Author: Bob Durrett
Date: 14:46:19 04/07/04
Go up one level in this thread
On April 07, 2004 at 17:41:23, Bob Durrett wrote: >On April 07, 2004 at 16:25:19, Alan Grotier wrote: > >>> >>>No matter what kind of computer and software you use you are likely to be >>>disappointed with the amount of time it takes to produce believable analysis >>>results. These are disappointing times in the chess-playing program world and >>>it may only be after we are all dead and gone before the programs can produce >>>instantaneous GM-level analyses. >>> >>>It takes quite a bit of innovativeness on the part of the user to obtain good >>>analyses since the current crop of chess-playing programs do not have infinite >>>horizons and the programmers have not yet found out how to make their programs >>>play good strategic chess. >>> >>>Just make the best of what you have and learn to live with it. : ( >>> >>>Bob D. >> >> Bob, >> An interesting comment.So if you had to put a qualifier >> to the analysis obtained from the present day engines >> where would you rate them? 1500-1800,1800-2200 range? <snip> On a very fast computer with adequate RAM [for hashing], at least 2600. However, that's just for finding the average MOVE in an average tactical position. The rating to assign to "ANALYSES" depends on the methods and skills of the user and the type of position(s) being analyzed. Kasparov could probably produce analyses using Fritz which could be called "3300 analyses." Bob D.
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.