Re: [computer-go] Ko and transposition tables

[Peter Drake <drake@xxxxxxxxxx>]

Okay, I'm going to give this a shot.  (I'll also store the depth.)  I  
wasn't keeping track of the simple ko point previously, but it's  
probably worth doing anyway, so that I can avoid making the move and  
then undoing it.

I'll also have a look at the Kishimoto dissertation.  Thanks for the  
help, everyone!

Peter Drake
Assistant Professor of Computer Science
Lewis & Clark College
http://www.lclark.edu/~drake/


On May 11, 2005, at 10:49 AM, Arend Bayer wrote:

> On Wed, 11 May 2005, Peter Drake wrote:
>
>
> > > > > One option is to XOR all previous-board Zobrist hashes and  
> > > > > store that
> > > > > value in the table as well.  This would take care of #1 and #2  
> > > > > below.
> > > > I may have some misconceptions about Zobrist hashes, but this  
> > > > doesn't
> > > > seem like it would work.  Suppose I play the move sequence A, B, C,
> > > > D.  This produces the following sequence of Zobrist hashes:
> Of course you are right this doesn't work. I think you should not  
> try to
> implement super-ko, just Japanese ko rules. Then adding a special hash
> for the "forbidden move" position is enough.
>
> Arend
>
> _______________________________________________
> computer-go mailing list
> computer-go@xxxxxxxxxxxxxxxxx
> http://www.computer-go.org/mailman/listinfo/computer-go/
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/