Author: Dann Corbit
Date: 08:33:31 07/15/00
Go up one level in this thread
On July 15, 2000 at 02:55:22, Drazen Marovic wrote: >On July 15, 2000 at 02:13:57, Dann Corbit wrote: > >>On July 15, 2000 at 01:59:21, Drazen Marovic wrote: >>>Listen I AM a GM!! and have probably been playing chess longer than you have >>>been alive i have played and beaten some of the best in the world when they were >>>the best. I know something that you will never know, i KNOW what GM strength >>>is. Since you don't know you really have no need to argue it is you who are >>>wrong. >> >>I won't argue that you are a better chess player than I will ever be, or even >>that you understand strength better than I do. But suppose that DJ falls to >>stonewall by any chess expert. Is it still a GM? > >The statement i made has nothing to do with that. If a GM player loses to a >weaker player does that mean that he's not a GM? That's exactly what it means, if he loses to that player consistently. >As i said previously, GM >strength and GM title are two different things. If a comp plays one game in 10 >at GM strength, then it plays GM strength chess(just not always!). It is not >even a question as to whether comps can play "GM STRENGTH" chess. Comps will >never be GM's because they are not human! Never say never. FIDE could reverse their position under new leadership and allow computers to become GM's. >"GM strength" is another issue. When >Nezhmetidinov(IM) played Poluguyevsky his immortal brilliancy, there is no >question that the game and the play was "GM strength". GM strength does not >mean GM title. If i can know the weaknesses of any player in the world (and >they have no gamesmanship i.e can change to play against my particular >weaknesses i could beat them consistantly(if i wasn't old:)). It would have no >bearing on whether they played "GM Strength chess". It would just mean that my >chess appeared stronger against them because i'd have more information about >them than most other players. If kramnik could get ahold of the preparation of >all the GM's before the tournament and knew exactly how they planned to play he >would win the tournament hands down. He'd be playing stronger against them >because he'd have more knowledge. It would not mean that the other players were >playing weaker than usual, or not playing "GM strength" chess. I wonder. Don't players prepare for their opponents? I find it surprising if GM's face the competition blind and don't look for weaknesses in who they are going to face. >>I have a degree in mathematics. Therefore, I suspect (unless you are also a >>mathematician) that I know some things in that area that you don't. >> >>I have not argued that DJ is not a GM. I have only argued that DJ is not >>mathematically proven to be of GM strength. In fact, I am correct about that. >> > >No you are not right Mathematics does not prove anything about "GM strength"! >Statistics judges the result not the qaulity of play. I am addressing the point "PROVE" not the point "strength." You are saying, "I can look at these games and know that the programs are as strong as GM's." I am saying that "PROOF" is a scientific measure. There is no scientific measure in an opinion, no matter how well reasoned. >If khalifman or Huebner >do not score a GM norm in this tournament it has no bearing on whether the >qaulity of their chess play was "GM strength". Straw man. Where have I said that not scoring a GM norm means you are not provably of GM strength? >If a rich individual pays off >enough people statistically he could get the "GM TITLE". Straw man. A "straw-man" seems to be one of your favorite techniques. You set up a silly and indefensible position and then beat it up. That's all well and good but it has no connection at all with the arguments I have put forth. >It would not be an >actual statement about the strength of his play. You may well know the recent >discussion of Myanmar GMs(though some are better than many think). This is actually a support of my position rather than an attack on it. If it is possible to achieve a GM title and yet not be a GM, that means that even achieving the title is insufficient evidence. How will we *really* know if someone is a GM? 1. They must meet all GM conditions as set out in FIDE against fair and equitable competition. It seems some matches may have been thrown. We could say that the experimental evidence is flawed or faked. This is not unknown in ordinary scientific experiments. 2. They should be able to demonstrate their rated strength in future competitions in a statistically significant way. If they could perform at GM level for dozens of contests against other GM's that would remove doubt, wouldn't it? In science, this is called repeatability and it is one of the fundamental cornerstones of science. If an experiment has the same outcome after many trials, then we consider the result significant. If other outcomes are more frequent, then we consider the unusual outcome a 'sport' or outlier. >>One million emotional arguements or artistic arguments or chess expert opinions >>do not alter that. Is DJ a GM? I think it *might* be of GM strength -- even >>that the evidence is leaning that way. But I not only think it has not been >>proven, I really do believe that once people figure out how to play >>anti-computer strategy it may not pan out to be the GM player you imagine. > >Because you can consistantly beat someone because you have extra knowledge of >their weaknesses does not mean that they don't play "GM strength" chess. Nor have I made this claim. However, it would not offer proof that GM strength chess was being played either. >>Have you considered the Anticomputer chess site of Raphael Vasquez? I think >>most GM's have no idea how to play against computers. Further, that there are >>many IM's that can do much better against computers than the average GM because>>they know how to play them. This is a sin of the GM, since if it is your life, >>you ought to understand your opponent. > >A Human GM is a sportsman with what is reffered to as gamesmanship, i.e move >fast now move slow now, sigh unexpectedly, sack when you think your opponent >does not like sacks. We change to suit our opponents. Comps are not GMs >because they are not sportsman and never will be. That is your definition of GM. Does it say a GM must be a sportsman anywhere in the FIDE rules? I will agree that comps will never be sportsmen. I won't agree that they won't ever be GM's. That's nothing more than a matter of definition anyway. One way or another, they will eventually become better players than the human GM's. >That is not determinitive of whether >they CAN play "GM strength chess". What I have addressed in my arguments is not whether thay can play GM strength chess or not. Rather, I have addressed whether or not this has been *PROVEN*. There is evidence that they are of GM strength. We have a computer earning the equivalent of a GM norm. We have an expert opinion that says they are of GM strength. This does not constitute mathematical proof. These are simply favorable observations. There are also negative observations. Given enough data, we will have mathematical proof, one way or another.
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.