Author: Thomas Lagershausen
Date: 10:19:02 03/19/02
Go up one level in this thread
On March 19, 2002 at 07:57:51, Uri Blass wrote: >On March 19, 2002 at 06:02:56, Thomas Lagershausen wrote: > >>On March 19, 2002 at 03:02:11, Jerry Doby wrote: >> >>>Mark my words. He will get a few draws but no wins. The same would apply to >>>40/2. Programs are just to strong now for mediocry has been Grandmasters like >>>Boris. A more interesting Challenge would have Been against Moro or Leko, >>>someone in the top ten. Maybe even Dreev who has demonstrated mastery against >>>computers on ICC. >> >>Your words are very funny.One lucky punch by Fritz and your are in heaven.On >>40/2 things are different.The GM has time to calculate and can make his better >>positional performance to work for his gamescore.It looks like you have missed >>the two losses by the real strong Rebel Century 4 with anti-gm settings against >>VanWely. >> >>Second, Gulko could easily draw the game in move 41 and goes for a win by the >>mistake 41..b3?.Humans can learn very fast.In the next game Gulko will play more >>accurate and much safer.See his first game.He won very easily a pawn and >>DeepJunior had never a chance for a win in this game. >> >>No one with GM-strength would have your opinion.They know so much about chess. >> >>TL > >You cannot talk for the GM-strength players. But i can talk for myself as a player that is able to beat sometimes this modern chessprograms. >I am sure that there are GM players who think that computers are strong. I believe this too.They are calculating very good variations. > >I do not know if gulko is going to win games but it is clearly possible >and history tells me that tiger did not lose a game in a tournament when GM's >played. Time will tell us.Not much games between GM´s and programs at the moment. > >Not all the players were GM's but the time control >was slower than the games against gulko and >Tiger did super gm performance. Some IM´s do this too in weak tournaments and loosing a lot of games against strong GM´s. > >Some facts: >1)programs are capable of doing super 2700-2800 performance >at tournament time control(Tiger and Deep Junior proved it in the past) Tiger do this like some strong IM´s and weak GM´s in other lowlevel tournaments too. DeepJunior 6 had played in Dortmund horrible games against Piket and Kramnik.The other GM´s didn´t play in there first accounter there normal chess against the "Golem". > >2)The time control against gulko is faster than tournament time control > >conclusion:They have a chance for 2800-2900 performance that means >probably not to lose a game against gulko and I do not see why >no GM is going to agree with Jerry. I don´t see any chance that the programs can play Elo 2800-2900.It is more realistic that under the timecontrol 60min/game +10sec GM Gulko is playing with the strength of Elo 2450-2500.So the matchresult could probalbly be a draw. TL > >Uri
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.