On Mon, 2008-12-22 at 14:05 +0100, Heikki Levanto wrote: > On Mon, Dec 22, 2008 at 07:20:24AM -0500, steve uurtamo wrote: > > don, this rule is very intuitive for a 19x19 board even if you include > > the 5th line. > > Yes, and much more effective, as there are more points it will disallow. > Especially on low number of simulations - if there are only 200 simulations, > and 361 emtpy points, some points will never get a chance! not that a few > simulations from any opening position will tell much about the relative > values of the moves... > > Still, I'd like to see some results on 19x19. I also share the gut feeling > that the improvement would be greater on larger boards. > > -H
I actually went straight to 19x19 to test it, using the reasoning that it would be a better test. However, expect it to take a LONG time. - Don _______________________________________________ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
