Usenet.com

www.Usenet.com

Group Index

Rec Thread Archive from Usenet.com

<-- __Chronological__ --> <-- __Thread__ -->

Re: Best Computer Chinese Chess Program



If you want to know the "best" or "strongest" engine there are a few
things to keep in mind:

1 - If an engine doesn't compete in a tournament, then it obviously
can't win. And just because an engine does compete at a tournament
doesn't mean that anybody can get a copy.
2 - An engine can potentially be optimized to win against computers,
but this could make it weaker against humans. You should look for
human-computer results too if you care about that.
3 - Tournament results aren't statistically significant. This is of
course true for all sporting events, but with computer chess we have
the opportunity to obtain statistically significant results by using
autoplayers.
4 - Often commercial engines are tweaked for tournaments. If you buy
the engine then you may not receive the exact version that played in
the tournament. For example the opening book may be different. Wins at
a tournament can sometimes be attributed to cooked books.
5 - Typically tournaments don't use uniform computing platforms. So
engines may run on different hardware. You should take a look at what
hardware was used depending upon what machine you plan to run it on.
For example if a 500 processor system wins, then you can't expect the
same level of performance on a single P4 at home.
6 - For Xiangqi you need to be clear about what rules were used at the
tournament. An engine that implements all of the AXF rules may search
slower than one that takes shortcuts. If you absolutely need all of
the rules to be followed, then this could be important. I imagine that
if masters know that a program implements a partial ruleset, then they
could take advantage of that. People learning the rules for tournament
play may want the computer to be able to explain rule violations too.
7 - Features like endgame tablebases may not increase playing
strength, but may be important if you're interested in studying
endgames or perhaps taking a lot of time to analyze games. Also you
might want a endgame visualization tool like that provided by Wilhelm
for international chess (http://www.geocities.com/rba_schach2000/)
8 - you may want other features such as adjustable strength play,
tutorials, integration with a database,... None of these features will
be reflected in tournament results.

"Mark Winands" <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>...
> We had already two computer chinese chess championships at the computer
> olympiad in Maastricht.
> 
> You can view the results at:
> 
> www.cs.unimaas.nl/olympiad
> 
> www.cs.unimaas.nl/olympiad2002
> 
> ELP won both tournaments.
> 
> There will be again championship in Graz this year.
> 
> Mark
> 
> "Metastabl" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]
> > Chris Kubica <[EMAIL PROTECTED]> wrote in message
>  news:<[EMAIL PROTECTED]>...
> > > What's the best computer chinese chess pgm? Where do I get it?
> > >
> > > Please reply to [EMAIL PROTECTED]
> >
> > You can take a look here for an opinion:
> > http://ai-master.com/cch000.htm
> >
> > If you can't read Chinese, then translate with babelfish
> > (http://world.altavista.com/) Or you can just click on the links for
> > each program and see if you find anything that you didn't already know
> > about.
> >
> > Regards,
> > Keith



<-- __Chronological__ --> <-- __Thread__ -->


Usenet.com



Please check out one of the premium Usenet Newsgroup Service Providers below for access to Usenet.