Author: Uri Blass
Date: 06:20:51 11/30/03
Go up one level in this thread
On November 30, 2003 at 08:58:10, Gerd Isenberg wrote: >On November 30, 2003 at 08:28:22, Matthew Hull wrote: > >>On November 30, 2003 at 07:05:36, Gerd Isenberg wrote: >> >>>What a WCCC. I guess all participants and TDs are a few years older by now and >>>need some vacations ;-) >>> >>>Anyway, big compliments and congrats to Stefan Meyer-Kahlen and his great >>>program Shredder. Also congrats to the very unlucky Frans Morsch and his Fritz. >>>Both programs really played great chess! >>> >>>About the 3-fold repetition issue - after Gian-Carlo's statement here, that >>>Jonny didn't know about 3-fold repetition at all, but only the fritz-gui, >>>i'll think the decision made by the TD was finally correct. >> >>It matters not that the engine could not detect a 3-fold, it is a draw according >>to the rules of chess, just like the 50-move rule or checkmate. > > >Simply no. There is no automatic draw, if a 3-fold repetition occurs. >You have to claim it before the position occurs: > >1.Stop the clock >2.Call TD >3.Tell TD that the program intends to force a 3-fold repetition > by doing that move. >4.The TD proves whether it results really a 3-fold repetition, > and has to agree if it is true. >5.Otherwise there is a time penalty. > >Obviously this is a FIDE rule for human chess, because humans may erroneous >claim a draw, e.g. not considering castle or ep states correctly or whatever. > >This seems a bit anachronistical to computer chess, >but even chess programs had and have bugs with this issue, same for 50-move >rule. > >The "main" point is IMHO that Jonny and Shredder didn't implement the correct >knowledge, but the fritz-gui, Jonny was playing with. The fritz gui was part of jonny(otherwise there was no reason to allow jonny to use it) If it was used in previous games then it should also be used later because you cannot change the players during the tournament. >I guess there was no explicite rule about the issue, if engine and external >interface disagree. IMHO the engine is the boss here. > >Gerd No The fritz gui did not implement the correct knowledge It declares this game as a draw when it is no draw because enpassent possibilities are changed. My first thought was a different position that even my program is wrong and I did not care to fix it but I found that Fritz Gui's error is clearly bigger. [Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "New game"] [Black "?"] [Result "*"] [SetUp "1"] [FEN "8/3p4/8/K3P1pr/8/8/8/7k b - - 0 1"] [PlyCount "9"] 1... d5 2. Ka4 Rh4+ 3. Ka5 Rh5 4. Ka4 Rh4+ 5. Ka5 Rh5 * 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.