this is an implementation of the Sieve of Eratosthenes
it's run as many times as possible in ~5 seconds to test per-thread performance
to try this program out, download the repo and run the executable inside its folder (only run executables from people you trust)
the higher the number of iterations, the faster your per-thread performance
iterations are reflected by score in points (points are adjusted for the time it took to run the benchmark)
some results i've seen to give context
processor | # threads | score (points) |
---|---|---|
i7-1065G7 | 8 | 820 |
ryzen 9 3900x | 24 | 767 |
i5-9600K | 6 | 693 |
i5-6600 | 4 | 423 |