Author: blass uri
Date: 21:59:19 05/28/00
Go up one level in this thread
On May 28, 2000 at 20:42:59, Michael Neish wrote: >On May 27, 2000 at 23:22:20, blass uri wrote: > >>Junior5.9 won 2432 player in correspondence game. >>I believe that chess programs are at IM level in correspondence games. >> >>I expect to win an IM (2456) with black in a correspondence game when all my >>moves are computer's moves(in the case of the game against the 2456 not all the >>moves were made by the same program and there were cases when I did not like one >>program and prefered another program) >> >>I believe that computer programs are at IM level inspite of the fact that the >>IM's have the right to use programs to help them and that my level is better >>than the average IM level at correspondence games. >> > >Coming from someone else I would have thought the post was just another troll! > >:) > >Actually I don't think I agree with you. As James Robertson said, it's only one >game. I replied that my impression is based on more than one game. > >You say that the player has to know when to override the computer's >recommendation whenever it is wrong. In other words, if the player follows the >computer blindly he will make bad moves every now and then and weaken his >overall strength. No I said that in this case the player is better than the average IM Steve Ham(2508 ICCF) has problems against Fritz and Nimzo with no help(In one game he admitted that the computer has a better position and in the other games he also see that it is not easy to win). So basically computers are IM strength in correspondence play >if the player using the computer is also IM strength and knows when to disregard >what the computer is saying when it's wrong. > >I might have excellent positional sense but weak tactical ability. I test out >my ideas by running them through the computer, find out that it doesn't set off >any losing tactics, and then go on to choose that move. I think that this is a mistake for most players because most players have not excellent positional sense and it is better to trust the program if you are not sure which move is best and not to analyze and use the program only for checking tactical mistakes. The fact that humans use programs only to check their ideas is the reason that programs are better than most humans in correspondence games. I think that choosing the computer moves almost always is a better idea. I think that the team human+program should be productive in the tactics by going backward and forward and in some other cases. 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.