Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Are Conventional Chess Engines Almost "Peaked Out"?

Author: Bob Durrett

Date: 12:19:45 11/07/02

Go up one level in this thread


On November 07, 2002 at 14:12:17, Joshua Lee wrote:

>Fritz Going back to a3 after 4 hours and 20 ply pretty much proves that there is
>still a bit furthur to go in strength of programs. On some moves it would take a
>good 70 minutes to play a correct move just for starters and in the instance of
>the missed draw in game 2 Rc8 is 41ply deep ....let me put it into perspective
>Fritz on my 1Ghz will search 500Kn/s to look at 41 ply it would need hundreds of
>years or if i had a 100+ Petahertz cpu
>Just to turn one loss into a draw. Ofcourse if we had that technology the game
>would be totally different earlier and maybe a win for deep fritz.
>So atleast for the next mmm 40 years the answer is no.
>in 39 years we should have Petahertz processors ateast if moore's law stays the
>same

I fear that you are addressing a topic different from the one I posted. : (

Everybody knows that chess engines will perform better on better hardware.

Bob D.



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.