Skylight: Need to have the mystery solved

[Alex Chow]

EDIT: Ron Lemen's example is actually the third picture now. It didn't appear before the edit.

This has been bugging me for quite some time. It's regarding the depiction of the blue skylight. Frankly, I'm very confused because I may be misunderstanding my sources on this.

As far as I understand from the multitude of sources I've read, the sky is a light source in a shape similar to a dome above us. What I don't understand is why there can be sky speculars and areas where its influence is greater. Not necessarily talking about the planes where the "dome of light" is unable to hit.

I've put up two illustrations demonstrating this; the third image is from Ron Lemen's tutorial and the first is from the PSG tutorial. Ron Lemen's example acknowledges the existence of skylight, but the right side of his face is conspicuously more intense. PSG's example also shows this happening but I'm pretty confused since, by theory, the yellow, green, and purple beams of light I drew in (amongst infinite numbers of it) will cause a sky specular. I've also put up a photo of this happening in real life. The sunlight hits the right side of her face but, again, skylight shows itself the most on the right side of her face. My assumption, given what I know, would be that the skylight would hit evenly across her face creating a flat form shadow across the right side of her face.

That, however, looks awful. Of course, I could just accept this idea as law and incorporate it everywhere in my drawings but I want to know what the heck's going on in fear that I might (see: will) misuse it. My first guess is that this is similar to the fresnel effect with swimming pools, where the reflection is affected depending on the viewing angle. My second guess is that somehow, somewhere in the "dome", the skylight is stronger. My third guess is that skylight isn't a dome at all. I don't want to jump into conclusions, however.

Thanks in advance.

EDIT: Messed up the directions of the light for the photo.

[briggsy@ashtons]

What you see is the additive combination of the light zones created by the two light sources, sun and sky. Within the area lit by the sun, the light zones created by the sky are almost completely swamped. Take a shiny sphere outside and alternately shield it and expose it to direct sunlight - you'll quickly see what's happening.

[frog from itchy]

Pretty much as briggsy says, the skylight is a more or less even dome but the direct sunlight completely overpowers it because it is just so much brighter.

Maybe if I get a chance I'll post a simulation created in a 3D package.

[Alex Chow]

What you see is the additive combination of the light zones created by the two light sources, sun and sky. Within the area lit by the sun, the light zones created by the sky are almost completely swamped. Take a shiny sphere outside and alternately shield it and expose it to direct sunlight - you'll quickly see what's happening.

Just to clarify, that would mean any planes in which the sunlight has even somewhat of an influence (even if it's very weak on those certain planes) will be enough to make the skylight negligible, right?

If what you say is true (which I'm sure is true since you wrote huevaluechroma.com and itchy, I'm assuming, wrote the tutorial at itchyanimation), does that mean the darkest planes on the right side of her face and the left side of the guy's face are still influenced by sunlight (albeit the regions take in less than the planes facing the sunlight)? Am I being tricked into thinking they are planes in which are not affected by sunlight simply because they're the darkest values in these particular examples?

If both of these questions end up being "yes", it completely changes my perspective on how I observe lighting since I can apply this knowledge in similar situations (see: 99% of situations) where a strong light and a noticably weaker light(s) is playing a role.

One more probably-stupid-but-worth-noting question. Ambient lighting is even weaker than both the sunlight and skylight. Does that mean that it can only be noticed in areas where it isn't completely swamped by either the sunlight or skylight sources?

[Jens]

Everything in that scene is affected by sunlight, since in that scene the sun is the only lightsource. You don't see shadows, all you see is the light that's reflecting off your subject into your eyes. ..

The entire color of the shadow side is base color from skin + reflected light from sky, + reflected light from other sources. You don't see what going on next to the kid, so there could be light reflecting of a screen for example, there's light reflecting off her hair, the blue towel etc.

Try the thing with the sphere , but also put something behind it, so that's there's a different kind of reflected light is hitting the shadow side. Put your hand behind it, a red screen and see what happens to the color of the shadow side.

edt: the tutorial on itchy is made by Anders Niklas Jansson, aka Prom http://www.itchstudios.com/psg/

[Alex Chow]

I know it's been answered 3 times, but I still don't get it. I took countless photos, tried to justify what's happening, though, I ultimately ran back into the same issue.

