I posted a bunch of concept work for the game "Hinterland" in the FINISHED section.http://www.conceptart.org/forums/sho...d.php?t=138847
Here's a step by step of one of those images.
I call firsties!
My SketchBook http://www.conceptart.org/forums/sho...d.php?t=139784
http://www.conceptart.org/forums/sho...d.php?t=192127"Everything must serve the idea. The means used to convey the idea should be the simplest and clear. Just what is required. No extra images. To me this is a universal principle of art. Saying as much as possible with a minimum of means."-John Huston, Director
I've never even considered starting out so mushy to nail the action - thank you!
Was about time a pro dumps something in this section!
Really?i work in Lab mode (sadly, not supported by Painter, only by Photoshop) because it organizes the color (and value) space much more like the way the human eye does.
Could you be a lil more specific about the lab mode?
I think i heard Mathias Verhasselt say he used it too but i must admit that when i tried it, i got some very strange (ugly) "marks" of color especially in my darks.
How d'you deal with that?
Also even if i only tried painter 3 (it came bundled with my wacom tablet) i really envy its way of mixing color as opposed to my Photoshop CS2.
Im very curious to hear more on the lab mode from you as you seem to enjoy it.
Thank you, i've been reading Handprints section on perspective for a little while now and indeed its a very informative ressource.
I am also aware of what you said regarding color theory, however my question was only referring to what you said about the lab mode in photoshop.
Sorry if it seemed like I ducked (or missed) your question. The modern standard for colormetrics is Lab, because it measures the color space perceived by the human eye in a way that is more like the way the eye works, because of how it evolved. That's why my response referred to color theory. When I say "color theory" I don't mean color wheels, or the various systems that artists have devised over the years to organize color in their art.
Do give you a blatant example of why Lab is more like the eye than RGB is - in Photoshop, make a bunch of "pure" colors, e.g. pure blue (0,0,255); pure red (255,0,0); green (0,255,0) and yellow (255,255,0). It doesn't matter whether you're in RGB or Lab mode (but you can't be in CMYK). Now, desaturate those colors by using image>adjustments>desaturate. Now they are all 50% gray - THE SAME VALUE. Are they really all the same value to your eye, though? If you use the color picker, and look at the values in the upper right of the dialog (L, a, b), you'll see that the blue in fact has a luminosity of about 30, the red about 54, the green 88, and the yellow 98 (almost white). If you switch to Lab mode and then run the same desaturation routine, you get different results. The blue is the darkest, followed by the red, then the green, and finally the yellow.
This may not seem like a big deal, but almost every color adjustment you make in Photoshop, including all the layer compositing methods and mark-making compositing methods rely on merging different color values. If you're merging the RGB values you are way off from how the colors actually look (e.g. yellow is a really light color, blue is dark). Why is this? Because RGB is simply a measure of how active the three phosphors in your computer monitor are - that's it. It has nothing to do with what those colors actually look like.
A lot of people think the eye, too, has three different kinds of cones: Red sensitive, Green sensitive, and Blue sensitive. This is only partly true. Originally the mammalian eye had only two types of cones: short wave sensitive (basically blue) and long wave sensitive (basically yellow). Most mammals still only have this, hence they are red/green colorblind. In humans, though, the yellow cone evolved and sort of "split" so that some are a bit more sensitive to longer wave lengths (red) and others are more sensitive to medium length waves (green). But it's critical to note that a) this is a very subtle split, and b) all the cones are in fact sensitive to almost all the wavelengths. There are exceptions though, e.g. the blue cones cannot get the very highest wavelengths.
Why does all this matter to artists? First off, yellow is a unique color. For you to see yellow, it's not just that both R and G need to be stimulated approximately equally, but they both need to be stimulated A LOT. This is not true of other colors. If Blue and Green are stimulated approximately equally, you'll see a bluish green, no matter what. If they're only stimulated a bit, you'll see a dark blue green. If they're stimulated a lot, you'll see a bright blue green. This cannot happen with yellow. There is no such thing as "dark yellow." If you try to make a dark yellow, i.e. hue approximately 60, then bring the brightness down, it starts to look green. If you move the hue up and down, you get a variety of browns and greens. Your colors go from red to green, without ever going "through" yellow. Yellow cannot exist toward the middle of the standard color wheel. In Photoshop, if you have a color that is really saturated, and you want to increase it's true value, at some point it will have to shift to yellow (hue) because other hues cannot exist at those high values AND high saturations.
