Author: Roy Eassa
Date: 16:09:40 09/14/01
Go up one level in this thread
On September 14, 2001 at 18:26:24, Bruce Moreland wrote: It's hard for me to understand calling a game a draw when in reality it is a win. The tables do not take sides -- they reflect the actual truth of the positions in question. A fully pre-announced forced checkmate in 51 (w/o pawn moves or captures) is now to be a draw?? >On September 14, 2001 at 17:40:00, Roy Eassa wrote: > >>On September 14, 2001 at 12:32:38, Bruce Moreland wrote: >> >>>The endgame databases don't contain values that are in accord with the FIDE >>>rules. >>> >>>Some have proposed that the FIDE rules be changed, but FIDE seems insistent >that the tables be changed. >>> >> >>Huh?? >> >> >>>Some programmers have suggested that the rules be changed for tournaments. >>> >>>I think that the FIDE rules are fine. >> >>You're joking, right? > >There have been a couple of times that people have shown that certain endings >must take longer than 50 moves to win. The first one was KP vs KNN, which was >the subject of some remarkable work by Troitsky, well before the advent of >computers. > >As a result of this work, there was a special exception to the 50-move rule for >this ending. This was all a long time ago. > >When Thompson did his 5-man table work, and someone else (one of the ICCA guys, >I think Jaap van den Herik) did work with KRP vs KBP with a pawn ram, it was >shown that these endings took longer than 50 moves, and exceptions were written >for them. > >I have a 1986 edition of the FIDE rules, and it mentions that three basic ending >classes are extended to 100 moves: > >1) KRB vs KR. > >2) KNN vs KP, if the pawn is blocked and not beyond a certain square, which >varies with the file of the pawn. > >3) KRP vs KBP, with an a2/a3 ram. > >It became obvious that there were more exceptions, for instance KBN vs KN. That >one can take over 50 moves, too. > >I think that at some point the chess players said enough. KRB vs KR, for >instance, has some pathological cases requiring more than 50 moves, which are >not likely to be encountered in actual play. The other exceptions were getting >hard to manage and there were getting to be more endings that would require >exceptions. > >So finally they just canned all of these exceptions. Now you get 50 moves, >period, so the data in KRB vs KR, for instance, is not in accord with FIDE >rules. > >I am in favor of the 50-move rule as it stands now. I think that if someone >wants to use tables, they should put up with this problem or fix it. > >This issue is a threat to come up at players' meetings at computer tournaments. >People who use the tables want the rules changed to accomodate them. I think >that is unfair to those whose programs play in accord with FIDE rules, meaning >everyone who doesn't use tables. > >bruce
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.