The upcoming GNU Go 3.6 release will be able to resign games. (This is already included in the latest development version 3.5.10.)
It would of course be nice if as many GTP clients as possible would support this feature by the time it gets released.
Arend
Arend wrote:
The upcoming GNU Go 3.6 release will be able to resign games. (This is already included in the latest development version 3.5.10.)
It would of course be nice if as many GTP clients as possible would support this feature by the time it gets released.
I plan to implement this in Quarry 0.1.8 (current version is 0.1.6, 0.1.7 will be released in a few days).
Paul
On Aug 26, 2004, at 20:19, Paul Pogonyshev wrote:
Arend wrote:
The upcoming GNU Go 3.6 release will be able to resign games. (This is already included in the latest development version 3.5.10.)
It would of course be nice if as many GTP clients as possible would support this feature by the time it gets released.
I plan to implement this in Quarry 0.1.8 (current version is 0.1.6, 0.1.7 will be released in a few days).
Same here, I hope to release a version of Goban that supports resignation in week or so.
marco
Marco Scheurer Sen:te, Lausanne, Switzerland http://www.sente.ch
I wrote:
The upcoming GNU Go 3.6 release will be able to resign games. (This is already included in the latest development version 3.5.10.)
It would of course be nice if as many GTP clients as possible would support this feature by the time it gets released.
Marco Scheurer wrote (on gnugo-devel):
Following Arend's request about support for resignation in clients on the GTP mailing list I just remembered that Goban has had in fact built-in support for resignation for quite some time... Is there a quick test to make GNU Go resign? I've been trying 9x9 with 9 handicap stones, and it passes after a few moves.
Resignation is only enabled on 19x19 (on the basis that small board games are fast enough not to get boring anyway). Try resuming the attached game, play a white move at E9, and GNU Go 3.5.10 will resign as black.
Arend
FF[4] supports board sizes up to 52x52, but differentiating between lower and upper case vertexes. Would it be beneficial to change GTP's current case-insensitive format to take advantage of this?
FF[4] supports board sizes up to 52x52, by differentiating between lower and upper case vertices. Would it be beneficial to change GTP's current case-insensitive format to take advantage of this?
Chris wrote:
FF[4] supports board sizes up to 52x52, by differentiating between lower and upper case vertices. Would it be beneficial to change GTP's current case-insensitive format to take advantage of this?
Not in my opinion. Extremely few games are played on larger boards than 19x19 and most of those are probably on 21x21. Adding support for sizes 27x27 to 52x52 but not for arbitrary sizes would be fairly pointless and is not worth introducing case sensitivity.
/Gunnar
-
Arend Bayer -
Chris Spencer -
Gunnar Farneb�ck -
Marco Scheurer -
Paul Pogonyshev -
Peter Strempel