Author: Vincent Diepeveen
Date: 10:45:55 12/22/04
Go up one level in this thread
On December 22, 2004 at 09:04:55, Francesco Di Tolla wrote: >>I do not see how they can have negative effect if they are probed only in the >>right conditions. > >well it could be that on average today the programs waste time to read data >from disk and gain little from it. >Of course it depends on when and how you read data. I have to agree with Uri here. In middlegame usually i get 1 or 2 probes to harddisk. Because i already use 1 processor more than the average program, especially The King. This extra processor that DIEP has, 'wastes' say 10-20 milliseconds to get some data. This is then used by the searching processor. Then i still have 1.99x more cpu power to my avail then The King. I do know that Johan says that probing is a waste of system time. It is however also the utmost nonsense. EGTB's are simply very important to have. And of course at blitz it is more important than in slow games, but the difference between blitz and slow games (i usually test at like 90 30 level at home) is not going to change the hard fact that in the endgame they are worth 600 rating points and overall for top programs they are worth at least 150 rating points. So saying that playing without egtb's is not making them play worse just means that the person who tested it has one of the parameters wrong. Not being able to use EGTB's is one of the reasons i'm not playing at a supercomputer with DIEP. I prefer to have EGTB's to my avail, it's a free 150 rating points. Vincent
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.