variable specular in chromed metals

[sevenfingers]

Try a falloff map in the specular channel. One with the scratched map, one with a plain one.
Should be fairly simple.

[Dogway]

Thanks for the reply.

The plain one for the actual pot you mean?

I did like that. I setted a map with scratches in all directions (so it would be applyable from any point of view) and then fall off. The fact is that between the ratio of influence of the glossy highlight, sure I get the scratches shining, but not only perpendicular ones, but vertical and all of them drawn in the map.

Am I misunderstanding? or is that simply?

I have some problem with max and cant give u a render, but hope it was understandable

[soulburn3d]

What you're trying to do is anisotropic reflections. Check out these tutorials for info on the subject:

http://www.neilblevins.com/cg_education/aniso_ref/aniso_ref.htm
http://www.neilblevins.com/cg_education/aniso_highlights/aniso_highlights.htm
http://www.neilblevins.com/cg_education/brushed_metal/brushed_metal.htm

- Neil

[Dogway]

Thanks Neil. I actually had seen your webpage, and it was really helpful, to understand the metals reflection behavior. Well, I have been thinking if it had to be with anisotropic. in fact the reflection (of the pot) is not distorted as long I can see. And furthermore the scratches are visible. I mean when it comes to a close up render. I wonder if you can set the anisotropic only to the specular (not reflections) and so, make visible those scratches, which have to be, as I understood, perpendicular to the highlight focus.

[soulburn3d]

in fact the reflection (of the pot) is not distorted as long I can see. And furthermore the scratches are visible. I mean when it comes to a close up render. I wonder if you can set the anisotropic only to the specular (not reflections) and so, make visible those scratches, which have to be, as I understood, perpendicular to the highlight focus.

A specular highlight is just a reflection of a bright light source, hence you can't have the anisotropic qualities of a surface only affect the specular in the real world, because the specular and reflection are exactly same thing. This could certainly be set in cg to affect each component differently, but you'll have immediately made a huge disconnect between how your image reacts and how the realworld reacts, and that may not produce the results you're after. And if you look at your example image, say at the brown line going up the wall in the center of the reflection, you'll note that the straight line edge is being distorted by the scratches in the same way the light source is being distorted, it's just because the light source is so much brighter you notice it more.

There's an example in that tutorial that discusses distance and the anisotropic qualities of a surface, in general the closer the thing you're reflecting is to the object, the less distorted it becomes, since the light source is far away, it's one of the most distroted things.

Also, because the light source is striking the pot right at the edge of the lip like that, the reflection is distorted even further due to the nature of the sudden change in the shape at that specific spot. If you moved the lightsource up so it was hitting the surface more on the flat area you'd probably see a less pronounced effect on the light.

In general, start with a chrome shader, make a map of the scratches that exist on your pot (both the larger scratches and the tiny scratches that are probably going around the potlid radially), use this map in the bump slot of your material which will affect both the reflection and the specular, and then play with the brightness of the specular quality of your light (possibly have a seperate light only for the specular highlight sitting on top of a light that only produces diffuse illumination, so you can affect their multipliers seperately), and you should be able to get an effect that looks very much like what you're seeing.

Hope that helps.

- Neil

[CGTalk Moderation]

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.