Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: GIGI

Author: pavel

Date: 07:00:05 07/12/02

Go up one level in this thread


On July 12, 2002 at 09:24:38, stuart taylor wrote:

>On July 12, 2002 at 07:59:46, pavel wrote:
>
>>On July 12, 2002 at 07:05:38, stuart taylor wrote:
>>
>>>On July 12, 2002 at 01:03:28, ERNIE COLLADO wrote:
>>>
>>>>The game I played with Spassky was at a simul held in NY around 30 yaers ago.
>>>>1974 or around there. Needless to say I was decapitated rather quickly. I will
>>>>try and look through my files and see if I can find the game. <Regards, Ernie>
>>>
>>>I'm sure that anyone who can win 1.5 from 20 games with Fritz 7 (at any of the
>>>serious modes with equal timings), can easily beat any GM at a simultaneous
>>>exhibition.
>>>Tell me I'm wrong! Anyone?
>>>S.Taylor
>>
>>Yes you are wrong.
>>I once almost beat Fritz7 with the aid of my anti-computer games database, but
>>at the endgame, I made a blunder and lost the game. (fritz7 had around -3.00)
>>You might say that I had aid of a special purpose database (which is true), if I
>>can do this (being 1500-ish) than think about the people who has excellent
>>understanding about anti-computer chess.
>>they don't necessarily need to be impressively high rated player.
>>
>>A good example would be, if you can visit ICC (even with guest account) try
>>checking out the games that computers loose (commercial programs) at blitz games
>>to un-titled human players.....sweet.
>>
>>I don't see any relation between beating Fritz7 and beating a GM easily in a
>>simultaneous exhibition.
>>
>>bad example...
>>
>>cheers,
>>pavs
>Nearly winning is also a very big difference in ELO than actually winning, but
>if the "nearly" is genuine, it only shows promise.
>Also, I mean at equal timing, and without taking back moves, let alone being
>aided.
>I'd still like to hear if others disagree whith me.
>

The game was in equal timing (40'40), no take back (If I could takeback then I
would not make the blunder, would I? :))
I had access to some 1000 anti-comp games, Fritz7 had access to millions of game
in it's opening book ;)
it's a pity that i didn't bother to save the game :(



>Anti-computer understanding can help alot, but it's not enough.
> Anyway, isn't it a shame to always try to outsmart a computer through
>anti-computer "tactics"? Is that all computer programs were made for? Still,
>their holes are getting smaller and closing up, aren't they?
>S.Taylor


there is nothing wrong with exploiting weakness in a program.
During GM games, players ussually study opening and tries to look for patterns
of his opponents games to look for weakness, as a preperation before the game.

Nothing wrong with that.
A win is a win, whether it's anti-comp or not is not the question.

the win of Smirin against CT14 was an excellent example.
Smirin didn't win by creating blocked position (as many would say an anti-comp
position), but won purely due to superior positional understanding.
Now if someone exploits that same kind of position against that version of
chesstiger repeatedly, would it be anti-comp ?

hey chess programs plays anti-human by playing open position, at which humans
are ussually weak at.

both are fair.


cheers,
pavs



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.