Hi,
I encountered a geme where gnugo miscalculates the score at end of game.
The actual score is B+79.5 while gnugo thinks it is W+10.5. In the game
white has a big group surrounded with no chance of making a second eye and
gnugo does pass, so I do not think it has any ideas to the contrary when
considering where to play. Yet the scoring algorithm does seem to have
some delusions.
I uploaded a screenshot and sgf of the game at
http://www.cs.tut.fi/~tikkanet/gnugo-bug.png
http://www.cs.tut.fi/~tikkanet/gnugo-bug.sgf
br,
Tuukka
_______________________________________________
gnugo-devel mailing list
gnugo-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/gnugo-devel
