Hi Greg,
I am revisiting the “Adaptive Shadow Testing for Ray Tracing” paper and am wondering how the algorithm
sacles up with multi-threading since two of the variables “the hit counter for source” and “the test counter for source” are source members and a source can be sampled for hit points all from multiple threads?
source[scp->sno].nhits++
source[scp->sno].ntests++
Thanks alot!