Author: stuart taylor
Date: 08:54:33 01/03/06
Go up one level in this thread
On January 02, 2006 at 04:18:17, Vasik Rajlich wrote: >On January 02, 2006 at 00:26:52, James T. Walker wrote: > >>Hello Tom, >>He doesn't seem too concerned about blitz performance but he has produced a >>great engine anyway. >>I played the 3 Rybkas vs 3 different opponents. The opponents all used their >>own books. The 3 Rybkas used Fritz8.ctg. Each Rybka played 120 games vs each >>opponent (360 games total). Ponder was on and each engine had it's own CPU on a >>dual core AMD. Here is what happened. >> >>Vs Fritz 8 >>Rybka Beta 66.5-53.5 >>Rybka Preview 73.0-47.0 (exceptional result) >>Rybka Beta 7 68.0-52.0 >> >>Vs Fruit 2.2.1 >>Rybka Beta 71.5-48.5 >>Rybka Preview 66.5-54.5 >>Rybka Beta 7 63.0-57.0 >> >>Vs TogaII 1.1 >>Rybka Beta 78.5-41.5 >>Rybka Preview 68.5-51.5 >>Rybka Beta 7 68.5-51.5 >> >>Totals >>Rybka Beta 216.5/360 >>Rybka Preview 207.0/360 >>Rybka Beta 7 199.5/360 >> >>You can see by the totals its a steady decline but probably within the error >>bars. Still a bit worrysome for a blitz only fan. :-) >>(sorry for the extra messages below) > >I will do what I can do make Rybka a good blitzer, without messing up my search >too much. Most people use their engines to analyze, and don't have the patience >to wait 2 minutes on each position. > >Vas For deep analysis, I'm sure people like to wait longer than 2 minutes, when they can see its opinion changing every few seconds. Only, if they want a spot check or spot opinion, many times one after another, then they would like to see something intelligent within about one minute. So you are saying that without blitz maximization, it would get a better evaluation in one minute, than if it were maximized for blitz? S.Taylor
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.