Closed rzuckerm closed 1 year ago
Clearing the bits in a loop improved the speed from 14 to 23 passes (about 1.6 times faster).
drag-race
Ha, that approach looks familiar. I recently applied a version of it in my second CUDA solution. :)
I'll approve and merge if CI passes.
Description
Clearing the bits in a loop improved the speed from 14 to 23 passes (about 1.6 times faster).
Contributing requirements
drag-race
as the target branch.