Author: Jorge Pichard
Date: 13:30:24 01/28/04
Go up one level in this thread
On January 28, 2004 at 15:45:12, Kurt Utzinger wrote: >On January 28, 2004 at 15:40:24, Jouni Uski wrote: > >>I played 100 games with own books in P4 2,4 level =60/5. Good old Fritz8(.0.0.8) >>didn't respect double world champion but won this match! >> >>1 Fritz 8 >>½½0½1½0000000½1½00101½010½10½110½11011111111000110½01½11½½½1½10½100½½101110½1½½½0½101½001011½½001010 >>51.5/100 >>2 Shredder 8 >>½½1½0½1111111½0½11010½101½01½001½00100000000111001½10½00½½½0½01½011½½010001½0½½½1½010½110100½½110101 >>48.5/100 >> >>Note, that there is one 8 and one 7 wins in row - stunning, when engines are so >>similar in strength. >> >>Jouni > > Hi Jouni > An excellent example to demonstrate that 10 or 20 games > do statistically prove nothing. > Kurt 100 games is better than 20 or 40 games, but they are so close that you probably need 500 games to see a bigger margin and probably after 200 games Shredder will be leading and after 300 games Fritz8 will be on the lead and between 400-500 anything can happen :-) PS: 20 games might not be sufficient but at least it gives you a clue, for instance if Fritz 7 is leading El Chinito 3.25 by a big margin after 20 games there is no need to continue; and if they are tied at least it tells you that they are close in rating, but again after 100 games the prediction is much better than with 20 or 40 games :-) Jorge
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.