[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gnugo-devel] Tuning (trevor_1_24.4)
From: |
Trevor Morris |
Subject: |
Re: [gnugo-devel] Tuning (trevor_1_24.4) |
Date: |
Fri, 01 Feb 2002 12:29:46 -0500 |
At 04:56 PM 2/1/2002 +0100, you wrote:
>
>Hi Trevor,
>
>nice tuning results!
>
>From your patch:
>+Pattern CB14
>+# tm New Pattern (3.1.23) (see arend:32)
>+# FIXME: only one of these points should be marginal
>+# is it safe to use amalgamate_most_valuable helper here? How?
>+
>+XO*O double cut.
>+O!O!
>+
>+:8,BX
>+
>+XA*O double cut.
>+ObOc
>+
>+; xplay_attack(*,A) && !xplay_attack(*,b,c,c)
>+; && (gprintf("CB14 matched: %1m %1m",b,c),1)
>+
>+
>
>The amalgamate_most_valuable helper has some problems with regards to
>owl-reading, see owl status of dragon N9 in arend:3. But I'll assume once
>we have a reliable connection reader, and get it well integrated, we'll
>have better solutions for this problem.
>
>I assume the constraint above should probably be
>+; xplay_attack(*,A) && !xplay_attack_either(*,b,c,c,*)
Yes, this is better. Much better without my bogus gprintf, too.
<urg>
-Trevor