[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gnugo-devel] owl:6
From: |
Gunnar Farneback |
Subject: |
Re: [gnugo-devel] owl:6 |
Date: |
Fri, 08 Aug 2003 17:10:47 +0200 |
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) |
SP Lee wrote:
> Then I think we can omit this test case safely. There is no aji to
> find at all on the surrounding.
Why should we do that? This seems like a great test case, although
difficult, regardless whether there actually exists a defense or not.
> Still I found this is not an easy problem, white can play A5 after
> black B6.
Now that move looks like something. Also it isn't in All About Life
and Death, so I can't look it up.
> Then black seems not to be able to kill white cleanly. For
> example, black B2, white A2. Or B:A2, W:A3, B:B3, W:B4. I could miss
> some variations.
I agree. I can't find anything better than ko now.
> Anyway I think gnugo doesn't have to know all the
> variations.
It suffices if it knows a few relevant ones. :-)
Graham wrote:
> I think just extending then tactic on p98 ALD Gunnar cited, is
> sufficient.
>
> eg (B 1st):
> C1,B1,A2,?B3,C4
No, it doesn't work. White can switch B3 to A3 and get a ko in this
position.
/Gunnar