Quoting Petr Baudis <[email protected]>:
On Thu, Sep 10, 2009 at 01:54:49PM +0200, Magnus Persson wrote:
This is very interesting. Here is a crazy idea, maybe it the same as
Marks but I want to take it to its extreme.
Since AMAF values are so helpful, perhaps one can let go of the idea
of sequential play following the rules of go, and basically play
moves in parallel on all empty intersection. Compute new state (do
captures) and repeat a fixed number of times and evaluate.
Since I do not understand GPU programming at all I cannot judge
whether this would make things easier to implement.
Also a lot of weird thing will happens when you do things in
parallel. What happens when two adjacent blocks are captured at the
same time. Are both removed? Are there tricks to remove the one that
was most likely to captured to begin with.
Well, after one iteration, _all_ blocks should be captured since there
will be no empty intersections, right? (Even if you don't play in
eye-like spaces, you will have this situation until some of these
actually form.)
The easiest heuristic would seem to be to capture the smallest group,
but I have hard time imagining this working well, _and_ how to choose
the smallest group efficiently (other than that, modifying my source to
do this should be fairly easy).
--
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/
--
Magnus Persson
Berlin, Germany
_______________________________________________
computer-go mailing list
[email protected]
http://www.computer-go.org/mailman/listinfo/computer-go/
