Bronze Patina

[binarycortex]

Paco had mentioned on IRC that a cavity shader would be pretty interesting. He offered up this example.

I got to thinking about it, and I was reasonably sure that this could be done with an ambient occlusion map. So I tried it.

I think it worked nicely. It shows up best on the swirls on the base.

[binarycortex]

Ok, another test. The only thing that changed is the second buddha and the room material is no longer the default glossy white, but a matte with an off white almost yellow color. This sped up the render significantly.

I think next I may try to have the AO map control the roughness so the darker areas are less smooth.

[paco]

Nice result

Actually it was Inlite who started this whole thing

[binarycortex]

Ok, It turns out you need to do some work on the AO map for this to work right.

Here is the original AO map.

I opened it in the gimp and used curves to add a lot more contrast to the map.

Next I applied a gaussian blur filter set to 10px, and used the levels control to bring up the white point a bit.

Which resulted in the following.

[binarycortex]

In order to control the roughness and have the patina less shiny, you need to make a copy of the contrasty blurred AO map and invert the colors.

If it ends up being too much, just adjust the color intensity as necessary.

[binarycortex]

Ok, I had to tone down the contrast of the inverted AO map for the roughness modulation. It works but I'm not sure I like the result.

The new inverted AO map.

The result.

[binarycortex]

And here it is without the roughness textured by the inverted AO map.

[binarycortex]

Ok, forget everything I just said about the inverted AO map. Just make a new mix texture set to greyscale and use the normal AO map to control the mix. then you can change the amounts of each slot as you see fit.

Here is the result from that.

And here is the mix setting that got me this result.

patina_mix.jpg (18.86 KiB) Viewed 1271 times

The bottom number controls the brighter areas on the AO map, and the top number the duller areas. 0.0 is the most shiny, and 1.0 is a super matte. Even 0.5 is almost completely matte.

[SATtva]

I'm very interested in the results (and actually the progress) of this thread. Great work Binary!

[binarycortex]

I'm very interested in the results (and actually the progress) of this thread. Great work Binary!

Thanks SATtva!

I think I'm going to call this final unless anyone has any suggestions. I'll throw some more samples at it before I post it of course.

128 S/p