[Computer-go] New Crazystone - when again?
Remi.Coulom at free.fr
Tue Apr 26 03:12:55 PDT 2011
I have similar logic in Crazy Stone, but in this particular case, it does not understand the semeai because of the ko. Should be easy to fix.
On 26 avr. 2011, at 07:53, David Fotland wrote:
> Many Faces’ old engine counts liberties in semeai that way, including approach moves, shared liberties when only one side has an eye, etc. This information is only used in the tree, not the playouts.
> From: computer-go-bounces at dvandva.org [mailto:computer-go-bounces at dvandva.org] On Behalf Of terry mcintyre
> Sent: Monday, April 25, 2011 10:13 AM
> To: computer-go at dvandva.org
> Subject: Re: [Computer-go] New Crazystone - when again?
> I also believe it resigned in a lost position. The mistake was S14. It should have played J12 or M12 instead, which should lead to a win.
> Rémi is correct.
> S14 is an interesting blunder. The preceding T2 move didn't exactly remove a direct liberty from white - it allowed the O1 approach move which had previously been a self-atari for black. This conforms to a more generalized use of the word "liberty" used in some discussions of capturing races.
> Does Bonobot - or any program - maintain a count of "moves needed to capture X, including approach moves and throw-ins, etc?"
> Computer-go mailing list
> Computer-go at dvandva.org
More information about the Computer-go