[Pachi] Low-hanging fruit - hacking ideas

Matthew Woodcraft matthew at woodcraft.me.uk
Sun Jun 5 19:51:26 CEST 2011

Petr Baudis wrote:
> I'd expect that pattern3_move_here is so slow because the hash lookup is
> almost always a cache miss.

If I'm reading 'perf annotate' output properly, the cost on my machine
is mostly from the unpredicable branch (and it's hard to do anything
about that).

That's from a single-threaded run. Out of interest, in tournaments are
you running with one thread per core or two?


More information about the Pachi mailing list