Author: Uwe Meißner
Date: 13:38:48 02/28/04
Go up one level in this thread
On February 28, 2004 at 01:22:10, Derek Paquette wrote: >On February 27, 2004 at 18:44:23, Uri Blass wrote: > >>On February 27, 2004 at 18:17:18, Bryan Hofmann wrote: >> >>>On February 27, 2004 at 18:11:39, Derek Paquette wrote: >>> >>>>Ok, I am a littel confused? >>>> >>>>These engines are code named so he doesn't know which engine he is playing >>>>exactly? or? >>>> >>>>what exactly is going on, >>> >>>Same here, but I do know this Nickel is not a GM level player OTB >>>http://www.fide.com/ratings/card.phtml?event=4623703 >> >>Totally unimportant. >> >>The games are to check if strong correspondence players with the aid of >>computers can beat computers. >> >>Note that Arno nickel does not use many computers to analyze so the opponents >>computer opponent have time advanatage and to make the task harder arno nickel >>also does not know the names of the chess programs that he is playing so he >>cannot prepare against specific program unless he guess correctly the opponent. >> >>Uri > >While that might be, and probably is true, where the programs do have a time >advantage, we are taking days on each move. He is an experienced player using >programs to assist him. All hiw knowledge and experience ASWELL as the programs >chugging away at the problems, now that I think about it longer i wouldn't be >surprised if he draws most and wins some. > >Has he said at all what his results have been just playing the programs on their >own at normal time controls? I don't think he did ever play the programs at normal time controls, because he is only interested in using them for analysis. But he said at the very beginning (and did so) that he would use no anti-computer strategies. And he did not want to profit from any book mistake etc. Then after about 20 moves played he stated, that it might be impossible to beat the engines in some types of balanced openings. That means, in a future match he would make changes in his openings in order to get better chances. The programs avoid a lot of mistakes, they would make with normal (say classical) time controls. Originally his aim was 4,5 : 1,5. By now I think 4 : 2 or 3,5 : 1,5. But it would be the first match-win vs. computers in such a tough cc experiment. Some other experienced cc players (4) managed only a 2 : 2 in 2002, when playing against Gandalf 5 (and they knew whom they were playing). Uwe
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.