Go Back > Great Internet Mersenne Prime Search > Hardware

Thread Tools
Old 2006-10-25, 00:26   #1
Oct 2006

3 Posts
Default P4 dual core problem

I have a P4 D805 processor running at the normal 2.666GHz and get iteration times of 0.092sec if running only a single instance of prime95 (50% CPU utilization) but if I try running two instances of prime95 (with affinities set to different cores and 100% CPU utilization) the iteration times more than double to 0.199sec. The CPU-Z program doesn't show any CPU throttling and I have varied fan speeds to change my CPU temperature from 47C to 67C (monitored by MBM5) with no effect on single or double instance iteration times.
Does anyone have a similar experience or an explaination as to what is happening?
tholloway is offline   Reply With Quote
Old 2006-10-25, 06:54   #2
S485122's Avatar
Sep 2006
Brussels, Belgium

175310 Posts

I had a D830 and have a D945. With both processors I had the same benchmark results running one instance of prime95 or two (one on each core obviously.)

Is it LL iteration times you speak of? If it is P-1 stage two or torture tests, you might not have enough memory and the OS would start swapping.

D8XX processors are quite sturdy, but I would try to keep the temperature under 65°C: thermal specifications from Intel is 64.1°C.
S485122 is offline   Reply With Quote
Old 2006-10-25, 12:21   #3
rx7350's Avatar
Feb 2006

14410 Posts

Check to make sure cpu affinity is set to '0' for one core, and '1' for the other core.
rx7350 is offline   Reply With Quote
Old 2006-10-25, 13:44   #4
Cruelty's Avatar
May 2005

22·11·37 Posts

Originally Posted by rx7350 View Post
Check to make sure cpu affinity is set to '0' for one core, and '1' for the other core.
Or "Let program run on any CPU", although that will cause a performance hit on P4 platform.
Cruelty is offline   Reply With Quote
Old 2006-10-25, 16:29   #5
Oct 2006

316 Posts
Default More Info on problem

Yes it is the LL iterations and the memory usage of each instance is only ~23mb. Also, I do have the affinities set to "0" on one and "1" on the other instance but it makes no difference which I add as the 2nd instance so both cores are performing the same with one instance.
One other piece of info is that I have tried it both on XP-pro and Linux-64 Fedora and get very similar results. When using Linux I get a CPU throttling warning only if I try overclocking to ~3.1MHz and add the second instance. That's why I'm not overclocking when trying to run 2 instances.
tholloway is offline   Reply With Quote
Old 2006-10-28, 20:16   #6
Oct 2006

3 Posts
Default Update - memory speed makes a big difference

I just swapped out my 266MHz DDR memory with a 333MHz DDR memory from another computer and this made a significant difference in iteration times. Single instance times went from 0.092sec to 0.085sec and double instance times went from 0.199sec to 0.131sec. It looks like memory speed makes a significant impact on interation speed, particularly with dual processors. I'm now going to look into upgrading to 533MHz DDR2 memory if I can find a good price.

Has anyone else noticed a difference with different memory speeds??
tholloway is offline   Reply With Quote

Thread Tools

Similar Threads
Thread Thread Starter Forum Replies Last Post
How to retire one core in a dual-core CPU? Rodrigo PrimeNet 4 2011-07-30 14:43
Dual Core to Quad Core Upgrade Rodrigo Hardware 6 2010-11-29 18:48
Dual Core P95 64Bit P4 Equivalent problem g0ods Software 9 2009-09-15 14:12
Importance of dual channel memory for dual core processors patrik Hardware 3 2007-01-07 09:26
Intel e6600 Dual Core Problem - How to use both cores with Prime95? Shoallakeboy Hardware 2 2006-11-06 17:55

All times are UTC. The time now is 17:08.

Sun Nov 28 17:08:30 UTC 2021 up 128 days, 11:37, 0 users, load averages: 2.06, 1.52, 1.38

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.