Re: [gnugo-devel] level of play

[Evan Berggren Daniel]

On Fri, 9 Jan 2004, david doshay wrote:

> However, in the code it says:
> --level <amount>  strength (default %d, up to 10 supported)\n\
>
> #define DEFAULT_LEVEL 10
>
> So, we thought we were running at the highest level possible
> when playing the above games.

You can put any number for the level, within reason.  It just means that
we don't really make any guarantees about strength improving beyond level
10.  If you set higher levels GNU Go will read deeper in a variety of
things.  It is fairly clear that GNU Go is stronger at level 12, and
stronger still at level 15.  However, it does get rather slow.

> Just to let you know about some other games we have run on our
> cluster, when we throw in lots of resources, we managed to win 4
> out of 8 doing a 16 move lookahead against a 4 stone handicap.

Very impressive.  Congratulations.  Any chance we could see the game
records?  Just the games themselves would be interesting, though seeing
what moves were looked at would too.

Do you know if these results generalize at all?  ie, is your machine that
much stronger when played against human or other computer opponents?  My
own metamachine experiments suggest that the effect is far more pronounced
against gnugo than against other opponents, though my techniques were
substantially different (and I only do 9x9).

Evan Daniel