Dielectric mistakes - my bad...

[Oliver]

OK, trying to get into the material nodes - and getting it wrong, as you see.
Double sided would have been nice, but I forgot. Well, still a nice image, I think.

Rendertime useless, rendered in a single background process while working and rendering with LW and AE... maybe tonight a correct render incl. rendertime.

Greets.
Oliver.

[Oliver]

That's the way I like it...

GItime 3,8sec (2 luminous polygons, no lights, just for something to refract)
Rendertime 4h 30min (FG RPE 128 MES 20mm Tol 0,1 Bounces 2)
AA2 (classic), Adaptive Sampling 0,01 (blue noise)
Xeon 5160@3GHz

Pretty stupid settings ... :o
Still learning.

Greets.
Oliver.

[Bliz]

I like 'em.

I especially like the rich colour of the first one.

[Castius]

I've been looking forward to this render with double sided. It's looks great.

About the render time.
You need to be careful of two things when you using AS. You generaly want to not drop this to low. And you should use limit dynamic range.

Since with each AS pass it will double it's samples. AS uses pixel contrast tolorance to decide what pixels to sample. So if you don't limit the range it may be almost impossible for it to calulate properly.

I try to never drop the AS below .05. I've only had to do that with extream DOF to resolve the boken effect.