Author: Rolf Tueschen
Date: 10:09:03 12/08/02
Go up one level in this thread
On December 07, 2002 at 02:04:49, Ingo Althofer wrote: >On December 06, 2002 at 17:36:24, martin fierz wrote: > >>... i >>can assure you that LOTS of CC players only turn on their engine and enter the >>opponents move when they get it, and really don't find their own moves. instead, >>they go with the recommendation of the engine, >> >>or, if several moves are close in >>value, they may choose the second or third-best. these are the guys who don't >>find their moves themselves. > >Hello Martin, > >I would like to distinguish severely between > >* those players who only use the (single) best proposal of their program > >and > >* those who let run the program in 2-best or 3-best mode - finally selecting >from the two or three candidate moves shown. > >Such a multiple-choice mode can lead to a trememdous increase in >playing strength. Years ago I made such multiple choice experiments >in "over the board" chess, calling the system "Double-Fritz plus Boss": >Fritz4 computed its two best candidate moves and me (=the human boss) >always had the final choice amongst them. In two events in 1996 (the seven >round open tournament in Apolda, and a 4.5:3.5-win in an 8-games match >against GM Gennadi Timoscenko) "Double-Fritz plus Boss" scored around 200 >rating points better than Fritz4's own (SSDF) rating. Mathematically this is a completely false conclusion for the statistics of your not so well documented and controlled "experiments", sorry. Rolf Tueschen > >Using computer proposals in multiple choice mode can be >a trememdous help (not only in chess). > >Ingo Althofer.
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.