Author: Uwe Meißner
Date: 02:51:24 09/11/03
Go up one level in this thread
On September 10, 2003 at 12:59:21, Uri Blass wrote: >There is a correspondense player(ICCF rating 2384 and Fide rating 2269) >who believes that he can achieve no worse than 5-1 against computers in the same >conditions that Arno Nickel plays(2585 Elo in cc) > >see >http://www.correspondencechess.com/bbs/messages/49460.html > >He told me by email that he is also interested in playing a similiar match. >I do not think that chessbase will agree (they have not infinite computer time) >but I thought that maybe it is possible to use some posters here as operaters(I >do not have computer time for that task) > >What is your opinion? > >Can he get 5-1 against computers? >Are there people with fast hardware that can give their computers to play >against him(we need 6 different fast computers when Tansel Turgut does not >know the programs that he is playing). > >Uri I think Tansel Turgut will not win even one game when he carries on to underestimate the engines as he did in his message to the TCCBM (discussion board of the ICCF). He just only wants to present himself and thinks, well if an ICCF GM like Nickel expects to win 4,5 : 1,5 , then will I have to win with a half point more. The context of Nickel's experiment is a little bit more serious than the simple wish to beat the machines. As I followed his articles through the years, he tries to give an intimate insight in the differences between computer evalutations (on a very high level) and human chess thinking (also on a very high level). - The games itself are the most important purpose not the naked result. And that is why the match is presented on the server chessfriend.com, where everybody can watch the games with live comments of GM Nickel. The reflection time will be 40 days per 10 moves (for each game), though everybody can expect that most games will be played faster in the opening. Nickel will not use more computers (only one), as it is not his intention to count better as the other computers, but to have better strategies. This means also, he does not try to play specific "anti-computer strategies", but quite normal chess as to be seen among humans. Imagine you play a correspondence game against a human, who intnsively uses computers, but plays quite normal openings out of the book. What you start to play and "anti-computer strategy" to beat him? No, if you are a really good player you would not start to make artificial moves, you would carry on to search for the best moves. So I just wanted to give you some serious background of the match, organised by the magatine "Computerschach & Spiele" (CSS) - not Chessbase at all! - and the server Chessfriend.com (CFC), which will start on 1st of October. There are also specific rules for the match, which are very important, e.g. to prevent humans from manupulating the computer-games (log-files after each move), rules for the modus of evaluation (best-move mode or multi-variation?) rules, how long the computers shall take for each single move etc. As you see, it would be interesting of course to have a parallel match for reasons of comparisons, but it is not so easy to copy the original version. By the way it might be, that GM Nickel will play a second match starting on December against Donninger's project "Brutus". This would be even more fascinating. 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.