Ekin Ozturk

I've identified the issue with the failing tutorial notebook, it's specifically related to the `prbvolpath` integrator and my incorrect implementation of accumulating radiance. I'm not quite sure what the correct...

No worries! I've made a list of the necessary features for NEE sampling volumes, hopefully it helps with deciding on the interface and I'm very happy to discuss it in...

I have significantly simplified the sampling of emission and reduced the complexity of the integrators in my latest code that I will push soon. Generally, I think it does make...

Does it make sense to add the 3rd volume sample by changing the `Point2f` sample to a `Point3f` sample? It seems the least cluttered way and is the most logical...

That makes sense! I will work on those and push an update asap.

Hey @dvicini, I am in the process of implementing the interface for Next Event Estimation of volume emitters, but I'm struggling with estimating the pdf of a direction sampled in...

Right, I think I understand better now. I was thinking I'd need to convert the volume PDF to a solid angle PDF in order to do MIS with the volume...

It seems that simply converting the solid angle PDF to a volume PDF does not work, or perhaps I'm doing something wrong. I've verified that integrating the volume PDF over...

The current implementation correctly tracks the ray entry and exit points in a given shape and calculates the exact directional pdf. In `llvm_*` modes this is quite slow compared to...

Looking further into, it seems that `cuda_*` modes gave the correct results when the ray intersection routines were disabled giving a pdf of 0 - this was due to how...