[pbmserv-dev] Re: [pbmserv] question about the game "Alak"

[Lyman Hurd <lhurd@yahoo.com>]

Probably need to move the discussion to the Dev
mailing list, but wanted to point out that all we
really need to track is a hash code of previous
positions.  I started to write some general code of
this sort when I was trying to help deal with the
question in support of Diam.  Never implemented it but
could dig it up...

Lyman

--- Martin Møller Pedersen <traxplayer@gmail.com>
wrote:

> On 10/7/05, Scott Huddleston
> <scotth@ichips.intel.com> wrote:
> > As of two or three years ago, the pbm server
> recognizes 99.99% of
> > the Chinese ko rule for go.  I added an algorithm
> that detects
> > any repeated board position that follows a
> sequence of single
> > stone captures.  This includes triple ko and any
> ko likely to
> > arise in real play.
> >
> > But this probably doesn't help Alak.  Alak
> diverges from go in
> > too many ways (no snapback allowed, go suicide
> survives in alak).
> >
> 
> I think that you have to save all earlier board and
> compare them to
> the current board
> before you can detect a repeat position.
> 
> /Martin
> 
> 
> To unsubscribe, send a message to esquire@gamerz.net
> with
> 	unsubscribe pbmserv-users@gamerz.net
> as the BODY of the message.  The SUBJECT is ignored.
> 
>