Re: [computer-go] Exact number of goban configurations.

["Jeffrey Rainy" <jrainy@xxxxxxxxxxxx>]

Hi,

> It should be possible to reduce these numbers by some smallish factor
> with a more efficient implementation. For people with clusters on
> their hands parallelization of the loop over the previous border
> states should work well. The algorithm can also likely gain some speed
> at the cost of higher code complexity by traversing the board
> diagonally, i.e.
>
> .
> 6.
> 358
> 1247
>
> and so on instead of columnwise. This reduces the length of the border
> most of the time and thus the number of border states. Unfortunately
> the same maximum border length is reached when passing the main
> diagonal(s) so it probably doesn't have a really drastic effect but
> still it might reduce the time for 13x13 by a factor 3 or 4.

I thought about that, and I'm not sure it will speed up. In fact, I bet it
would be a considerable slowdown. The number of border states will probably
be much higher since diagonally 'adjacent' stones are not connected and thus
have more possiblities. Example :

W: white with liberty
B: black with liberty
w: white without liberty yet
b: black without liberty yet
_: empty

_WwW_____ is one possible border state with a diagonal border

_WwW_____ is not a possible border state with a horizontal border, since
whites stones are all connected and all already have liberties or not.

.........
.W.......
.Bw......
..BW.....
.........
.........
.........
.........
.........

The same probably applies to connectivity between stones.

J.














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