Roughness at glancing angles Vray vs. Modo

[GidPDX]

Does Vray already calculate miro roughness at glancing angles?

I was using a gradient to have the roughness fall off at glancing angles, but it's producing dark edges around my material (which is a metal).

Maybe this is correct and MODO is wrong? I have no idea, it's just different between MODO and Vray and I've narrowed down the issue to this use of a gradient to control roughness across the surface as the culprit.

If I just switch it off for Vray and use the roughness settings on the vray material it more closely matches MODO's shader with the same roughness controlling gradient on.

Thoughts?

[vladimir.nedev]

Does Vray already calculate miro roughness at glancing angles?

I don't understand what you mean here.

I was using a gradient to have the roughness fall off at glancing angles, but it's producing dark edges around my material (which is a metal).

Maybe this is correct and MODO is wrong? I have no idea, it's just different between MODO and Vray and I've narrowed down the issue to this use of a gradient to control roughness across the surface as the culprit.

If I just switch it off for Vray and use the roughness settings on the vray material it more closely matches MODO's shader with the same roughness controlling gradient on.

There are differences between how the MODO material and the V-Ray material do glancing reflections. From what I remember:
- in the MODO material, the glancing reflection amount is a gray color with intensity equal to the "fresnel" channel
- in the V-Ray material, the glancing reflection is just the reflection color

Maybe that's what you are seeing ? Can't be sure without an example scene.

Also, at the moment the V-Ray material doesn't have "glossy fresnel", so with high roughness (low glossiness) values, it will (in most situations) reflect more light at glancing angles, compared to the MODO material.

Greetings,
Vladimir Nedev

[vandigital]

I do believe this is a shader model limitation that Allen over at luxology was able to overcome in 901? You also wont see edge darkening in maxwell materials something they fixed a few years ago. Had to do with energy loss or some such... idk ill have to look it all up again. But its defiantly something that can be fixed on the dev side. I think Luxology referenced the Disney models for their material.

Not the same exact problem here but maybe some insight:
https://eheitzresearch.wordpress.com/240-2/

[GidPDX]

Vladimir-

I think what I'm seeing is what you're stating about the reflection color and Fresnel. I'm lighting with an HDRI map that has some very dark areas, and when roughness hits zero at the glancing angles, I'm seeing that dark BG in Vray vs. in MODO where the Fresnel kicks back in (or something of that nature).

If I just the reflection glossiness settings on the material directly, the dark edge predictably goes away.

vandigita

I think you're right and I do recall that being talked about when Allen put the new shading model in.

It's taking me a bit to get use to creating metals in Vray with the material shader. I can get "a metal" but getting the exact look can sometimes be a bit odd. This is half just down to my inexperience with Vray, and half being that the IOR isn't really tied to anything I can reference. 22? 58? 100? These are weird settings.

I do like a lot of other things about the shading though.

[John_Do]

I suggest you to don't use the IOR when you're creating metals, unless if you can use K number also. The fresnel IOR gives incorrect results above approximatively n=8.

example ( copper ) :