Author: Peter Berger
Date: 03:35:50 08/30/04
Go up one level in this thread
On August 29, 2004 at 17:36:18, Stuart Cracraft wrote: >On August 29, 2004 at 15:15:39, Peter Berger wrote: > >>Movei played on my Athlon 64 3200+ ( 2.0 GHz) notebook, Windows XP, 64 MB Hash >> >>MChess 7 played on PIV 2.2 GHz, Windows 98SE DOS mode, 60 MB hash. >> >>Time control was game in one hour. >> >>A race to 6 wins, similar to old-time human worldchampion matches. Movei won >>with 6-5 with zero draws, congratulations. >> >>Interesting games IMHO. To some extent they feel like human-computer games when >>MChess often looks like the human. But Movei has made progress too when it is >>about knowledge. In some games it seemed to show clearly improved understanding >>of king safety compaired to older versions. > >Any reason why you didn't disable thinking on the opponent's time and run >the match entirely on one box? Yes, there are a few. a.) MChess can only use a hashtable of decent size if the system is run in DOS-mode - so no multi-tasking possible. b.) I prefer matches with pondering enabled in general. c.) In a previous match at game/30 Movei got slaughtered by MChess 7, so I didn't expect the hardware advantage in this match to be too serious: http://chessprogramming.org/cccsearch/ccc.php?art_id=353377 Actually one conclusion I drew from the three matches between Movei and MChess is that choice of openings is extremely important, partly because of MChess's unusual way of learning. I have no idea at all whether Movei or MChess is stronger ( even if it is too few games anyway). Peter
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.