Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Rebel10/Fritz5 GMs

Author: Shaun Graham

Date: 05:08:14 07/26/98

Go up one level in this thread


 . The top computer chess programs are powerful. How good when
>playing many games against top humans is another matter. Perhaps the GM must pay
>more attention to what he's doing. After all, one of the computer's strength is
>its memory. It doesn't forget and it can't be bluffed and it doesn't get tired.
>Perhaps the GM sometimes plays at a handicap.
>

Must pay more attention to what he is doing?  That is an excuse, paying
attention is a component of chess skill.  If you say that a GM is performing at
a handicap because he/she can't remember as well, and doesn't get tired.   Well
then any time the average GM is playing kasparov he is playing at a handicap.
Further what you are saying is that he is weaker, because endurance and memory
are components of chess skill.  If anyone is at a disadvantage it's the
computer.  Gm's can prepare for a computer all day long, the only preparing a
computer can do against a human is with the opening book.  There's a problem
with even this, because the computer isn't deciding in such cases which openings
to use against GM's but it is in fact people,(usually non GMs) deciding.


>On July 25, 1998 at 15:38:04, odell hall wrote:
>
>>
>>A Very Good and interesting point! I have also followed some of the statements
>>of the "computer chess experts" regarding the question of Grandmaster strength
>>of Modern chess programs. My understanding of their comments is that We do not
>>have enough Games at 40/2hrs against humans to make a claim of grandmaster
>>strength for programs. However My question is How many Do We Need? It seems to
>>me that rebel 9 more then proved it was a grandmaster in the Aegon 97 tournament
>>with it's 2619 performance ratings where it scored 1 out of 2 against
>>grandmasters and defeated every international master it played, some of these
>>were in fact strong IM"s. Also let us not forget Rebel 9 outstanding performance
>>at santo domingo tournament where it scored 13 points!! to win the tournament
>>over Seven international master at 40/2hrs. It appears that such performances go
>>beyond the abilitites of an alleged "weak international master". True We do not
>>have a whole lot of games at 40/2hrs between computers, and humans but the ones
>>that we do have demonstrate clearly that computers programs are performing at
>>the grandmaster level, so why not give credit where credit is due? Is it perhaps
>>because of deep seated prejudices that computer are a threat to chess? If I am
>>wrong here would someone show me where all the games are where top programs are
>>not performing at the GM level, because all the games that I have seen Indicate
>>that they are.



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.