What "color" are shadows?

[rr113]

Normally I would say shadows are a darker version of whatever color they fall on. But I've been puzzled by the street next to mine with so-called ""sodium lights that are orange. I understand they are a pretty pure orange. The shadows that they cast on the sidewalk, which is a warm light gray, are definited blue, the complement. This reminded me of those watercolors with violet shadows, the complement of yellow sunlight. I always thought that the violet shadows were made up, but now I'm wondering about the "theory of shadows" and why those shadows are blue.

Richard

[LadyDreamkiller]

I'm actually glad to see this question posted, because it's one that I had been wondering about myself. Especially since laying in shadows on my paintings is something that I Seriously need to work on.

[jdadson]

Originally Posted by rr113

Normally I would say shadows are a darker version of whatever color they fall on. But I've been puzzled by the street next to mine with so-called ""sodium lights that are orange. I understand they are a pretty pure orange. The shadows that they cast on the sidewalk, which is a warm light gray, are definited blue, the complement. ...Richard

That's an optical illusion called simultaneous color contrast. If you think about it, the "real" (aperture) color can't be blue if the only light illuminating anything is orange. (Oh my. Did I dangle a participle?) It is hard to overstate just how dramatic simultaneous color contrast can be.

There have been at least two long threads on shadows recently.

[Einion]

Originally Posted by rr113

Normally I would say shadows are a darker version of whatever color they fall on.

Hi Richard, generally that's exactly right. In addition they tend to be lower in chroma - duller.

Originally Posted by rr113

But I've been puzzled by the street next to mine with so-called ""sodium lights that are orange. I understand they are a pretty pure orange. The shadows that they cast on the sidewalk, which is a warm light gray, are definited blue, the complement.

Yep, I've noticed the exact same thing myself - but only when there is a secondary source of illumination such as the evening sky. When it's fully night and the sky is dark (no moon or moonglow) the shadows are just that horrible brown/grey colour you'll be familiar with.

Originally Posted by rr113

This reminded me of those watercolors with violet shadows, the complement of yellow sunlight. I always thought that the violet shadows were made up, but now I'm wondering about the "theory of shadows" and why those shadows are blue.

This topic is a very contentious one here as there's a lot of confusion in some posts between what is actually there and what is seen to be there when you want to look for it and how you then choose to paint it (which has more to do with the second than the first ). Anyway, here's a handful of good previous threads that you might like to go through:
http://www.wetcanvas.com/forums/showthread.php?t=268000
http://www.wetcanvas.com/forums/showthread.php?t=265446
http://www.wetcanvas.com/forums/showthread.php?t=259369
http://www.wetcanvas.com/forums/showthread.php?t=257430
http://www.wetcanvas.com/forums/showthread.php?t=256279

Einion

[jdadson]

Check this out. The blob in the middle is exactly the same hue as the bright orange.

[jdadson]

All other things being equal (which they never are), a shadow will be the same hue and saturation as the local color, but less bright.

[LadyDreamkiller]

That was a really ODD example, Jive - at least for me. When I looked at the "blob" what I saw were blue toned greys first, and then brown-grey tones when I looked again.

[jdadson]

Originally Posted by LadyDreamkiller

That was a really ODD example, Jive - at least for me. When I looked at the "blob" what I saw were blue toned greys first, and then brown-grey tones when I looked again.

When I look at it, sometimes the blob looks blue around the edges and brown inside.

If you google for optical illusions, you can find some that are outright astonishing.

[WFMartin]

Originally Posted by jdadson

That's an optical illusion called simultaneous color contrast. If you think about it, the "real" (aperture) color can't be blue if the only light illuminating anything is orange. (Oh my. Did I dangle a participle?) It is hard to overstate just how dramatic simultaneous color contrast can be.

There have been at least two long threads on shadows recently.

That's EXACTLY what causes it. Yes, they do appear blue (or "cyan-ish"), and that's the way they OUGHT to be painted, because that's the way your EYE, or anyone else's eye would see it, under those conditions.

The appearance has nothing to do with extraneous light. Although a "blue", extraneous light source COULD cause the appearance, it can be equally (or even MORE) apparent with only one, orange light source available.

Cameras won't see it that way. Film won't see it that way, Why? Because, as you say, it's an optical illusion, and that takes place in your eye/ brain. No, it's not actually blue--it isn't there! But, since it looks as though it's there, should you paint it? It's your choice.......but I certainly would!!!

That's what separates the "copying-from-a-photo" artists from the "copying-from-a-photo-while-still-making-it-appear-real" artists! I do my best to be the latter.

