I wrote an non-RTX on-GPU raytracer a while back (naive compared to this) and it's super-interesting to read about the advances in compressing BVH structures.<p>But the changes also highlights a change in focus from just implementing this naively(RDNA3 technically not too much removed from the naive raytracer I wrote) to moving it to something carefully engineered and optimized for memory bandwidth (with savings circuits even built into silicon?).