Author: Robert Hyatt
Date: 11:13:28 06/13/01
Go up one level in this thread
On June 13, 2001 at 11:49:04, Mark Young wrote: >On June 13, 2001 at 06:30:04, José Carlos wrote: > >>On June 13, 2001 at 03:42:41, Bruce Moreland wrote: >> >>>On June 12, 2001 at 23:56:26, Terry McCracken wrote: >>> >>>>Many IM's and GM's have felt the teeth of these and many other programs like >>>>Fritz, Junior, Tiger etc. >>>> >>>>However, it's true they lose to much lower rated players, even to me >>>>sometimes!;) >>> >>>Humans have a hard time coping with this. >>> >>>A human who plays chess is an apple. A computer that plays chess is not an >>>orange, but it's not quite an apple, either. You can compare them, but not >>>perfectly. >>> >>>A ludicrous example: Compare an unarmed human with a tank. A tank can blow up >>>a house at a range of over a kilometer, and can survive machine gun fire, >>>whereas an unarmed human cannot blow up a house at a range of even one meter, >>>and would fare badly against machine gun bullets. >>> >>>On the other hand, there are ditches that a tank cannot get out of, while a >>>human might have an easy time with these ditches. >>> >>>You can't deny the tank its strengths just because there's a ditch a human can >>>handle better. >>> >>>End of ludicrous example. Computers are getting better at being uniformly >>>strong, but they are not now, and will probably never be, perfect human analogs. >>> There will always be the possibility that a weak human player can say "duh" to >>>one of these machines -- sometimes. >>> >>>bruce >> >> Everytime this kind of argument arises, I have the same impression: There's a >>problem of definition. What do we call 'GM strength'? >> >> A. If we speak of 'quality' of chess (whatever this can mean), most chess >>players will probably agree to one of these possibilities: >> A.1. Computers are not GM strength because they show lack of understanding >>too many times to be considered GM's. >> A.2. Computers and humans do not compare (like the soldier and the tank). >> >> B. If we speak of 'quantities' (namely ELO rating), we can certainly compare >>humans and programs, and say _with numbers_ if programs perform like GM's or >>not. >> >> So, as long as we don't specify the definition of 'GM strengh', there will be >>discussions comparing apples with oranges. > >There can be no other definition then ELO. > >We don?t consider GM Kasparov or GM Fischer the strongest grandmaster because of >the beautiful games they produced, even though they produced many, but so has >many other grandmasters. We consider them the best because of the Elo ratings >they generated, meaning wins, losses, and draws. > >There is on other measure in chess. You don?t get points for beauty, only for >wins and draws. > > I disagree there. I don't pay a lot of attention to Elo at all. I recall Fischer's _games_ and tournament victories... not his rating at the time he was playing. He was the best of the 1970's because he won the WC title, not because he had a big Elo. He could have been rated 2500 and he would have _still_ been the 1972 WC... Elo ratings do _not_ tell you exactly who is the strongest player around, until we see the time where _everybody_ plays everybody an equal number of times. Human vs human chess is not a transitive relationship. Elo is just a statistical approximation based on expected game outcome between two opponents in a rating pool. It does not guarantee a winner. Comparing a 1972 Elo to a 1992 (or 2002) Elo is pointless. The rating pools change. Too many take this as an absolutely perfect statement of fact. It is just a way of predicting heads or tails more accurately since there are more than two opponents anyone can play. >> >> José C.
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.