Re: [computer-go] Computer Olympiad photo

[Marco Scheurer <marco@xxxxxxxx>]

On Thursday, December 4, 2003, at 02:50 PM, Vincent Diepeveen wrote:

> My poor nameless go program searched 6 - 7 ply in 19x19 at 100-200Mhz
> hardware a few years ago.
>
> Yet that's pretty useless when a pattern takes 150 plies to show its 
> use...
>
> The computer chess guys making 9x9 go programs will have the problem 
> that
> the first few moves of the game can be decisive and they lack go 
> knowledge
> there...
But like in chess it could benefit from an opening book for these first 
moves. Of course I don't know if that plus search would be enough to 
reach the stated goal.

And by the way, 1-dan on 9x9 does not mean much since official ranks 
are evaluated using 19x19 games. What is the criteria? Beating how many 
players of what rank how many times? With what handicap? Should the 
program, when benefiting from a large komi or some handicap stones, be 
able to beat a higher dan? Should it be able to beat kyu players when 
benefiting from handicap stones? These are things expected from a 1-dan 
human...

marco


Marco Scheurer
Sen:te, Lausanne, Switzerland    http://www.sente.ch

_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://computer-go.org/mailman/listinfo/computer-go