Author: blass uri
Date: 19:46:01 12/27/98
Go up one level in this thread
On December 27, 1998 at 09:03:25, Hans Christian Lykke wrote: >On December 26, 1998 at 20:19:18, blass uri wrote: > >> >>I read that the result of Junior5 against fritz3 was 17.5:2.5 and not 30.5:9.5 >> >>Is there a mistake in the result of Junior5 against Fritz3 or maybe there were 2 >>testers who did 20 games Junior5-Fritz3? >> >>Uri > >Both Thoralf Karlsson and I played J5-F3 I do not understand it. I told thoralf Karlsson that I think that programs without learning feature should play the same number of games against programs that can learn Thoralf replied me: Our ambition is to play the same number of games against the same opponents for all new programs. That is 40 games against other programs on P200 MMX and 20 games against P90-programs. Unfortunately we don't always manage to reach this goal. If this is the case then what is the reason for playing 40 games with Junior5 against Fritz3(P90) I can understand if they play less then 20 games against a P90-program because they have not enough time or volunteers to play but what is the reason for playing more than 20 games against a p90-program? Fritz5 also played many games against Fritz3(P90) I want to know if these games were also played by different testers or by the same tester The result can be different because if different testers test the same program then it cannot learn from one match of 20 games about the other match. Uri
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.