Re: [gnugo-devel] owl:11

[kevin yong]

Hi, Evan:

  i did some work on regress century2002:180.

  problem: O11 wrong, T9 right.

  source of the problem: During make_worms() reading,
it tried to attack B worm S12 and failed. Factually,
it is wrong because W:T9 would make seki. then i go
through all the variations about attack B worm S12,
and found the problem: after sequence:

  W:R11 B:S10 W:T9 B:Q10

  W should make no further move here and it is seki.
but W keep trying to move inside here and made total
dead stone = 4, therefore made B live.

  recommendation: enhance reading.c to handle seki
case.

  like to hear your comments.

kevin.




  

  


______________________________________________________________________ 
Post your free ad now! http://personals.yahoo.ca