gnugo-devel
[Top][All Lists]
Advanced

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

[gnugo-devel] Bug report


From: Tom Zych
Subject: [gnugo-devel] Bug report
Date: Sun, 01 Dec 2002 08:43:04 -0800

Hi, I was playing gnugo 3.2 and we played a pretty weird game.
There seem to be bugs in both the engine and the scorer. I tried
to compile 3.3.12 but couldn't get it working (not too experienced
in this area). So this may be fixed already.

The command line was:

gnugo --boardsize 9 --color black --handicap 2 --komi 0.5 -M 64 --seed 6

Here's the game:

(;FF[4]GM[1]
SZ[9]
GN[GNU Go 3.2 Random Seed 6 level 10]
DT[2002-12-01]
KM[0.5]HA[2]AB[gc][cg]AP[gnugo:3.2]RU[Japanese]
 ;W[cc]
 ;B[gg]
 ;W[eg]
 ;B[ef]
 ;W[ch]
 ;B[dg]
 ;W[fg]
 ;B[ff]
 ;W[gh]
 ;B[hg]
 ;W[bg]
 ;B[bf]
 ;W[dh]
 ;B[bh]
 ;W[bi]
 ;B[ag]
 ;W[hh]
 ;B[ih]
 ;W[]
 ;B[]
 ;W[])

Now, for one thing, I know darned well I'm not good enough to
crush gnugo that badly with only a two-stone handicap. And it
seems odd that it would pass when it did. So that's the engine.

As for the scorer, it said all of white's stones were dead. I'm
not sure if that's right or not, but in any case, the scorer said
the score was B+10.5. If all of white's stones are dead, the score
should be B+78.5:

70 territory
1 capture
8 dead stones
-0.5 komi

I don't know how my executable was compiled. It's Debian's
gnugo 3.2-1.

Hope this is helpful. Thanks for your great program!

Cheers,
--
Tom Zych
This email address will expire at some point to thwart spammers.
Permanent address is rot13(address@hidden).




reply via email to

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