Closed tippesi closed 3 years ago
This might be caused by low ray counts per probe and therefore a low accuracy (see 256 vs. 32 rays per active probe). Need to investigate the accumulation and make sure the ray rotation matrix is actually random and rotates rays in a uniformly distributed manner across the sphere
This can be mitigated a bit by using a dynamic hysteresis or using a higher ray count. Dynamic hysteresis is implemented now, might need some adjustments in the future
There seems to be an energy loss when using DDGI. Further investigation is required