Author: Uri Blass
Date: 21:54:22 04/30/03
Go up one level in this thread
On April 30, 2003 at 19:05:22, Kurt Utzinger wrote: >On April 30, 2003 at 13:47:23, George Tsavdaris wrote: > >>On April 30, 2003 at 13:29:09, Kurt Utzinger wrote: >> >>>On April 30, 2003 at 12:28:54, WILLIAMS POLK wrote: >>> >>>>Would you give it 2400 elo? Can you still Draw against it like the other >>>>engines? >>> >>> 2400 Elo=yes / second question: so far played no games myself. will >>> follow but I think a draw isn't impossible. >>> Kurt >> >>Why do you think that Shredder (7/7.04) is only 2400 elo? According to >>the results of Fritz 8 and Junior 8, the elo of the todays strongest >>programs is around 2800 on a fast machine of course. Don't you >>believe that on a pentium 3 MHz Shredder 7 is at least 2720 elo? >>Of course the real question is if <Shredder7+ Pentium X MHz> is 2400 elo. >>For example Shredder 7 on a Pentium 80.000 MHz is surely 2800 elo as >>chess is not played by Shredder or by pentium but with both of them. > > It might be that I did not understand the question correctly. I thought it > was meant if Shredder7 would [at least] have 2400 Elo and therefore > answered with yes. People know my opinion: computer ratings are 200-300 > points too high, but I do not want to again argue about this. I do simply > not understand why it should be possible for 2000-Elo-players to get 2-3 > draws out of 10 games if the top programs have 2700-2800 Elo. Maybe it is possible for the same players to do even better against kramnik and kasparov in case that these players do not know the opponent and do not try to play in a different way against them(for example by avoiding positions with big chances for draw). It is known that GM's often agree to GM draws in 10-20 moves and I think that with that style 2000 players have chances to draw against them. Neither the > matches Deep Fritz vs Kramnik nor Deep Junior vs Kasparov have convinced > me. Both programs made too often bad moves and only psychological > reasons and severe public pressure for the players prevented clear wins > in favour of the human beings. There is another factor that people seem to forget. Not winning on purpose in the hope that a draw will make more interest in another match. I think that matches against humans when humans get so much money only for losing is a bad idea. I think that there is enough circumstances evidence to say that kramnik did not win on purpose. He did mistakes that even 2000 players are going to feel bad to do them giving a piece by 1 ply mistake and resigning in position with practical chances. The try of kramnik to justify his decision by saying that he thinks that objectively he is losing(I was not convinced about it) is absurd because chess is a practical game and you should resign only when you are sure that there are no practical chances. It is always better to resign too late than to resign too early and if there is a doubt you should continue. If you see that you lose material then the next trivial question is if there is a chance for a fortress or a stalemate combination and only if you see that there is no chance then you should resign(if you are in time trouble and has not time to calculate it then you should continue). I think that it should be obvious for 2800 players. 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.