[Top][All Lists]

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

RE: [gnugo-devel] nando_3_9.4

From: Portela Fernand
Subject: RE: [gnugo-devel] nando_3_9.4
Date: Mon, 16 Sep 2002 11:59:20 +0200

Arend wrote:

> I will try to have a look at that;

Thanx, much appreciated.

> for now, do you know why
> > Breakage: 3 PASSes, no FAILs
> >
> > ---- File [blunder.tst]
> > 12 PASSED
> > 20 PASSED
> > ---- File [trevor.tst]
> > 290 PASSED
> you lost the other 3 or 4 PASSes you had reported earlier? (I.e. where
> did you add changes apart from caching?)

Just after I coded the changes with the caching, I had only 1 PASS left,
blunder:12. After some tuning in owl_does_defend() I got blunder:20 back and
after some tuning with blunder_size() and owl_confirm_safety(), trevor:290
was ok.

Maybe it has to do with the effort I did to prevent the new OWL code from
reporting GAIN/LOSS codes when (for unknown reasons) it wasn't able to tell
which worm was involved. So before I implemented my changes about caching,
it was possible that a dragon had an attack code GAIN (or defense code
LOSS), but no worm information in the field I added for this purpose. In
such cases, the previous implementation just avoided to generate move
reasons, but it must have had an influence on the dragon status. This could
explain the difference.

I will try to verify this, but I don't know when I'll do that, I have a busy
schedule this week.


reply via email to

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