Re: computer-go: Programming Go in Java!

["Phil" <PhilippGarcia@xxxxxxxx>]

For some strange reason the mailing list put Ray's name on the e-mail I
sent.

If you want a copy of Benson's paper, just send me (philippgarcia@xxxxxxxxxxxxxxxxx)
your mailing address.

Phil

----- Original Message -----
From: "Ray Tayek" <rtayek@xxxxxxxxxxxxxxxxx>
To: <computer-go@xxxxxxxxxxxxxxxxx>
Sent: Monday, May 01, 2000 3:37 PM
Subject: RE: computer-go: Programming Go in Java!


> I have a copy of Benson's paper. I'll be happy to snailmail this to anyone
> who
> wants a copy. Just e-mail me your address.
>
> Phil
>
> ----- Original Message -----
> From: "Ray Tayek" <rtayek@xxxxxxxxxxxxxxxxx>
> To: <computer-go@xxxxxxxxxxxxxxxxx>
> Sent: Monday, May 01, 2000 1:01 PM
> Subject: RE: computer-go: Programming Go in Java!
>
>
> > At 04:51 PM 5/1/2000 +0100, you wrote:
> > >...Thanks for the code, but I do not really understand what is
happening.
> > >...Any chance of a quick lesson into how it works?
> >
> > i assume you mean benson, since the other is just inner classes (i was
> > thinking that these could give you a board to move on as well as being
> > something to add benson to).
> >
> > hmmm, can't seem to find my copies of benson's papers (wish i could get
> > them online). i sent someone copies of what i had by snailmail, maybe
they
> > can send you copies.
> >
> > in a nutshell, benson finds all sets of stones that are unconditionally
> > alive (i.e. the other guy can move forever and you never have to
respond).
> > this is good for terminating searches.
> >
> > benson, calls a connected set of stones an x-block (x=black or white).
> > benson determines unconditional aliveness by making sure that each
x-block
> > has 2 small x-enclosed regions (usually eyes, but not always - some may
be
> > "fake" eyes) that are "healthy" for it.  these x-enclosed regions must
be
> > enclosed by x_blocks in the set x_blocks under consideration for
> aliveness.
> > furthermore, these x-enclosed regions are such that the other player can
> > not fill them up without capturing himself (i.e they are "small" in some
> > sense).
> >
> > in the suport class, we take the initial x-blocks (y) and find every
thing
> > that is region connected to it ([y]) - (tranansitive closure). then we
> > throw way any xblock that does not have 2 healthy regions. what's left
is
> a
> > set of x-blocks that are unconditionaly alive and this is called the
> > support of y.
> >
> > i can probably answer more detailed questions by looking at the code,
but
> > maybe we ought to do this off line as perhaps not all on the list are
> > interested in benson.
> >
> > hth
> >
> >
> >
> > ---
> > ray  http://home.pacbell.net/rtayek/
> > vice chair orange county java users group http://www.ocjug.org/
> > hate spam? http://samspade.org/ssw/
> >
> >
>