On Thu, 2006-12-07 at 08:15 -0800, Peter Drake wrote:
> On Dec 6, 2006, at 9:36 PM, Don Dailey wrote:
> 
> > The equivalent C version (after I took out some optimizations) is
> > doing 13,745.70 games per second on an old pentium 4.
> 
> I'd really love to know what I'm doing wrong. I was never able to get  
> more than about 6,000 games per second on a 2 GHz machine, even in C++.

Yes, I'm wondering about that too.   I think even my implementation is
slower than some others are getting - but there can be many differences.

The machine I'm now working on is a 2.4 GHz pentium 4, a little better
than what I had before but I'm upgraded next month to a really fast dual
core AMD.

Are you trying to keep a lot of information updated?   Mine only tries
to play random games as fast as possible.   It does not have the ability
to undo moves - this is easily handled by copying state when you need
this feature.    

There is also about a 2 to 1 slowdown if you are not finding the random
legal moves properly.   I don't know if you are or not, but look at the
recent emails on this, one of mine and one by Lex Luthor - or was it
Lukasz Lew?

I have a feeling one or both these things are hurting you.  

I think even the Java implementation should do better.

For reference, the machine I'm now working on is an old, but not too old
2.4 GHz pentium 4, a little better than what I had before but I'm
upgraded next month to a really fast dual core AMD!   I can't wait.    

I think YOUR machine actually should be performing just a little bit
better than mine in general - but I'm not 100% sure as I'm not a
processor guru.


- Don





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