Author: Vincent Diepeveen
Date: 16:07:51 01/31/05
Go up one level in this thread
On January 31, 2005 at 17:56:09, Mark Rawlings wrote: >On January 31, 2005 at 04:37:52, Uri Blass wrote: > >>On January 30, 2005 at 21:25:09, Mark Rawlings wrote: >> >>>On January 30, 2005 at 13:28:17, Duncan Roberts wrote: >>> >>>>On January 30, 2005 at 12:53:11, Helmut Conrady wrote: >>>> >>>>>On January 30, 2005 at 10:56:55, Mark Rawlings wrote: >>>>> >>>>>>Very interesting! >>>>>> >>>>>>Is the KRNKNN mate in 262 still the longest tablebase mate?? >>>>>> >>>>>>Mark >>>>> >>>>>Yes these 523 plies are the longest DTM so far. I expect this record holds in >>>>>the 6 piece endings. >>>>> >>>>>Helmut >>>> >>>> >>>>was it always suspected, it would be the longest, or did they know only after it >>>>was done. >>>> >>>>do you have any thought on longest 7 piece and how long ? >>>> >>>> >>>>duncan >>> >>>I was always thinking that the egtbs with pawns would have the longest mates, >>>but I guess that's not the case. >> >>I guess that you do not consider the 50 move rule. >> >>Mate in 150 moves with no pawns(6 pieces) is simply a draw if my memory is >>correct and the same for mate in more moves. >> >>1.*** >> >>50...capture >>51.*** >>100...capture >>you already have 4 pieces and it is a draw or mate in less than 50 moves(I am >>not sure what is the longest mate but if I remember correctly it is less than 50 >>moves). >> >>I am not sure what is the longest mate for every tablebase of 4 pieces and I >>think to write a program to calculate it based on movei structure by having a >>loop on all possible 4 piece positions and probing the tablebases. >> >>Knowing the correct values for every position can help me to check if I have no >>bug in the program that probe tablebases. >> >>Uri > >Right. I was not considering the 50-move rule since the tablebases don't take >that into account. It's really more of a theoretical topic for me. (I never did >like the 50-move rule, since it's always possible that the rule could be >changed.) > >Mark The rule will NEVER be changed again in FIDE, it's 50 moves and not a move more. There was a short lived experiment to make it 75 moves for specific endgames, but after a few years this caused so much problems that it was changed back to 50 moves for all positions and NEVER again will get changed within FIDE. I remember 1 game Kroeze-Diepeveen which lasted 7.5 hours, for the dutch 'open' championships in Dieren, a few years ago. Getting to those 50 moves just took forever for me to draw. In fact i had 1 game against Victor Korchnoi too. Dead draw. After 7 hours of play i fell asleep and blundered and he still won. That should have been a 50 move rule game too. Too bad i blew it already after 7 hours. I had at the time 2150 as a rating so i guess he didn't like drawing me, him being a 2690+ rated person at that time. Note i'm FM nowadays but get little chances now against such big guys :) Nowadays most time controls are shorter. A 7 hour game is quite difficult to get now. The FIDE level now is 40 moves in 2 hours + 1 in 20 + 15 or 30 minutes all. So a game never can last longer than 6.5 or 7 hours now. That's why a month ago i had a game which took 6.5 hours nearly. Very quickly i managed to win KQKR there :) A year ago in KNSB i had a position which i could play on forever kind of. I had night + 3 pawns (f,g,h) against opponent king + rook + 1 pawn (h pawn). 50 moves is really a lot there. Basically i shuffled for 30 moves. By then his clock was so low that when i exchanged myself, he had only 2 minutes left or so and had to move instantly. 50 moves against someone who has all time of the world is really a lot. He made a mistake and lost. I was real happy, i played for my team and they needed that point to win the match. Additional it is a big d***head at ICC who shouts around the biggest nonsense there, as he can keep anonymous there. With such timecontrols as we have now in tournaments and the prospect that it won't get slower, 50 moves really is a lot. It will never get more moves. Vincent
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.