RE: cotton balls & elephants (was Re:[computer-go]Patternmatching -rectification & update)

["Mark Boon" <tesuji@xxxxxxxxx>]

> -----Original Message-----
> From: computer-go-bounces@xxxxxxxxxxxxxxxxx
> [mailto:computer-go-bounces@xxxxxxxxxxxxxxxxx]On Behalf Of Frank de Groot
> Sent: Saturday, November 20, 2004 2:11
> To: computer-go
> Subject: Re: cotton balls & elephants (was Re:
> [computer-go]Patternmatching -rectification & update)
>
> I had an idea that is perhaps naive but it involves iterative deepening.
>
> You calculate all possible moves 3 or 4 ply ahead and then do a very
> thorough analysis of the terminal nodes.
> Then you do the same but only on a very smnall number of terminal
> nodes (10
> or so).
>

In Go, all possible moves 4 ply is a number with 10 zeros. How are you going
to do 'very thorough analysis of the terminal nodes.', and finish within the
hour?

> You can go extremely deep that way, but I am not sure whether it would be
> practical.
>

_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/