Author: Michael Henderson
Date: 16:51:27 01/12/05
Go up one level in this thread
On January 12, 2005 at 19:37:29, Steve Maughan wrote: >Dann, > >>Things that seem impossible quickly become possible. > >I recon about 300 years before a computer will solve chess. This assumes > >1) 10^120 possible positions >2) Alpha-beta cutting this down to 10^60 sensible positions >3) Moore's law holding ad-infinitum i.e. doubling in speed every 18 months >4) Assume that we're happy to let a computer ponder for one whole year to solve >the game (i.e. 31.5 million secs) >5) Today the fastest quad PC can analyze 5 million pos / sec today I don't mean to be rude, but alpha-beta is used to find the exact score and path for one position, e.g. solving the starting position, which fits in your model and analysis. Some other nodes will have bounded scores (and positions not solved as a result). So I guess it depends on how one interprets "solve chess". Michael
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.