I want to replicate a velvet-type fabric on theatre seating. I don’t want to use something as render intensive as displacement even though I know that’s probably the best technique for cloth.
I was trying a noise map in the glossiness slot and it works somewhat but the variation between light and dark areas isn’t nearly as strong as it is in the real photos. Maybe I need a special normal map? Any suggestions on how to recreate this look?
I’ve never done velvet, but try using your noise in the anisotropy slot instead, and just keep the glossiness a constant low value. That will give you a stronger variation, and behave more interestingly/realistic when the camera moves.
Fastest way: falloff angular map in the diffuse slot, with 2 versions of the fabric texture; a darker one in the perpendicular slot and lighter in the parallel slot.
Thank you for your tips everybody.
@dgruwler - I like that anistrophy test you sent. Is that just a texture in anistrophy or are you also using a texture in glossiness?
@Grigio and Vizioen … I did consider the diffuse trick that you mentioned. And I will probably end up doing that. What I wonder though -- is that really what’s going on? Is the color different on the sides of the “threads” versus their tips? To me the color variation seems like it is coming from reflection differences since the lighting and direction has such a strong effect on how it looks. If it was just a diffuse component, the changes wouldn’t be so dramatic. Still, since I want “fast” I realize faking it with diffuse is probably not a bad idea.
Your “concrete” example looks quite good. BTW -- why are you running it through HDRI map rather than a regular bitmap? Is the texture actually an HDR or are you using some special function of the HDRI that doesn’t exist in bitmap?
i’m using vrayHDRI (it’s the equivalent of the native max bitmap but for Vray), because it’s better suited in conjunction with Vray and in terms of memory, and the filtering is a bit better (I tend to use elliptical) afaik. Someone might correct me though but that’s what I remember after some reading here and there. You could achieve the same result with a regular bitmap map here normally speaking. Filtering might be a little bit different though.
Thank you all…
Not to hijack my own thread -- but Vizioen -- are you saying VrayHDRI is more efficient with memory than Bitmap? And the filtering -- is superior to pyramidal and summed area?
- Chaos Forums there are more threads than this one, just search for elliptical.
@davision it’s just a texture in the anisotropy, with a constant glossiness. As for why velvet looks like it does, I don’t know exactly how it works in the microscopic level, but I do know that the result is anisotropy (velvet is often used specifically as an example of anisotropy), and so that should be the key to getting a realistic looking render, at least so far as the bright/dark patches go.
You can get very far in a still image with just the diffuse if you have a good eye, but it’ll never look right when the camera moves or repositions.
