Re: [computer-go] Designing faster, better influence functions

[Erik van der Werf <E.vanderWerf@xxxxxxxxxxxxx>]

Mark Boon wrote:
> And then there's the matter of inefficiency of course of calculating
> influence which is then discarded.
Maybe this can be compensated by performing the calculations in the 
Fourier domain, or is 19x19 still too small for that?

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