Author: Uri Blass
Date: 09:29:10 12/06/03
Go up one level in this thread
On December 06, 2003 at 11:48:08, Mike Byrne wrote: >On December 06, 2003 at 11:42:19, Mike Byrne wrote: > >>On December 06, 2003 at 11:33:42, Joachim Rang wrote: >> >>>On December 06, 2003 at 11:22:15, Alex Szabo wrote: >>> >>>>My self-play experiment between crafty versions 19.6 and 17.14 resulted in the >>>>19.6 version being stronger by 11 (plus or minus 2) rating points. >>>> >>>>The experiment was for 20,000 games at a 20+1 time control. It was run on a >>>>2x2GHz pentium4 linux system with xboard 4.2.6. I can include other details if >>>>there is interest. >>> >>>20.000 Games at a 20+1 time control? How many month did your test last? It seems >>>to me impossible to play 20.000 games with Crafty 19.6 since this version is >>>only a few weeks available. >>> >>>regards Joachim >> >>I thought the same thing - but then I did the math and I felt it was possible. >>His 20 +1 must mean 20 seconds plus 1 second - not 20 minutes plus 1 minute. >>You could play 20,000 games of 20 seconds plus 1 second in about 500 hours. >>Just about 3 weeks , if played continously. > >I retract this post -- On second thought, I don't think it works. Why not? 20 seconds per game+1 second per move when game take 60 moves and the programs do not use all the time may mean 2.5 minutes per game. 2.5 minutes*20,000=50,000 minutes It is 833 hours and 20 minutes if my calculation is correct. When you have 4 dual that play at the same time then you get less than 210 hours that is less than 9 days. I have a problem with this way of testing because 4 dual that work on the same time may mean double games and it is better to do different tests at different dual. 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.