gnugo-devel
[Top][All Lists]
Advanced

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

Re: [gnugo-devel] report on gnu go 3.5.2


From: Gunnar Farneback
Subject: Re: [gnugo-devel] report on gnu go 3.5.2
Date: Wed, 28 Jan 2004 00:17:01 +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:
> In both cases black can push through and cut on S15 and S15,
> respectively. Of course this is bad as white can take the ponnuki, but
> if white insists on defending the 3-3 invasion stone, he can only get a
> ko.

Yes. These are really about "live with the stone or somewhere else
inside black's moyo", but we don't have a way to express that.

> I would suggest a restricted_genmove C16 C15 test for the first
> position.

An alternative might be to relax the "correct" answer, e.g.
[(1|2|3) C15]. I'm now assuming that white can't even get a ko after
C16.

> Not sure about the 2nd

Would changing it to owl_defend R15 be sufficient to make it correct?
Otherwise it doesn't worry me too much if it's allowed to get away
with a good but technically incorrect move.

> (S15 is another move to make life, btw).

There are probably several alternative moves missing among the rest of
the test cases as well. You're welcome to review all of them.

/Gunnar




reply via email to

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