> > One more thing is that after re-compiling with this new pattern, > > gnugo passes after white responds (typically C2 or C1); is there > > something else I need to do to let gnugo know how to finish the > > invasion? > > Once there is a stone in place the responsibility falls to the semeai > analysis code in semeai.c and owl.c. Unfortunately that has been known > to make mistakes now and then. Occasionally it's necessary to help it > with explicit patterns like S10b and similar but if possible we prefer > to avoid that.
I believe the semeai and owl code breaks down pretty bad with this pattern. It's obvious the best move for black is B2, but when white responds with C1, gnugo gives up. The response W-C2 B-C1 W-B3 B-A3 is found with default options. But when you turn the level up, W-C1 is found. Adding S15b S15c would help there, but W-C1 B-A3 W-B1 B-C2 and white has options at W-B3 B-A2 and W-A2 which black doesn't know how to deal with. (Of course the correct response is B-A1 W-B3 B-D1 W-C1 B-A2.) My point is that adding patterns for this would not be pretty. _______________________________________________ gnugo-devel mailing list gnugo-devel@gnu.org http://lists.gnu.org/mailman/listinfo/gnugo-devel
