Author: Heiner Marxen
Date: 16:21:03 02/04/01
Go up one level in this thread
On February 04, 2001 at 16:43:52, Jonathan Lee wrote: >To the perfectionists out there, how important is the machine to find the fewest >number of moves of a checkmate? >The computer finds a "Mate in 10" for example in less time than say a "Mate in >5" later on. (The Mate in 5 would be the fewest number of moves.) >Obviously the title's variables are: >X is a positive integer and (Y is a positive integer AND X>Y). As long as the goal is to win the game, any forced mate will yield the same result: 1-0. Therefore this is normally not considered a problem at all. >Which programs have this very minor problem? I do not know of any chess playing program that guarantees shortest mates, at least not in normal playing mode. A special mate finder mode may well do that, but it is not suited for the purpose of winning a game, since finding the shortest mate, not just _some_ forced mate, can be very expensive. Within the last week you can find several examples (e.g. from Leonid), where Chest (which is fast in proving shortest mates) is by far slower than playing programs finding some (longer) forced mate. >Jonathan (94th message) A related topic may be to make progress, after a mate has been found. If the program still accepts any forced mate, and does not continue to search for a mate less deep than the last announcement, it might continue to announce a mate in 3 over and over, never making any progress. Still, it need not be the shortest mate, any less deep (forced) mate will do. Heiner
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.