I just realized I kinda knew how to shade.

[aegis]

http://www.youtube.com/watch?v=q33XmjqFPxU

edit: images.


before:

after:

[MidKnight]

Heh.
Big improvement over my attempt.

[Argh]

12 minutes. Could use about five more on some of it, imo.

[aegis]

I said shade, you know, like where there are shadows and stuffs
you textured it.

and what's with the rust?

[KDR_11k]

Don't use black lines for indents, that looks cartoony.

[Argh]

It's shaded, lol, down to shadows and a reflection I just added textures because it's so boring to shade it without considering the albedo / reflectivity of the material and some surface detailing.

[EDIT]

1. Never, ever apply a Contrast like that, you just washed it out and ruined a lot of the details, especially in the lens area. Use Curves, and use a little more discretion in the selection area, it's not like dragging a circle and doing Invert Selection is hard.

2. A shadow from a lightsource that's above and in front of the object would not cast that shadow.

3. You washed out too much of the color. Bare metal hardly ever looks like that.

4. If you wanted stronger glints on the reflection area, you need to hand-paint it, if you don't have matching textures at different lighting levels, and keep it subtle.

5. You didn't seem to get that I used two different lighting angles on the "metal", because they're at different angles to the light- different conic angles, basically.

6. Don't do a ninja-edit with moderator powers, that is confusing.

Anyhow, good try at fixing it up, my take was far from perfect.

[aegis]

if the light areas are too washed out, you need to get your monitor/eyes checked - I can still see the majority of the detail.

[Argh]

No, this monitor's color-calibrated for web, etc., and has a very good contrast.

[aegis]

I'd guess it's not calibrated right, unless you used expensive equipment to do so and it's an IPS panel

it's also possible you're sitting too high on a TN panel.

[Argh]

It's not my equipment, and not my eyes. You used a contrast, where you should have used Curves, and you did it in a way that didn't work. You lost a lot of color value! It's not rocket science, dude, contrast operations are invariably lossy.

If you wanted to heighten values, you needed to work it more subtly- just put the two side-by-side, and yours is clearly a radical departure, color-wise.

If you know what you're doing, we'd never see a difference in overall values, it'd just look better overall as you fixed goofs, like my failure to complete the deep shadow of the seam line, etc. Sorry if that seems harsh and all, it's just how it is.

[aegis]

I actually used neither.

[Argh]

So, you used a Dodge tool, or whatever. It all works the same way, guy. 8 bits of gray per channel, if you compress them or raise them globally, it's lossy, and you lose color values. Look at them side-by-side, please.

[aegis]

you lose color value the instant you change *anything* other than mapping channels to each other and layering over the top.

[Argh]

Yes, all edits are destructive, in that sense. But when you compress the channels, you are washing out the top or bottom end, which is why your version looks both a lot grayer and lighter, and lost a lot of blue.

Here... I guess I'll spend that 5 minutes... you'll see what I mean, I hope.

[Argh]

There. No main color values have been destroyed- i.e., the metallic blue is still correct, etc. I even added a little overspill of green light passing through the lens

[aegis]

it looks like some color was destroyed on the bottom shadow, though, and why doesn't the top of the disk have anywhere near the lightness of the metal part of the dome?

[Argh]

Yeah, it's impossible to avoid as you head towards black, you're compressing the available range, which is why black and white are both a bitch.

why doesn't the top of the disk have anywhere near the lightness of the metal part of the dome?

Because the outer circle is nearly flat from the POV. I didn't copy your donut, it was too easy.

[aegis]

why doesn't the top of the disk have anywhere near the lightness of the metal part of the dome?

the light doesn't totally make sense

if you had any other angle of light source above the object, the back wouldn't be so dark directly behind the obvious light source.

[Argh]

Do you need to see it in cross-section, to understand what I did?

Here, it's not a donut, again I'm sorry if that's confusing, I didn't know that we were supposed to just copy your design completely, and just reinterpreted the circles as entirely different angles. I'd do a donut version, but this was kinda boring.

[aegis]

*that* light doesn't make sense either.
the only thing really affected by light is the dome - you'd need a very odd light source to actually get that.

edit:
shading is affected by distance from the light source and shadows/reflections cast upon the surface.

since the disc has a shadow cast on it, it must be getting light from the light source.

however, it doesn't really have any variation in shading... so it's all the same distance from the light source?
but the only way to do that would be to put the light directly above the disc... which would remove the large shadow.