In the end you see that color is not at all symmetrical, as it is typically shown in the standard artist color wheel type of paradigm. There is a whole palette in the warm area of the wheel, so to speak. You can make an entire painting just using warm colors where a true neutral gray will read as blue. You can't do this on the other side of the wheel.
Why? Because, going back to the eye - our eyes "care about" warm colors more than cool (long more than short). We're hyper sensitive to warm colors. As noted re the uniqueness of yellow, we can make a new color not only by the *relative* levels of the long wave cones, but by the total overall light as well. There is sort of a third dimension in the warm colors.
And lastly, in art making and visual perception, value is king. Value is what is critical in terms of how the eye reads the world, whether a 2d image or the real thing. To see how this works, take any painting or photo you want, and in Photoshop turn it to grayscale. It still looks pretty cool. Now turn it to color only, by making it a color layer, and putting a 50% gray layer under it. The result is almost no image at all.
So value is key, and color is value (e.g., again, yellow is a much lighter color than blue, because the perception of yellow in the eye is red and green at max stimulation).
Most artists' color theories desperately try to apply a fabricated kind of symmetry to this thing (the eye) that is, like most natural phenomena, a complete "mess." Most artists assume that true complimentaries (colors that truly cancel each other out) should also be mixing complimentaries (when you mix those two pigments they should produce a perfect gray). The Lab organization still has it's problems, but it is based on a much stronger analysis. That's not saying much, I mean, RGB doesn't even try to be. Again, it is just a numeric representation of how the monitor works.
Lab works like this:
The "L" dimension represents "luminosity", and tries to closely approximate how light or dark a color appears to the eye. So as noted yellow is close to 100, while blue is fairly dark.
The "b" dimension is the blue vs. yellow axis, which, surprise surprise, is inherently also a dark vs. light dimension.
The "a" dimension, which I think of as "horizontal" is the red vs. green axis.
I hope that helps!
Here are a couple of examples to illustrate the above.
BTW, with the new upload manager (well, new-ish) I can't figure out how to intersperse text with images (i.e. in between images). Does anyone know how to do that?
Target attachment url example:
Use this vBCode:
Text before first image
Added text after first image
More text added
Hope that helps, and thanks for all the helpful information!
First off thank you very much!
Ive been too lazy and should have followed your advice and gone back to color theory instead of stubbornly asking about lab colors. Sorry for that.
I havent even finished to read half of "the dimensions of colour"...
About the content of your reply, not much to say except that it enlightened me!
Ive only been painting for a year now but rather quickly i realised there was something wrong with the RGB grayscales.
Definitely not.In Photoshop, make a bunch of "pure" colors, e.g. pure blue (0,0,255); pure red (255,0,0); green (0,255,0) and yellow (255,255,0). It doesn't matter whether you're in RGB or Lab mode (but you can't be in CMYK). Now, desaturate those colors by using image>adjustments>desaturate. Now they are all 50% gray - THE SAME VALUE.
Are they really all the same value to your eye, though?
You cant imagine how many times i followed that very process and stopped short at the result wondering what the hell was wrong.
Then i would usually blame it on the fact that my eye was probably not trained enough to recognise the actual value of a color...
Yet when sketching with colors i would instinctively use orange then yellow to increase the value of a warm highlight.
This was the missing piece of the puzzle haha! thanx.
One last question and i stop hijacking your thread, what about the weird marks in lab mode i mentioned earlier?
About the attachment manager, i know Bartdeco beat me to it but i think there is an easier way. Just click the lil paperclip and choose wich attachement you want to insert in your post.
Once again thanx for taking the time to make those posts, i aprreciate it.
Hell yeah!I hope that helps!
If you ever feel like it, feel free to drop by my thread and tear me a new one.
Thanks both for your help with the attachments!
Re the odd marks in Lab mode, I really don't know what you're referring to. It doesn't seem to make sense that the color mode would cause this. Perhaps you were working with a particular kind of layer compositing, or type of brush? Can you post something to show me?
Alright here is an example of what i mean by "strange marks".
Used the same brush for both (flow at 70% and opacity set to pressure sensitivity) with pure black on a red background.
The difference, as you can see, is anything but subtle...