Re: [gnugo-devel] tuning patch

[Evan Berggren Daniel]

On Sun, 13 Apr 2003, Paul Pogonyshev wrote:
> i'm somewhat afraid to mark the move in AS7 as an antisuji. instead i
> added a large negative shape value (-10). however, M19 in the appended
> game is still valued 1.28 points. probably the best solution would be
> to revise antisuji moves as non-antisuji if they also have some vital
> move reason (e.g. OWL_ATTACK). the same may also hold for large negative
> shape values.

Some antisuji moves (eg in joseki) are attacks that work, but have bad
consequences (eg sacrifice a corner for overwhelming thickness).  We need
to be cautious about that sort of thing.  I think the appropriate response
is to have antisuji moves never be played, but be very cautious in
applying that classification.  I think revising large negative shape
values for vital moves is reasonable.

Evan Daniel