(BTW., keep in mind that the theoretical maximum for Meow Hash v0.4 was 16 bytes/cycle, not 18 bytes/cycle - graphs that show it as 18 bytes/cycle were taken without CPU boosting disabled, which messes with the RDTSCs and makes things look faster than they are)
-
-
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Is the drop around 500k due to hitting some hardware cache limit?
-
..and if so, it looks like L0 runs out at around 1000 but that doesn't bother meow, and L1 runs out at around 500k and affects everyone..?
- Show replies
New conversation -
-
-
Is this data/test synthetic ? Any more details on the test ?
-
It's meow_bench (https://github.com/cmuratori/meow_hash/blob/master/more/meow_bench.cpp …). It's the best hash test that I know of - and one of the only ones that tries to eliminate branch prediction favoritism. I'm using v0.5 but it's basically the same as the v0.4 in that github.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.