Re: [computer-go] Testing move generation and tree traversal

[Maarten Mariën <maarten.marien@xxxxxxxxx>]

> I agree so far
> 
> >39,929,136 ways of playing a 4ply game,
> 
> The obvious, and wrong, number to give here is 81*80*79*78 = 39,929,760.
> This has to be reduced by the cases where White can't play on a 1-1
> point because there are black stones on both the adjacent 1-2 points.
> How many such cases are there?  I make it 4 (the number of corners where
> it can happen) * 2 (the number of orders the two black stones could have
> been played in) * 79 (the number of points where the first white stone
> could have been played) = 632.  This leaves 39,929,128.  Your number is
> eight more than this.
> 
> Nick

Eight is the number of possibilites for White's first stone to have
been captured, so that White's second stone has an extra possibility.
(i.e., the last *78 has *79 for those eight cases)

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