If you can implement the intersecting bounding cone idea without impacting frame rates that's going to be even smoother on WebGPU but it would be interesting to

If you can implement the intersecting bounding cone idea without impacting frame rates that's going to be even smoother on WebGPU but it would be interesting to see the difference apples to apples with this type of implementation.