Author: Dann Corbit
Date: 16:59:33 11/26/01
Go up one level in this thread
On November 26, 2001 at 18:57:13, Russell Reagan wrote: >http://www.theregister.co.uk/content/3/23012.html > >25 GHz processors by 2005! They said 'maybe 20GHz' if I understood the article correctly. >What will this mean for chess programs? 25x faster than today. >Endgame tablebases? No change. That is greater disk that is needed. You could compose them faster, however. >Will the worlds top GM's still have a slight edge over the top computer >programs? At 25GHz, consider that a very fast program on a 1GHz chip does 1 million NPS. Consider 25x that, to get 25M NPS. That's 1/8 the sustained throughput of Deep Blue, and 1/40th peak. Kasparov gave Deep Blue a run for the money. Of course, the algorithms will be different etc. Already, chess programs are plenty strong enough for 99.999% of mankind. At some point, they will clearly be better. When is an interesting question, but I don't see any reliable way to extrapolate it. What if there is a revolution in human chess thinking?
This page took 0.01 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.