Author: Robert Hyatt
Date: 10:46:59 12/17/02
Go up one level in this thread
On December 17, 2002 at 12:36:52, Vincent Diepeveen wrote: >On December 17, 2002 at 12:10:34, Gian-Carlo Pascutto wrote: > >>On December 17, 2002 at 12:04:44, Matt Taylor wrote: >> >>>On December 17, 2002 at 11:40:37, Gian-Carlo Pascutto wrote: >>> >>>>On December 17, 2002 at 10:31:02, Robert Hyatt wrote: >>>> >>>>>It doesn't necessarily get slower. And once you move to windows .new, or >>>>>the new linux kernel gets fixed, this won't happen at all as both of these >>>>>systems will understand that two threads need to run on two physical processors >>>>>rather than on two logical processors on the same physical processor... >>>>> >>>>>what is your point? This is a sudden revellation to you? :) >>>> >>>>Yes. The thing is, Microsoft is _specifically_ advertising Windows XP >>>>(which was used for this test) as 'HyperThreading optimized'. >>>> >>>>This test shows that that should be taken with a few mountains >>>>of salt, considering the OS cannot even schedule correctly! >>>> >>>>This means that _right now_, there are severe issues with HT >>>>on a dual system. And they won't be resolved until a next generation >>>>of OS comes out. >>>> >>>>-- >>>>GCP >>> >>>Windows .net Server = Windows XP Server >> >>The tests were run on Windows XP Professional. Is the XP Server >>scheduler smarter? >> >>-- >>GCP > >in general 'server' is hell slower for computerchess as it is doing anything >first before executing the engine processes :) IN general you can't tell the difference. I have benched NT workstation and NT server, and could find no performance difference at all which is not a surprise. Unless you fire up a web server application and other such stuff, which would kill any machine anyway...
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.