On Tue, Sep 15, 2009 at 08:49:15AM -0700, David Fotland wrote:
> Simple playouts with no eye fills and mogo 3x3 patterns and basic uct beat
> Gnugo 40% (at version 120)
..snip..
> All win rates are on 9x9 vs gnugo 3.7.20 level 10 with 5000 playouts. After
> this I switched to testing 19x19, and stopped tuning for 9x9.
Wow, are you sure there's no mistake in this? It is enough to have only
_5000 playouts_ in basic uct (do you use any priors?), playouts using
mogo 3x3 patterns (hane, cut1, cut2 and the four side patterns?) to beat
gnugo _level 10_?
I'm getting 10% at most with just 5k playouts, and I'm finding this
performance rather hard to believe... but maybe I'm suffering some
very bad bug that distorted my perception since the very beginning.
--
Petr "Pasky" Baudis
A lot of people have my books on their bookshelves.
That's the problem, they need to read them. -- Don Knuth
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
