Author: Helmut Conrady
Date: 11:17:57 05/12/05
Go up one level in this thread
On May 12, 2005 at 14:09:32, Dieter Buerssner wrote: >On May 12, 2005 at 06:08:22, Rafael Andrist wrote: > >>On May 12, 2005 at 04:46:43, Walter Faxon wrote: >> >>>Have any egtb's been generated with the current rule in mind? >> >>Yes. Search the CCC archives for "DTZ" > >One might note, that pure DTZ is not enough in general - for example for kqpkp. >But a slightly changed DTZ generator (perhaps together with some additional >logic in the access routines to take the half move counter into account) can do >it. > >However, in positions, that don't result from an immediate capture or pawn push, >the problem will be pretty unsolvable. (Normal TBs cannot do it either, even >when not considering the 50 moves rule, because you might have had a repetition >already on the path to this position, and the shown mate might go through >another repetition. In games played from start to end by an engine with TBs this >will never be a problem). > >Regards, >Dieter The best metric according to the rules is DTZ50. Rafaels program "Wilhelm" can access thoes databases very well. (You might know that.) I have written something about this in the German magazine CSS. Helmut
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.