OTOY Forums

Hello everybody,

I'm new to Octane, still familiarizing with it, and I was wondering: in the documentation I've been checking so far the concept of samples per pixel seems a bit of a black box, meaning I couldn't find any actual breakdown of how Octane builds up to a value of, let's say, 500 spp.
As compared to Arnold, I'd be able to say ( 3x3 diffuse samples + 2x2 glossy/reflection samples + 3x3 refraction samples + 2x2 SSS samples ) x 8AA samples and have an estimation of how much rays I'm shooting in the scene, also to have a generic balance of how many light and shadow samples are worth using as I go testing my renders.

Not a deal breaker really, I can survive with getting the generic hang of it without breaking it down further, but can anybody point me to some reference for this?

Thanks!

Hi,
please, have a look at this video from Disney on how Path Tracing works:
http://youtu.be/frLwRLS_ZR0
ciao beppe

Hi bepeg4d and thank you for the link,

Unfortunately this is just a path tracing 101, let me rephrase my question, maybe it wasn't put down properly.
Is it possible to control or see the number of samples being used per-pixel for each different type/cetegory of ray being solved (diffuse, spec, transmission, etc..) ?
Or is it just one master number of samples that's constant across all pixels of the whole picture ?

example: gray-shaded matte geo everywhere with a glossy sphere bouncing in the middle. fixed camera and 1 directional light. the only heavy-sampling portion is the motionblurred hilight of the sphere.
how do you shoot more rays on the spec areas only, keeping the diffuse sampling low and cheaper? or do you just raise that 'master' samples-per-pixel number upsampling all pixels?

Thanks!

After some more digging around, I found the answer:
No, Octane doesn't offer control over separate types of samples.
Presumably, a decision was made to prioritize ease of use vs control, relying on MIS and adaptive sampling to optimize potential waste of rays.