[Top][All Lists]

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

Re: [gnugo-devel] defend both, attack either, worms[], dragons[]

From: Gunnar Farneback
Subject: Re: [gnugo-devel] defend both, attack either, worms[], dragons[]
Date: Tue, 10 Dec 2002 23:15:40 +0100
User-agent: EMH/1.14.1 SEMI/1.14.3 (Ushinoya) FLIM/1.14.2 (Yagi-Nishiguchi) APEL/10.3 Emacs/20.7 (sparc-sun-solaris2.7) (with unibyte mode)

Arend wrote:
> My experiments with revising the influence interface revealed some problems
> with "attack either" and "defend both" move reasons. I think the only way
> we will ever get their valuations right is by strictly using the
> following definition:
> 1. Both "attack either" and "defend both" must concern two tactically stable
>   worms.
> 2. "attack either" means opponent cannot defend both worms at a time but could
>   if we don't play.
> 3. "defend both" means opponent could play a move after which we cannot
>   defend both (but we could defend each of them, of course -- otherwise the
>   worms wouldn't be stable).

Sounds right.

> Hence I finally decided to kill both the worms[] and the dragons[] array,
> together with find_worm() and find_dragon(). They seem to be a relict
> from 2D-board times to me.

Yes, they are.

> (Well I see that this also gives some consistency with regards to the
> way connections, EITHER_MOVE data etc. is stored, but unless someone
> can convince of a good reason why this consistency is necessary I would
> really love to get rid of them.)

No, there's no substantial inherent value in that consistency.


reply via email to

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