Author: Uri Blass
Date: 10:25:11 12/24/05
Go up one level in this thread
On December 24, 2005 at 12:55:56, Gabor Szots wrote: >On December 24, 2005 at 11:28:11, Uri Blass wrote: > >>On December 24, 2005 at 11:07:38, Albert Silver wrote: >> >>>On December 24, 2005 at 10:00:44, Nolan Denson wrote: >>> >>>>While playing on playchess, I get my best results when playing with slightly >>>>Tactical. At one point it played about 43 games in a row with no loses. The >>>>rating of the other accounts were from 2400-2900. My rating got as high as 2860 >>>>using slightly tactical. I am using a 3.73 EE processor. >>> >>>In my personal tests, ST had the clearly worst results compared to the other >>>settings. >>> >>> Albert >> >>Note that it is logical to expect ST not to be the worst because if ST is worse >>than SP then I expect also VT to be worse than ST. >> >>It is the same as changing the value of the queen. >> >>If you have 4 possible values 9.5 9.7 9.9 10.1 then it is not logical that 9.9 >>is worse both relative to 9.7 and relative to 10.1 >> >>If the 9.9 is worse than 9.7 then it means that the value of queen is smaller >>than 9.9 and in this case 9.9 is certainly worse than 10.1 >> >>Uri >>Uri > >I don't understand this argument. A lot of functions have local minimums or >maximums. > >Gábor Yes but local minimum unlike local maximum is not logical when trying to optimize a parameter for playing strength. 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.