realistic" rendering.

In case there was any misunderstanding about this: I personally am a huge proponent of understanding the science of light, color, perception, so I as an artist can make informed decisions, and try things I might not have thought of on my own, exploit the way we see, etc., but also I use my gut a lot regarding color. I'm thrilled you have already mentioned that same site. It's awesome and in my experience unique.

And thanks for the kind words about my donkey picture.

-Chris

ekah
08-26-2005, 10:50 PM
In case there was any misunderstanding about this: I personally am a huge proponent of understanding the science of light, color, perception, so I as an artist can make informed decisions, and try things I might not have thought of on my own, exploit the way we see, etc., but also I use my gut a lot regarding color. I'm thrilled you have already mentioned that same site. It's awesome and in my experience unique.

And thanks for the kind words about my donkey picture.

-Chris

Thank you for that explanation. I don't think that we are in any disagreement. :) Yes, I did misunderstand your post, and my apologies for jumping into the wrong conclusion about what you said. Another lesson for me to think before typing. :)

Again, your donkey images rules! :D

Chris_Beatrice
08-27-2005, 12:06 AM
Heh, usually I'm the one who "types before I think."

Zoontjens
08-29-2005, 01:08 PM
thanks this helped me!

That site indeed is the best one I could find up to now..discovered it a year ago but did notread all of it...must-read-now...

vrexplorer
09-05-2005, 01:02 AM
Color scheme tool (http://forums.cgsociety.org/showthread.php?t=273299)

CGTalk Moderation
09-05-2005, 01:02 AM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.