Author: Peter Berger
Date: 14:15:48 09/28/05
Go up one level in this thread
On September 28, 2005 at 07:41:31, Heinz van Kempen wrote: >On September 28, 2005 at 07:26:50, Jonas Cohonas wrote: > >>So Fritz 9 has this score: >> >>6 Toga II 1.0 ( 2727 ) 10 + 0 = 1 - 9 5.0 % 2367 >> >>So far against Toga?? >> >>I personally find it hard to believe that something isn't wrong here (in the >>set-up or the like). >> >>Maybe you should re-run the games against Toga? >> >>Regards >>Jonas > >Hi Jonas, > >please check the games. We all did this. They are available for download with >comments (depth search, NPS). All games look fine and we see no need to repeat >them. > >http://www.husvankempen.de/nunn/downloads.htm > >It looks like a statistical anomaly and we are having many of them. This is >normal. The match against Toga continues. > >Best Regards >Heinz There are people who win the jackpot in the lottery every week, so it clearly happens. But are you aware *how* unlikely this result is assuming independent events (the CEGT setup should guarantee this) and somehow similar strength of engines, which can be assumed here? Especially if you are really seeing *many* of these anomalies as you say and this is normal in your matches, there *should* be something wrong here. I hesitated to post this - I am only interested in the mathematical aspect of it ( although my knowledge is a bit rusty). Peter
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.