General problem: glossiness (roughness) and environment

[Micha_cg]

Hallo,

I try to find an easy workflow to get a rough surface ...

http://www.asgvis.com/phpBB2/viewtopic.php?t=697

... and have tested the usage of a single reflective layer. Some months befor I have done this test with Maxwell. I have not got the real roughness effect (moon), but it was possible to set up a rough reflection so that it looks like a diffuse surface without any difference. Here my Vray test: the rough metal in blinn mode looks very dark - to dark.

Image 1: comparsion of a white diffuse material vs. a rough reflection layer


My experience from my exterior renderings is, that sometimes it help to get the right balance between background and GI env lighting, to set the background intensity at approx. 2.5x higher than the GI env.
The next image show a diffuse white amterial and a reflective (100%) layer material. The diffuse mat is lit by the GI env, the reflective mat by the background env. If I set the background at 2.5 and the GI env at 1, the materials looks approx. like the same.

Image 2: GI environment intensity 1 and background intensity 2.5


I have the feeling, there is a bug in Vray.

[Micha_cg]

General problem: glossiness (roughness) and environment

... an other problem is, if I use a plastic surface with fresnel reflection and use a low glossiness, than the edges are looking dark. OK, I know, that the diffuse layer is dimmed there the fresnel reflection is bright, but if a glossiness is used, than the diffuse layer should not be dimmed. I think on a plastic sphere: if I would scratch it with a fine sand paper, than it could look like a diffuse material.

Image 3: strange fresnel dim effect at rough plastics

[Micha_cg]

General problem: glossiness (roughness) and environment

Hmm ... I don't know why, but the problem of image 3 is gone after reopen of the scene.

[Micha_cg]

General problem: glossiness (roughness) and environment

Interesting effect for image 1. My test are done per full adaptive sampling. But now, I have manual set the subdivs of the rough reflective layer higher as default 8 and get a bright surface. If I set subdiv 50, blinn and phong looks quite the same and if I set subdivs 100, the brightness is the same. So, the problem here is, that the adaptive sampling bring the strange effect of the dark surface.