Author: adam wilks
Date: 13:24:56 09/28/05
Go up one level in this thread
Its logical to think that computer programs will get stronger gradually as time progresses. I’ve done this "deep blue test" of 1996/7 numerous times on a desktop pc with shredder 8/9,Hydra, Fritz and Chessmaster 10k settings. If tested on default settings most programs will chose stronger lines and not consider alot of the Deep Blue moves as "search techniques" have improved. This is not to say Deep Blue didn’t have a search feature. It did. On the contrary, Deep Blue was predominantly a "brute force" approach to chess. Furthermore, this "brute force" approach has become less prominent as programmers adopt highly sophisticated/selective search techniques. Time acts as a "technological hardware cut-off" in computer chess when more knowledge can be implemented in to chips like Hydras FPGA chips or via software as compilers are alot faster, debugging easier than 10 yrs ago as there’s more knowledge to refer to. I think pda testing is the future as its portable and like carrying a grandmaster in your pocket. Take Pocket Fritz`s draw with Michael Adams a few yrs ago in Mainz. If a Pocket Pc can draw a grandmaster then what’s the point of Hydra playing Adams? Lastly, Hiarcs 9.50a on a Tungsten T3 has beaten two high 2600 grandmasters. People finally recognize Machines annihilation of man on the board, will it be 10 yrs when the average pocket pc will be two pawns up by move 30? Does that mean by the time im 22 Hiarcs will be one pawn up by move 15? Man can only win by: 1. Computer programming was banned this month so "humans can try to catch up with their own creations". (Highly unlikely as its human instinct to go where no one else has gone). 2. Or, everyone started reading Ernest F Pecci`s book "A Psychiatrist matches wits with Fritz" and learning “quiet” moves. I wish that within the next 50 yrs there will be a "normal" group of grandmasters like today but also another "normal" group whom specialisze in beating chess programs/dedicated computers whom recieve equal credability and accolades as chess GMs` whatever that may currently be.
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.