Author: Heinz van Kempen
Date: 13:37:02 09/30/05
Go up one level in this thread
On September 30, 2005 at 08:53:47, Thomas Logan wrote: >On September 30, 2005 at 08:38:13, Thomas Lagershausen wrote: > >>On September 30, 2005 at 08:18:08, Thomas Logan wrote: >> >>>On September 30, 2005 at 07:43:29, Thomas Lagershausen wrote: >>> >>>> >>>>http://www.husvankempen.de/nunn/rangliste.html >>>> >>>>Fruit WCCC`5 = Fruit 2.2 >>>> >>>>We have a new No. 1 in computerchess. >>>> >>>>TL >>> >>>Fruit may be the strongest engine in the world or it may not be >>> >>>In my view however this is not decided by a 40/40 nunn rating list >>> >>>Play them at classical time control >>> >>>let them each use their own book >>> >>>then you will have the strongest in the world >>> >>>Tom >> >>First,i am not interested in booktesting. I am interested in the strongest >>machine that give me good advices in positions that are not included in >>openingbooks, for example the recent games of the top 8 in argentinia. > >Hi Thomas > >What leads you to believe the nunn test suite is nuetral ? > >Can you dismiss the posibility one of the engines involved was tuned using this >while the other was not ? > >Also while you examine for analysis purposes can you not also considered others >examine for the strongest possible chess entity they can get on their computer > >To me both are valid > > >>Second, classical time control is a matter of your hardware.If your hardware is >not the fastest you can´t say that 40/40 games is not the same as the games you >play on your pc. > >Well I matched Toga II at 40/40 vs Shredder and Toga won > >then I matched them at classical time control Shredder won > >I believe I can say 40/40 is not the same as classical time control > >Dont get me wrong Fruit is very strong and it may have surpassed Shredder at its >best but lets find that out by putting the best Shredder foward > >personally to me a 40-40 nunn Shredder is not it > > >Tom Hi Tom, apart from the fact that currently and already for a few months very raraly Nunn or Noomen positions were used in CEGT you would need around one hundred machines to give the same statistically reliable data with tournament time control. I would doubt that we would get considerably different ratings. From time to time I also play matches 40/120 and rarely the moves differ after 3 minutes from those chosen after 1 minute. And when, you can still not be sure if only by centipawns, so just an alternative move. Best Regards Heinz
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.