Author: Mike Byrne
Date: 22:12:20 02/24/03
Go up one level in this thread
On February 24, 2003 at 09:59:12, Mike Hood wrote: >On February 23, 2003 at 16:23:16, Mike Byrne wrote: > >>Ruffian , among others, like Shredder 7, gets confused by this position, It >>thinks it's mate in 18 for black and plays Kxd7 -- which, unfortunately leads to >>a Mate in 18 for BLACK!. The problem here is that the EGTB does not recognizing >>castling rights and the program is not examining them either. >> >>Proper play is Kd8 and black mates in 18 >> >>[d] 2k5/3P4/8/8/8/8/1r5p/R3K3 b Q - 0 1 >> >>Ruffian 1.0.1 analysis >> >>New game >>2k5/3P4/8/8/8/8/1r5p/R3K3 b Q - 0 1 >> >>Analysis by Ruffian 1.0.1: >> >>1. -+ (-#17): 1...Kxd7 2.0-0-0+ Ke6 3.Kxb2 Kf5 >>2. -+ (-#18): 1...Kd8 2.Ra8+ Kxd7 >> >>(Byrne, Secane, PA USA 23.02.2003) >> >>Proper Analysis: >> >> >>If black plays Kxd7, then it is mate for white in 18 with 0-0-0 as shown by >>Fritz 8. >> >> >>[d]8/3k4/8/8/8/8/1r5p/R3K3 w Q - 0 1 >> >>Analysis by Fritz 8: >> >>1. +- (#18): 2.0-0-0+ Ke6 3.Kxb2 Ke5 4.Kc3 >>2. -+ (-#16): 2.Ra7+ Kc6 3.Rh7 h1Q+ 4.Rxh1 Rb1+ >> >>(Byrne, Secane, PA USA 23.02.2003) > >Hey! Copyright infringement! > >It was me who devised this position, and I posted it in CCC on November 9th >2001! (Topic: "Endgame Tablebase Problem without Castling") > >Oh well, it is a good position, but admittedly very unnatural. One of the >parameters of Eugene's EGTBs is that there is no castling. This is a sensible >limitation for 5-piece EGTBs, but by the time the first 20-piece EGTBs are >generated Eugene might have to rewrite his code :) Nice work - I thought it was a thing of beauty. Now the question is - where I did come across this position (and it wasn't here)?
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.