[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gnugo-devel] reading patch
|
From: |
Gunnar Farneback |
|
Subject: |
Re: [gnugo-devel] reading patch |
|
Date: |
Sat, 27 Oct 2001 08:53:11 +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) |
Trevor wrote:
> Um, on looking at this again, I do believe that it is correctly a
> ko, and not unconditional.
>
> To wit, in reading:154 -
> B G4, W F2, B H2, W H3, B H4, W J2, B H1, W J4
>
> A bit complicated, but none-the-less, it is a ko.
Agreed. To correctly detect this the ko depth needs to be increased to
11 (default 8).
Anyway, my patch does solve test case trevor:3 since it realizes the
danger of filling in at E3 on move 28. To detect this blunder the ko
distinction isn't important.
Btw, what my patch really does is to stop the tactical reading from
spending nodes and reading depth on the pointless atari at D5 and
similar meaningless moves.
/Gunnar