RadeonRays_SDK icon indicating copy to clipboard operation
RadeonRays_SDK copied to clipboard

BVH Implementation Question

Open romanlarionov opened this issue 7 years ago • 0 comments

I was wondering what influenced you to choose the BVH "intersectors" that you did. Are there any examples of BVH construction/traversal papers that you purposefully chose to omit/didn't feel were worth the time?

For instance, have you considered wide-BVHs (4-8 child nodes) for improved SIMT and with persistent CL kernels?

I'm curious to hear back.

romanlarionov avatar Jan 04 '18 21:01 romanlarionov