Author: KarinsDad
Date: 15:44:53 04/30/99
Go up one level in this thread
On April 30, 1999 at 17:14:36, odell hall wrote: > >On April 30, 1999 at 15:54:40, KarinsDad wrote: > [snip] > > >Well Let me ask you two things? First What would you define as evidence? Well, for one thing I would think that since GMs need to get their norms via GM standard time tournaments, that at least the same requirements should be used for computers (the time control, not the tournament setup). >Secondly How many grandmasters Must fritz beat before you would consider it to >be a grandmaster? I respond to your question with a question. How many GMs has Fritz beat in standard time (40/2 G/1) match (6 games or more) situations (as opposed to G5s, G25s, G30s, and G60s)? I do not dispute that computers may be GM strength. I dispute that you have shown me the evidence. IMs defeat GMs ALOT. However, IMs do not get their GM norms by beating GMs one at a time and in situations where the programmers may have set up the opening book versus that particular GM. If a program was to play 6 different GMs without preparation and win 4 of the 6 tournaments at standard time controls (relatively equivalent to what an IM has to do to get his 3 GM norms since programs are not allowed to participate in FIDE events), then you would have fairly conclusive data. Until you have these numbers of tournaments with one program, you do not have a conclusion. You have an opinion. By the way, you have illustrated more data with your message below. Unfortunately, you have left out a lot of the details, therefore, it is not enough either (although if you had the details, it may be enough to convince me otherwise). I am not totally familiar with all of the matches you mention (still nowhere near one program playing 6 matches at standard time controls). However, a few of them are not based on 40/2 G/1. KarinsDad :) The Problem is as long As computers are not allowed to >participate in Fide Events and Achieve legitamate norms There will never Be any >"evidence". But this doesn't stop me from using good common sense, If one >consistenly beats grandmasters then one is a grandmaster! Show me a >international master on the face of the planet that could beat Judit Polgar in a >match under any conditions or circumstances? If you take the performance of top >programs as a whole they have more than proven themselves to be grandmaster >strength. You speak as if there is no evidence , when there is overwelming >evidence. Humans in order to get the GM title need only 3 Grandmaster norms!!! > This means basically that they only have to achieve a grandmaster performance >in three tournaments, to get the title. ONLY three tournaments!! I don't think >even the most skeptical person if they are honest with themselves would deny >fritz5 could easily achieve a grandmaster Norm if allowed to compete. Ok let me >list some Evidence 1. Hiarcs Defeat of 2485 elo rated Deen Hergott in a six >game match 2. Rebel winning of both Samuel Cups I and II in 1997 -1998 over >Several International masters with over a 2600 Performance rating!! 3.Numerous >indivisual encounters between computers and grandmasters at 40/2 where the >computer was the victor 4. Rebel Annand match 5. Matches Played at 40/2 between >Crafty and Grandmaster Larry Christian Crafty won. 5. Fritz5 Defeats 2577 elo >Judith polgar (game\60, Game\30. The Strongest Player in the History of the >Game says they are 2600! This all means nothing??? >Oh I forgot to mention the outstanding Perforances of computers at the Aegon >tournament With Mchess and Rebel with performances ratings over 2600el0. >No doubt I am leaving alot out!! When you have all these results and people are >still having doubts, this makes me wonder what people are looking for. yet a >human only has to score three norms to be considered a GM. No doubt had all the >above events been Qualified by Fide as Norms, Computers would have earned the >title long ago. Could you show me a human international master that could >achieve any of the above? I ask you again, if this is not suffient evidence >than what would be? Does programs have to Grow wings and then Fly to Heaven and >Beat God? Perhaps then you would still say there is no evidence!
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.