[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gnugo-devel] owl.c question
From: |
Gunnar Farneback |
Subject: |
Re: [gnugo-devel] owl.c question |
Date: |
Sun, 17 Feb 2002 23:14:07 +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:
> So the specification would be that such a helper is allowed to decrease the
> pattern value? (This is what I guess from the code.)
The real truth is that that piece of code most likely was copied from
the corresponding code in shapes.c some time during 2.7.xxx. I don't
think the ability to modify the pattern value (other than down to
zero) is very interesting. Sometimes I have wanted to be able to use a
helper to generate a different move (typically of the back-filling
variety) than the one marked in the pattern, though.
> Another question: There are a few patterns with class b or B in
> owl_attack_pats.db.
Those are bogus.
> I cannot find documentation on this.
It's documented at the head of owl_defendpats.db:
# B : This move is not connected to the goal dragon and should
# therefore not be included in the goal.
#
# b : This move is maybe not connected to the goal dragon and should
# therefore be included only after closer examination.
> They set the same_dragon variable. Is this used at all for owl
> attacking moves?
No, it doesn't apply to attacking moves.
/Gunnar