Bill

[Einion]

Originally Posted by LadyDreamkiller

That was a really ODD example, Jive - at least for me. When I looked at the "blob" what I saw were blue toned greys first, and then brown-grey tones when I looked again.

Here's another, it's one of my favourite illusions because it's so pronounced but there are many others of a similar type.

Originally Posted by WFMartin

That's EXACTLY what causes it. Yes, they do appear blue (or "cyan-ish"), and that's the way they OUGHT to be painted, because that's the way your EYE, or anyone else's eye would see it, under those conditions.

Really?

Originally Posted by WFMartin

The appearance has nothing to do with extraneous light.

That's exactly right.

Originally Posted by WFMartin

Cameras won't see it that way.

Yes they do - both film and digital recording recreates colours in a scene overall with greater accuracy than any human painter.

...

It's a simple truth that if one paints an illusory effect as they appear to be you get the colour wrong. It's the actual colour of the area, in the context of the actual colour of the surrounds, that creates the effect in the eye/brain - so unless you use something close to the real colours you won't recreate the effect. I can't put this any plainer. Bottom line is, you paint the illusion as seen you distort/exaggerate the relative difference.

Does anyone else doubt that this is the case? I'd like to stop having to needlessly debate this so if you do doubt what I've said above is correct I'd like you to do a small bit of work to see whether it is or is not the case for yourself. It's not complicated and it's not too difficult but it will require a little effort on your part but I assure you you'll benefit from it if you wish to understand this sort of visual phenomenon correctly.

Einion

[WFMartin]

Originally Posted by Einion

Here's another, it's one of my favourite illusions because it's so pronounced but there are many others of a similar type.



Really?


That's exactly right.


Yes they do - both film and digital recording recreates colours in a scene overall with greater accuracy than any human painter.

...

It's a simple truth that if one paints an illusory effect as they appear to be you get the colour wrong. It's the actual colour of the area, in the context of the actual colour of the surrounds, that creates the effect in the eye/brain - so unless you use something close to the real colours you won't recreate the effect. I can't put this any plainer. Bottom line is, you paint the illusion as seen you distort/exaggerate the relative difference.

Does anyone else doubt that this is the case? I'd like to stop having to needlessly debate this so if you do doubt what I've said above is correct I'd like you to do a small bit of work to see whether it is or is not the case for yourself. It's not complicated and it's not too difficult but it will require a little effort on your part but I assure you you'll benefit from it if you wish to understand this sort of visual phenomenon correctly.

Einion

Well, Einion,

You and I've been debating this for a long time, and, this time, although I will forever believe that you're incorrect, I also believe I have FINALLY understood your premise.

If I am correct,... what you're claiming is that if one were to paint the colors of the light source, subject, and cast shadow in the EXACT COLORS that they are, in real life (ignoring the optical phenomenon), ON A CANVAS, that they will somehow actually CREATE the same illusion that was before your eyes in the original scene, as created by colored LIGHT??? By using paint on canvas? With pigments?

Is THAT what you are contending here??

Well, if I'm stating your position correctly, at least you can rest assured that I finally UNDERSTAND what you're talking about.

I certainly will never agree with your premise, but at least I now understand it, I think.

The reason that I don't believe your premise to be valid, is that I believe the effect to be much more exaggerated when caused by colors of LIGHT (the orange street light, and the cast, blue shadow, in this case) impinging on your eye, compared to the appearance of the three elements (the light source, the subject, and its cast shadow), of paint on canvas. I don't believe that the effect of pigmented paint on canvas, and illuminated by some totally uncontrolled museum light source is enough to activate the illusion. My simple opinion is that....I don't think so! I believe that the "effect" has to be imitated (and, yes, even exaggerated) by the painter, rather than expecting (hoping and praying) that the painting, itself, will miraculously display the phenomenon on its own.

Heck, I wish that WERE true. It'd make painting a LOT simpler, I think.

That is quite honestly, my opinion, and I can do nothing more than adhere to it, I'm afraid. I suppose we can just agree to disagree on that, because I don't feel confident enough in your contention to change my mind for it,..Sorry.

Regarding film recording the exact image, accurately. I have worked with films of all sorts in the litho trade for over 40 years, and among those color films have seen some of the greatest distortions of color known to mankind between various brands and types of films.

Camera color films are absolutely NOTORIOUS for DISTORTING colors imaged upon them. Among the name brands are Fujichrome, Agfachrome, Kodachrome, Ektachrome, and the list goes on and on, (with "indoor" and "outdoor" versions available for each, as well), and each and every one of them is generally selected by professional photographers specifically FOR their extreme color biases--ESPECIALLY in the the shadow (dark) areas. Given this premise, I could (using Fuji Film) prove to anyone that the cast shadow caused by an orange light, is "green", and just as easily, on Kodak Kodachrome film, prove to that same person, that the cast shadow caused by an orange light is "purple".

