Author: Christophe Theron
Date: 12:17:23 11/02/98
Go up one level in this thread
On November 02, 1998 at 12:55:57, Reynolds Takata wrote:
>>>Weakens? Chess master is a monster and the strongest combinatively. I'm a
>>>master(USCF 2249) and i own all of the top programs except tiger and mchess. CM
>>>has all other programs beat when it comes to the majority of combinations. CM
>>>is going to win the Komputer Kup tourney, on the Shep's Computer chess site it's
>>>won more tournaments run there than any other program. The only reason i think
>>>it didn't do as well in the Mclane's tourney i believe is do to some sort of
>>>hardware conflict there. This because he kept reporting CM crashing, when i
>>>have had CM a long time and never had a crash. CM is in the same category as
>>>far as strength is concerned as any of the top programs and stronger than the
>>>majority. The thing that is so good about chessmaster is that it's actually
>>>trying to win, ALL the other programs will gladly concede to a draw if you want
>>>one, but not chessmaster, it's motto is "OFF WITH YOUR HEAD!". Many people want
>>>to charge the programs with making positional mistakes, the master understands
>>>that positional weaknesses are a necessary part of the game if you really want
>>>to win. Chess is about give and take. The trick is to be the first to make use
>>>of the weaknesses and chessmaster seeks the initiative so hard it usually gets
>>>there first.
>>
>>I'm not sure I can argue with you, as you give only subjective reasons.
>
>Performances in tournaments are not subjective reasons.
As stated 1000 or more times, how can you compare a 10 rounds tournament result
with the hundreds of games played by EACH program in the SSDF?
>Further me being a
>master and having played almost all the top programs, is not a subjective reason
>either.
Aren't you giving your feeling? You played several programs and it looks to you
that you can beat them more easily than CM. This is subjective.
>What you are overlooking, and i have posted here before is that chessmaster is a
>slow program, and it performs better on faster hardware, on a P90 it is not
>equal to other programs, CM needs at least a P233 to perform at it's best.
>Sometimes something that is also a slight hindrance(about 20 elo) is not
>chessmasters raw strength, but a non-optimized book. CM with fantastic play
>usually overcomes the probs of the book though.
The best thing to do when your program is not the best, is to argue that it is
better with faster hardware.
First it is hard to prove that it's false, because too few people own the fast
computers you are talking about.
Next, if program A scores say 30% against program B on 486dx2-66, it is a normal
trend that it will score 35% on P100 (I mean same hardware for both program),
and maybe 40% on P200MMX.
Push the speed at will, and you come to the point that program A scores 47.5%
against program B, so you are under the margin of error you can expect from even
a long match, and now you can happilly claim that program A and B are equal.
>>You may be impressed by the combinations it finds, but even in combinations
>>there are better programs.
>
>Not unless it's J5, Tiger or R10.
Ok, so let's make an experiment.
Please post here a set of at least 10 positions where CM has found a great
combination. Please post the positions in EPD format, so that everybody is able
to load them and try them.
I promiss I'll give Tiger results on these positions. I hope other testers will
post Junior5, Fritz5, Nimzo98 and other programs results as well.
If your point is correct, we will find out. If not, at least we will have some
interesting positions to add to our collections.
Do you accept the challenge?
Christophe
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.