[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gnugo-devel] Opponent not passing
From: |
Trevor Morris |
Subject: |
Re: [gnugo-devel] Opponent not passing |
Date: |
Sun, 03 Feb 2002 15:15:19 -0500 |
At 10:01 AM 2/2/2002 -0800, Daniel Bump wrote:
>
>The patch trevor_1_24.3 contained something called
>opponent_not_passing with the remark:
>
>/* If we're way ahead, and opponent has not been passing when we
> * think there are not even dame left, let's start playing aftermath
> * moves, too. This funtion returns 1 if we should play an aftermath
> * or capturing move.
> * See, disaster in games/nngs/gnugo-3.1.22-heeroy-200201252006.sgf
> * See also TurboGo in the 2001 European Congress (Dublin), where GNU
> * Go took some really unnecessary risks in passing over & over.
> * http://www.britgo.org/results/computer/egc01/tg-gg.sgf
> */
I still think this is a good idea, in the following very restricted
scenarios:
1) we're way ahead (say 50 pts. or so)
2) opponents last move was not a pass.
3) we've already passed a few times recently.
(I can't currently reach the www.britgo.org site; is this game
already in the regression set? If not, it probably should be.)
In order to speed the end of the game, if the opponent is gayly
playing on the inside of my territory, hoping for a blunder, and
I'm way head, I would certainly just respond, and start consolidating
all the possible weaknesses he might be looking for.
The discussion regarding owl-threats makes a lot of sense, and I
agree with most that Dan & Arend have said. I'd add the following
observation, too. There seems to be a very close correlation
between the three stats: cputime / owl nodes / reading nodes.
So, it may make sense to turn off owl-threat reading only when
the number of reading or owl nodes reach a certain threshhold.
I am all in favor of not using the opponent's last move as a basis
for evaluation, though it is certainly tempting (for example, in
running owl-threats only against dragons near the last stone played).
I find GNU Go's endgame very refreshing, as it tenuki's quite freely
and often appropriately, a typical weakness of even much stronger
human players.
-Trevor