If there is truly one thing that a color film does NOT do is to "record an image accurately". And, I must say, with all due confidence, that is a FACT, not an opinion.

Seems as though we're not going to resolve our differences on this one, either, but at least I believe I understand your position on the first one, which I never have been able to, before. Sorry, but I'm just a dumb printer.

Bill

[LadyDreamkiller]

Ok, you've managed to prove to me, at least, once and for all that what the brain percieves and what is necessarily actually There can be rather far removed from each other, Einon!

Which means that for now, at least, I'm definately going to be doing some experimenting with simply using darker tones of the same color for laying in shadows. Once I have a chance to see how that's working out in my painting, maybe I'll consider using what the eye "sees" rather than what's actually "there" .

[Einion]

Originally Posted by WFMartin

You and I've been debating this for a long time, and, this time, although I will forever believe that you're incorrect, I also believe I have FINALLY understood your premise.

In all honesty Bill it's not like I made it hard to understand if you'd care to check back; I've said as much, in as many words, more than a couple of times I'm sure.

Originally Posted by WFMartin

Is THAT what you are contending here??

I'm not suggesting it's possible to get the colours exact, close is good enough. Want proof? Check some of the links I provided above.

Originally Posted by WFMartin

I certainly will never agree with your premise, but at least I now understand it, I think.

But what if you're wrong?

Originally Posted by WFMartin

My simple opinion is that....I don't think so! I believe that the "effect" has to be imitated (and, yes, even exaggerated) by the painter, rather than expecting (hoping and praying) that the painting, itself, will miraculously display the phenomenon on its own.

A small amount of tweaking to compensate for various things might well be necessary but some of the illusory effects are very pronounced - shadows on skin being way less blue than they look being the classic example perhaps. Again, this is more evidence in its own right if one looks at high-quality portraiture and figure painting.

Originally Posted by WFMartin

If there is truly one thing that a color film does NOT do is to "record an image accurately".

Who said that? Not I.

Originally Posted by WFMartin

Camera color films are absolutely NOTORIOUS for DISTORTING colors imaged upon them.

And painters don't?

Originally Posted by WFMartin

That is quite honestly, my opinion, and I can do nothing more than adhere to it, I'm afraid. I suppose we can just agree to disagree on that, because I don't feel confident enough in your contention to change my mind for it,..Sorry.

Did you miss this part?

Originally Posted by Einion

I'd like to stop having to needlessly debate this so if you do doubt what I've said above is correct I'd like you to do a small bit of work to see whether it is or is not the case for yourself.

Einion

[FriendCarol]

Throwing in my 2 cents -- yeah, I'm sure at some point in the past I've stated my agreement with Einion's position: Painting the colors as they are (not as they look!) has the best chance of reproducing the effect of the way they look. Yet I agree with Bill that the effect certainly works differently in paint.

If the painting replicates the illusory colors instead of the actual colors, it always looks very 'painterly' (non-representational) to me.

Perhaps a very slight tint of the illusory color helps a painting represent the colors of the scene more accurately, I don't know. I'd be interested in seeing a demo, Bill, if you have spare time to make one:
[a] simple scene (with shadows) painted as you normally would paint it (i.e., with strongly complementary shadows)
[b] the same scene with the shadows painted as closely as possible to 'match' the actual target color in the shadows (ignoring all that weird stuff that makes exact matching impossible)
[c] a compromise: just a touch of the complement in the shadows.

I honestly don't know what the result will be (i.e., which painting will look more 'accurate,' or more effectively evoke the scene), so it would be a very interesting experiment. Of course, one problem is we would not have a standard for comparison (photos would be worthless as the standard, I think we all agree?). But we could all simply report whether a, b, or c seemed 'truer' to us. How about a poll?

[LarrySeiler]

Outdoors in atmospheric light...I have learned that a warm light generally produces cool shadows...and a cool light warm shadows. Color will generate that feeling...

also...an object producing shadow is blocking the sun from striking an area. This very often allows the influence then of light and color directly above to make its presence known.

Some might miss taking into consideration also that color, light and shadow, various areas will read differently looking directly at it...from looking or taking it in peripherally.

The eye is often quite sensitive to take in hints of color not looking directly at an area, and those hints will vacate to allow more the local color to dominate when you shift your gaze to look at it directly.

This makes an interesting possibility. You could paint more local color at your subject of main interest...and paint more the peripheral hints of color, values and detail of adjacent less dominant elements/areas.

Larry