Author: Dan Kiski
Date: 07:25:44 12/16/98
Go up one level in this thread
On December 16, 1998 at 09:42:33, Harald Faber wrote: >On December 16, 1998 at 09:11:49, Dan Kiski wrote: > > >>I hate to say it but a test of 20 or so games at one set time control don't >>prove anything regardless of games with other programs. > >Are you talking about my idea of testing CM against the other top programs? No, I was refering to cm6555 v cm cummings. > >I know that 20 games don't say much, they almost say nothing. If you want to >have nearly what you call PROOF then you need to have at least 500 games EACH >MATCH. > >What my intention is to get a more certain impression of CM's strength. CM6k >won't be tested by SSDF and has no autoplayer so it is a farce to play manually >even 100 games. That justifies, at least for me, this small tournament with >reduced time controls which indeed is bigger than some other Xmas tourneys or >so. >And when we see after the 240 games that CM6k/CM6555 wins every match with 13-7 >*I* would go so far and say that CM6k/CM6555 IS STRONGER than the rest. You >don't need to but I will/would.
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.