On Sat, Aug 15, 2009 at 08:33:31AM -0400, Jason House wrote:
> On Aug 15, 2009, at 8:22 AM, Don Dailey <[email protected]> wrote:
>
> >
> >
> >2009/8/15 Jason House <[email protected]>
> >On Aug 14, 2009, at 11:02 PM, "David Fotland" <fotl...@smart-
> >games.com> wrote:
> >
> >>Moves often merge two groups.
> >>
> >>I count liberties incrementally as I make moves, so no need to
> >>search to count.
> >>
> >
> >How do you detect shared libreties to avoid double counting.
> >Simple addition does not work for real liberties (and I think
> >you've said previously that you track real liberties.
> >
> >I don't know how David does it or if there are special tricks,
> >but you get real updates without that much extra work - you just
> >have to look at a few points around the newly placed stone.
> >
>
>
> It's not that simple. Shared liberties can occur far away with
> longer changes. I've come up with a scheme that looks a few points
> around, but only works for chains up to about seven stones. Consider
> two parallel, linear chains of 5 stones each, separated by a space.
> A stone placed at one end to connect them is likely miss the more
> distant shared liberties.
>
> XXXXX
> +
> XXXXX
>
> It's also possible to construct more complicated cases (with non-
> linear chains) where there are no locally shared liberties, but some
> exist further away.
Isn't it easier to use the gnugo way, keeping just up to 4 or 5
liberties listed explicitly? This covers many common cases and makes you
do hard work _only_ in case the chain has many liberties and one of the
explicitly listed ones disappears. All else including merging or
1,2-liberty patterns is trivial.
--
Petr "Pasky" Baudis
A lot of people have my books on their bookshelves.
That's the problem, they need to read them. -- Don Knuth
_______________________________________________
computer-go mailing list
[email protected]
http://www.computer-go.org/mailman/listinfo/computer-go/
