Author: Robert Hyatt
Date: 19:57:15 02/01/01
Go up one level in this thread
On February 01, 2001 at 20:17:34, Robert Pawlak wrote: >Hello all, > >I was wondering what kind of performacne increase one could expect (if any) when >running a chess engine that is _not_ designed for multiple processors on a >multiprocessor machine. > >For example, say I have a performance of X on a single processor machine. What >performance can I expect with two processors? I can see where this might >approach 2X for a program designed for dual operation, but what about one that >is not? Let's assume also that I am running win2K. > >The reason I ask is that I've looked at dual machines running Matlab 5.x, which >I've been told is not compiled for dual processors. And when you look at the >processor utilization under NT, it shows the same utilization for both >processors. > >Thanks! >Bob The only thing you will gain is the ability to run more than one program without having one interfere with the other. A single program will be _no_ faster if it is not designed with parallel programming in mind...
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.