This actually brought up a really interesting idea for me. I noted that looking at most floors from above gives a generally diffused view, but when you look at it from a harsher and harsher angle, such as putting your head on the ground, the floor will almost start to look like a mirror.
I always wondered why this might be, but I think this image showcases why on the left hand side. The glossy has concentrated light rays, yet the diffused has very spread out ones. When you view the floor at a steep angle, you are allowing the floor to have less and less ability to spread out the light rays before they hit your eyes. This is a similar concept to how polygon faces will interact with light based on their angle to the light source. This was a super cool idea and I was thinking that it might actually be relatively easy to implement, since I think it's a byproduct of a rendering technique we've already learned. Super cool!
el-refai
This explanation of microfacets also explains why really smooth surfaces are glossier/can even have reflections. By sanding a surface to remove these microfacets or at least have them all on the same level that makes the object glossier as the microfacets don't reflect the light in odd directions.
doihead
When trying to render these materials with raytracing, it seems like it would be incredibly expensive to send out enough rays so that these effects become visible with monte carlo estimates. Would it be possible to average this data somehow or encode it such that it can be rendered with just a few samples?
Refangs
I never really gave much thought as to why different materials or surfaces had glossy or diffuse looks. It's cool to see that kind of explained with the distribution of microfacet normals being the primary factor (if I'm understanding correctly)
This actually brought up a really interesting idea for me. I noted that looking at most floors from above gives a generally diffused view, but when you look at it from a harsher and harsher angle, such as putting your head on the ground, the floor will almost start to look like a mirror.
I always wondered why this might be, but I think this image showcases why on the left hand side. The glossy has concentrated light rays, yet the diffused has very spread out ones. When you view the floor at a steep angle, you are allowing the floor to have less and less ability to spread out the light rays before they hit your eyes. This is a similar concept to how polygon faces will interact with light based on their angle to the light source. This was a super cool idea and I was thinking that it might actually be relatively easy to implement, since I think it's a byproduct of a rendering technique we've already learned. Super cool!
This explanation of microfacets also explains why really smooth surfaces are glossier/can even have reflections. By sanding a surface to remove these microfacets or at least have them all on the same level that makes the object glossier as the microfacets don't reflect the light in odd directions.
When trying to render these materials with raytracing, it seems like it would be incredibly expensive to send out enough rays so that these effects become visible with monte carlo estimates. Would it be possible to average this data somehow or encode it such that it can be rendered with just a few samples?
I never really gave much thought as to why different materials or surfaces had glossy or diffuse looks. It's cool to see that kind of explained with the distribution of microfacet normals being the primary factor (if I'm understanding correctly)