Here are some photos with edits on the best examples I could take. My finger is more than 5 feet above the ground (makes ground reflected light almost negligible) and not near anything like a car or telephone pole. Also, the reflected light cannot be from my hand because it would have increased the saturation of the left side of the finger, not lower it (which is what should happen when the sky blue hits the orangeish skin)

The first photo is the uneditted (but cropped) photo of my finger.

The second is what happened when I covered my finger from the sunlight.

The third is my analysis, or at least my flawed perception of this. I cannot justify the existence of the dark region in between the area lit by sunlight and area lit by skylight.

The fourth is an editted version of what I would have done if you told me to paint a finger under these conditions.

Thank you for all the help thus far, but clearly I am still having trouble understanding this. Hopefully, my concern is more understandable with this post.

EDIT: Photos didn't appear. Now they are here.

[briggsy@ashtons]

Sometimes an apparent dark band like that can be simply a contrast effect, but in this case it looks like there definitely is some more light hitting the left side of your hand. Without being there, I can't say definitely what's causing it, but my guess is that most of this is reflected light from surrounding objects (perhaps excluding the fingernail, where I think I see reflected skylight). If I was an ant sitting on the shaded side of your hand, what would I see? Where would the strongest light(s) be coming from? Are there distant sunlit objects off to your left reflecting more light to your hand than you realize? Even if your hand was surrounded by a completely random assortment of planes, you might expect an effect something like this, because the planes to your left would tend to face the sun, and so reflect more light to your hand than the others.

[h2rra]

You can't forget subsurface scattering if you are studying thin fingers facing direct sunlight :P That's why you have this saturated core shadow and yellowish-red fingernails. It seems that you somehow blocked alot of reflecting light in the second picture.

[J Wilson]

That's your core shadow. You have the sun as a light source on one side, and the sky coming in from the other. In between you have that narrow band where neither has as much influence because they can't quite reach it.

Why doesn't the sky light the core shadow too? Who says the sky lights everything evenly? Part of the sky could be cloudy, but I'm guessing it's as simple as not as much light is reaching your finger from the direction the camera is shooting from. First of all, YOU are there blocking out part of the light. Also I'm guessing there is probably a house or line of houses behind you. You are facing a street, and light travels unhindered down the street, but there tends to be things on both sides of the street that might block out some of the intensity of the light.

In my opinion, it's not so important to understand every light source in a scene perfectly, only to recognize them when you see them. In the photo of the young girl for example, I don't need to know what it bouncing light into her shadows, only to recognize that SOMETHING is and treat it as such.

[briggsy@ashtons]

Amaranth, have you tried shielding your hand in various directions to establish exactly where that light on the left is coming from? Use something black.

[Alex Chow]

Amaranth, have you tried shielding your hand in various directions to establish exactly where that light on the left is coming from? Use something black.

I actually did another photograph session but used an orange instead. There is actually another light coming from the left that I thought wouldn't be a factor, and that's a white garage door around 4 meters to the left reflecting the sunlight (at least I believe it should be the secondary light). I do not have my camera here right now to place the photos onto my computer but I'll do so if I really need to.

Whoops. :S

Wilson may also be correct that I could very well be blocking the skylight, or h2rra could be right in that it's subsurface scattering (though I was under the impression that SS should "glow" a brighter value than that).

I'll try to bring this altogether now to end the "mystery".

From what I've read, this means that Ron Lemen's image of skylight (the third image on the opening post, with the skylight clearly affecting the left side of his head intensely) is a stylized depiction to give viewers a clearer idea of skylight's existence unless, as Wilson suggested, that there's something partially blocking it from several angles. Prom's depiction of skylight speculars (first image) can only exist on planes which is not affected by the very intense sunlight or even by nearby objects reflecting the sunlight intensely enough, right? The general conclusion is that something is reflecting light onto the right side of the girl's face and the left side of my finger.

And J Wilson, I think it's very important for me to understand every lighting factor contributing to how an object looks because, at my stage of the art game, I have to reduce my misconceptions as much as possible (hopefully to nothing) before I start twisting the rules in my own illustrations, even if it means I have to analyze the entire environment again and again to understand every light source perfectly. I would have not double-checked the environment I took the photos of my finger in if you guys didn't question the left lighting on the finger. This just means that if I were to draw a building around 4 metres to the side of a character now, I'll know to make sure that it should contribute to the overall lighting. I dismissed that as "light which wouldn't possibly reach me" and further studies proved me wrong.

This thread has helped me tremendously to do that and it motivated me to make specific studies on lighting. Thanks everyone.