[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gnugo-devel] owl attack moves and semeais
From: |
Paul Pogonyshev |
Subject: |
Re: [gnugo-devel] owl attack moves and semeais |
Date: |
Wed, 7 Apr 2004 23:38:36 +0300 |
User-agent: |
KMail/1.6.50 |
Arend wrote:
> IMO, we should always apply the following logic when deciding whether a
> stone is alive (assuming, for simplicity, it's the opponent to move next):
>
> - if it's tactically capturable => dead
> - otherwise, if owl alive => alive
> - otherwise, if there is a single semeai that this dragon can win => alive
> - otherwise, dead
>
> [Corollary: reading code may never claim successful attack when in fact it
> is not, but is allowed to claim no attack when in fact there is; owl may
> never claim life when the dragon is in fact dead.]
Seems logical.
> This patch tries to apply this logic for owl attack moves: If the attacked
> dragon is involved in a semeai, we have to check whether it can win the
> semeai after the owl attack move has been played. If not, we only award
> a strategical attack move reason.
I assume regression results (with node counters) are on the way, right?
Paul