Updating accel structure from GPU #157
-
Hi, is it possible to efficiently update the positions of procedural primitives in the accel structure from device side kernels? In the examples I only see building the structure up front once and then reusing it. My use-case is to reuse the fast ray-aabb tests for collision detection and I have a realtime gpu physics simulation id port to luisa if that were possible :) |
Beta Was this translation helpful? Give feedback.
Answered by
Mike-Leo-Smith
Oct 4, 2024
Replies: 1 comment 3 replies
-
You can launch a kernel to update the AABB buffer and then encode into the stream a ProceduralPrimitive::build() command and an Accel::build() command. |
Beta Was this translation helpful? Give feedback.
3 replies
Answer selected by
cat-state
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
You can launch a kernel to update the AABB buffer and then encode into the stream a ProceduralPrimitive::build() command and an Accel::build() command.