Author: Uri Blass
Date: 10:46:49 05/23/03
Go up one level in this thread
On May 23, 2003 at 11:15:37, Stephen Ham wrote: >On May 23, 2003 at 04:59:53, Angelo Ciavarella wrote: > >>In response to my question as to whether the champion CC players can outperform >>computers,FM Tom Brownscome of the U.S. Chess Federation said that most people >>believe that the world's best CC players are better at CC than the current best >>computers.He also said that most top-level analysis is done by master-level >>players working with computers. > >Dear Angelo, > >Clearly this is a topic of interest to you, given your repeated posts. As a >dedicated Correspondence Chess player myself, I have some interest in this >matter as well. Thanks for your interest. > >Indeed, as of the present time, I think that most human CC players do believe >that the very top CC players, playing without move generating chess engines, are >superior to the best chess engines in CC. Personally, I think they're wrong. >Before I try to state my case, please allow me to provide some perspective for >you. > >I tried to test this myself, since my rating hovers around ICCF-2500. So in >December 1999, I played four CC games versus the top rated chess engines of that >time, Fritz 6a and Nimzo 7.32. They were assisted by one of the fastest >computers of that day. I played without any computer aid. See below for the >match conditions and the result. > >http://www.correspondencechess.com/campbell/ham/ham.htm > >In hindsight, in order to try to entertain the masses, I hurt my prospects. I'm >not making escuses...I lost fairly. However, I moved rapidly, at the rate of >approximately 1-move/day, and also published my notes live (just annotating the >games took a lot of time). Also, my only time for analysis was after my family >went to bed, so during the late night hours, I was already fatigued. I had an >advantage in 3 of the 4 games, but failed to win a single game. Why? Human >error. Machines just don't make the mistakes that we mortals do. Alsthough >insuffucient games were played from which to draw any conclusions, I must say >that I was shocked at how difficult it is to defeat the silicon monsters. For >the record, I did not play in anti-computer style, since the point of the >experiment was for me to play as I would against a human. After all, in CC these >days, we never know who/what our opponents are! > >My point though is that since late 1999, there have been large gains in chess >engine strength and hardware speed. Also, I'm not anywhere near being considered >one of the "best" CC players. Maybe someday...;-) > >Since then, Uri Blass has won the Israeli CC championship using only chess >engines to generate his moves. Uri selected moves generated by a number of >different engines, so there was some indirect human interevention. Still, the >point being that the games were won strictly by chess engine moves. I need to do some correction here. Almost all the moves were generated by computers. There were some moves in the Israeli final that were not book moves and were not generated by computers(but I used a lot of computer analysis to decide about them). I believe that they did not change the results by more than 1/2 point out of 8. they happened in 2 games that I won and in 2 games that I drew. In the games that I won I believe that they only made the win faster. In the games that I drew in one case the reason was that I wanted to win the game and the computer move was leading to a draw but my move also probably leaded to a draw except the fact that I got an inferior position. In another case I found that line that the computer suggested is probably losing and chose another move. One move later when the computers still evaluated my position as better I decided to offer my opponent a draw because I felt that the gravitation of the position is against me. In other words I did not see a forced winning line for my opponent but found winning lines for my opponent in comp-comnp games. My opponent did not believe that he is better so he accepted the draw. 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.