Author: José Carlos
Date: 01:47:51 08/17/01
Go up one level in this thread
On August 16, 2001 at 09:41:18, Thomas Mayer wrote: >Well, in my opinion this depends to several things: > >Maybe we will have a future, where chess programs play in tourneys and >sponsorship is there to make it interesting for the GMs to play with to programs >seriously... > >THEN i believe in this: At the moment programs can achieve ratings of 2700 and >higher in those tournaments with humans... when the GMs start to play serious >with the programs and start to study them because they need to, i think there is >no doubt that the ratings will lower to say 2500 or so... but well, the >chess-programmer do not sleep - so in 2010 we are maybe back at 2700... - I am >TOTALLY sure, that nor in 2010 neither in 2030 a chess program hit 3000 in >tournament time control... > >Greets, Thomas Not with actual hardware, probably, but hardware gets faster and faster. I expect in 2010 hardware to be so fast that programs will crash _most_ GM's. In 2030 programs _definetly_ will be beyond 3000. But maybe we could invent a new game (derived from chess) where branching factor will be too big to compute even with super-fast machines, and where human inference capabilities can still rule for some more years... José C.
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.