croemer 5 days ago

Parallelism isn't speed. Zstandard decompression is insanely fast, no need for occupying multiple cores.

1
Retr0id 5 days ago

Bloom filter lookups are embarrassingly parallel to the point that you could occupy 0 CPU cores and do it in a GPU fragment shader.

There are probably other reasons why this is a bad idea but I'm curious to try it.