gnugo-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [gnugo-devel] arend_1_25.3: less (??) experimental influence


From: Daniel Bump
Subject: Re: [gnugo-devel] arend_1_25.3: less (??) experimental influence
Date: Tue, 12 Feb 2002 21:39:08 -0800

I notice that although games have resumed on NNGS, the web page
at: 

http://www.lysator.liu.se/~gunnar/gnugo/nngs/

stops on February 2. Will this extremely useful page be restarted?

Thus far the match between GNU Go and itself has progressed to
23 games. The experimental influenced version has won 16 out of
these. The scores:

W+33.5
B+43.5
W+13.5
W+15.5
W+12.5
W+9.5
W+39.5
W+33.5
W+16.5
W+29.5
W+31.5
W+23.5
W+28.5
B+23.5
W+14.5
W+54.5
B+43.5
B+6.5
B+29.5
B+4.5
B+0.5
W+1.5
W+1.5

Arend wrote:

> Hmmm, looking good. However, if you want to see the weakness of
> --experimental-influence, try playing a 9 stone handicap game with
> GNU Go as white. During fuseki, it will tenuki at any moment it possibly
> can, ignoring peeps etc.

On the other hand even with standard influence GNU doesn't play
9 stone games well.

Perhaps we should make this option standard in 3.1.25! Then if
tunings are broke we can try to fix them.

Dan




reply via email to

[Prev in Thread] Current Thread [Next in Thread]