On Mon, 2007-03-26 at 12:33 -0800, [EMAIL PROTECTED] wrote:
> > Your idea is useful if it can be show to be superior in some
> > way to other move generation techniques.  It may be superior
> > in speed or some other metric.
> >
> 
> Main superiority I can think of: It becomes easy to combine conditions
> like (example) "All points with one black piece above and one white piece
> to the left", arrive at a new bitmap from that, and logically combine that
> with more complex conditions.

Yes.  I wrote a bitboard chess program a few years ago and once you
start
thinking in bitboards it changes the way you think about things.  It
has definite advantages as well as some disadvantages.   

I  have heard it said in computer chess that you can write a world-class
chess program using any reasonable data structure.   I think this is
true of go too.  But the data structure may have an effect on the
strengths and weakness of your programs.    You just tend to work around
the hard things and look for opportunites to take advantage of the easy
things.   

There is also a saying that if all you have is a hammer, everything
looks
like a nail.   I think there is a bit of this in your choice of data 
representations.

- Don



> Randomly combining these might lead to faster ways for programs to learn
> to recognize patterns we pick up by natural visual examination.
> 
> (I incidentally made an error in describing this; bits 1-19, 21-39, and
> 41-59 were a more natural choice than 0-18 etc.)
> 
> Forrest Curo
> 
> 
> -----------------------------------------
> This email was sent using AIS WebMail.
> http://www.americanis.net/
> 
> 
> _______________________________________________
> computer-go mailing list
> [email protected]
> http://www.computer-go.org/mailman/listinfo/computer-go/

_______________________________________________
computer-go mailing list
[email protected]
http://www.computer-go.org/mailman/listinfo/computer-go/

Reply via email to