Author: Chris Taylor
Date: 02:29:13 07/09/99
Go up one level in this thread
On July 08, 1999 at 23:22:55, Christophe Theron wrote: >On July 08, 1999 at 17:58:37, Chris Taylor wrote: > >>One person said something along the lines of, average time, as set on comp v >>comp, is not a true reflection of results? Like it is not a proper way to go on >>in testing machines, against each other? WHY? > >It was me, and Jim Walker in the other post in reply to yours has very well >explained why using average time control is not serious. > >To be short: the only way to fairly test programs against each other is to use >real time controls (I mean X moves in Y minutes, or all the game in Y minutes). > >Average time control is a toy and should NOT be relied on for serious testing. > > > Christophe Thank You for putting me onto a real lemon! I did not relaise average move was so bad, and I can't remember reading about it on these pages. I have got my two computers to play via a cable, so any testing will be as you suggested. THANKS. Chris Taylor
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.