Author: Thorsten Czub
Date: 07:23:24 06/06/03
Go up one level in this thread
On June 06, 2003 at 06:52:14, Sergei S. Markoff wrote: >Using of Botvinik ideas, I think, is only possible from very critical point of >view. The main problem that Botvinnik did a completely unproved conclusion that >computers must use in chess methods that is similar to human ones. But the >architecture of our electronic machines is completely different, very unlike >human brain. Botvinnik spent a lot of his own time and time of his adepts trying >to implement human methods of chess-playing. But this goal needs not only near >infinite programming skills, but a fantastic methodology of human knowledge >extraction. Our conceptions of our knowledge is often very methaphysically. It's >only the illusions. > >When reading Botvinnik your can found that modern computer chess is under a >great influence of his ideas. But this is only influence, not direct using of >any concrete ideas. The Botvinnik idea is to build search on knowledge. Not >divide search and evaluation. It's a very mechinistic division. Botvinnik's >program must be very intellegent, it must discover some lines much deeper that >other ones. Should we try to reproduce how humans play chess ? or should we better use computer methods too create a strong chess program ? IMO we should try to make a chess program that plays like humans do. there have been efforts. Take e.g. Mephisto III by Nitsche and Henne. (this program got a world-championship title 1984) Take Botwinniks Pioneer. Take Chess System Tal. (It beated Genius !)
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.