Author: Robert Hyatt
Date: 12:19:33 02/25/02
Go up one level in this thread
On February 25, 2002 at 02:07:25, Uri Blass wrote: >On February 25, 2002 at 01:49:32, Gian-Carlo Pascutto wrote: > >>On February 25, 2002 at 00:02:50, Robert Hyatt wrote: >> >>>I don't "believe" it is impossible to find it from the Kf1 position, I >>>_know_ it is impossible for a program to find this perpetual. >> >>Okay, but then please explain why Fritz, with a 21 ply nominal >>search & a 53 ply of extensions returns _exactly_ that: a draw >>score. >> >>-- >>GCP > >I hvae some possible explanations: > >1)maybe you do not need full 60 plies. >I saw no proof that you need 60 plies to see the draw. > >It is better to post the line of 60 plies so people can find if >this is really the only line for black to force repetition and that black has no >alternative to force repetition with a shorter line. > >2)hash tables may help and program can see 60 plies forward when the maximal >depth is 53 plies thanks to hash tables. > >Uri (3) the program just can't see far enough to know what is going on, and it is making decisions about who will try for a draw and who will fight the draw, based on incorrect information. This should not be hard to understand. It is something I cover in a basic AI course here... Mates are forced and easy to prove. Draws are optional and not nearly as easy to prove since the side with the "draw in hand" can always choose to vary if it spots a way to do even better...
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.