climateprediction.net home page
About my processor @ 100%

About my processor @ 100%

Questions and Answers : Preferences : About my processor @ 100%
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile old_user68500

Send message
Joined: 2 Apr 05
Posts: 1
Credit: 0
RAC: 0
Message 11574 - Posted: 2 Apr 2005, 14:53:00 UTC

I'm not sure if I am reading it wrong or if it bad for my computer but...when I bring up my Windows Task Manager it says that my CPU is running at 100%. Will this hurt my processor over time?
If so, how could I go about adjusting it?

Thanks for you time and any help,
Rick
ID: 11574 · Report as offensive     Reply Quote
crandles
Volunteer moderator

Send message
Joined: 16 Oct 04
Posts: 692
Credit: 277,679
RAC: 0
Message 11575 - Posted: 2 Apr 2005, 15:15:18 UTC
Last modified: 2 Apr 2005, 15:21:56 UTC

Hi Rick and welcome.

There are a few threads like <a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/forum_thread.php?id=1869"> this one</a>

search for threadmaster for a few other threads. Threadmaster is a program that should allow you to adjust the maximum % processor usage.

The general reaction seems to be overheat or spikes in current can kill a processor. If you have adequate cooling system and clean it out once in a while the heat shouldn't be a problem. Spikes depend on your electricity supplier (and whether you run heavy machinery off same electricy supply and whether you have ups or other protection) and probably are not affected by running processor at 100%. More you have it running the more chance of encountering too much of a spike but this is a risk most of us take.

Processors probably do wear out other than by overheat and spikes but by the time they do, the processor is obsolete and you may be glad of the excuse to upgrade ;)

Running at 100% may well be risky for laptops as they are less likely to have adequate cooling. Some people still run laptops taking the precaution of lifting it up away from surface particularly at the back.
ID: 11575 · Report as offensive     Reply Quote
Profile old_user35334

Send message
Joined: 8 Jan 05
Posts: 13
Credit: 115,849
RAC: 0
Message 12063 - Posted: 24 Apr 2005, 3:51:14 UTC

Generally speaking semi-conductors do decrease in life (or worsen in MTBF) when they are operated at higher temperatures for long periods of time. However I could not find any solid data on the intel public website to draw any conclusions. There was a comment I found stating a 250 year MTBF for hard defects but not enough data to use that number.

Most of the intel CPUs are specified to a max of 70C or thereabouts.

http://support.intel.com/support/processors/pentium4/sb/CS-007999.htm

When I am running CPDN at 100% CPU, my processor is about 50C. This rises to about 55C when the heatsink gets dirty due to accumulated dust. Personally I have not worried too much about this in terms of reliability as it is well under the chip specs.

You will see statements that life decreases by 50% for avery 10C rise in temp, but this does not mean much without the baseline data on MTBF at a particular temperature.

There have been a large body of people running Distributed computing projects over the last 5 years, but not much regarding studies on impact on CPU life that I could find.

I think that as long as your CPU is in the 45 to 55C range then you are in the same situation as the rest of us. Your motherboard may have a utility to monitor this. (My asus MB has a temp and fan speed monitor programme). Its more likely that another component in the PC will fail first.

I agree with Crandles that you want to keep an eye on the cooling that you use and that a laptop may be more risky. Also - make sure that you have a surge suppressor on your pc to help with electricity spikes.

ID: 12063 · Report as offensive     Reply Quote
Profile old_user35334

Send message
Joined: 8 Jan 05
Posts: 13
Credit: 115,849
RAC: 0
Message 12964 - Posted: 31 May 2005, 3:18:26 UTC

well i spoke too soon - my pc started resetting a few days after that post when under load.

took a while but after running memtest, prime95, reseating my RAM etc. I decided it must be the psu.

sure enough when i opened the PSU, there was a burnt patch where a resistor had blown in the PSU. was a generic 400W (LPK2-30) only 16 months old. I'd been running BOINC about 4 months when it died.

Bought an Antec truepower2 550W to replace. Works fine - PC runs about 3degrees cooler (47C CPU now at 100% load. It also has speed control for its fan and the case fan. I added a second case fan and everything seems quieter and cooler.

I would guess that the load of the continuous running was too much for the PSU. New one has MTBF of 80000 hours. Should expect ~10% failure rate a year. Hopefully I am running it cool as it is oversize.

ID: 12964 · Report as offensive     Reply Quote
Jean-David Beyer

Send message
Joined: 5 Aug 04
Posts: 1045
Credit: 16,506,818
RAC: 5,001
Message 18630 - Posted: 22 Dec 2005, 20:35:06 UTC - in response to Message 12964.  

well i spoke too soon - my pc started resetting a few days after that post when under load.

took a while but after running memtest, prime95, reseating my RAM etc. I decided it must be the psu.

sure enough when i opened the PSU, there was a burnt patch where a resistor had blown in the PSU. was a generic 400W (LPK2-30) only 16 months old. I\'d been running BOINC about 4 months when it died.

Bought an Antec truepower2 550W to replace. Works fine - PC runs about 3degrees cooler (47C CPU now at 100% load. It also has speed control for its fan and the case fan. I added a second case fan and everything seems quieter and cooler.

I would guess that the load of the continuous running was too much for the PSU. New one has MTBF of 80000 hours. Should expect ~10% failure rate a year. Hopefully I am running it cool as it is oversize.


I have a dual 3.06GHz Hyperthreaded Intel Xeon machine that I run 4 BOINC applications at a time (now that the BOINC client has been fixed to obey that requirement -- it still has other scheduling problems) 24/7/52. I can measure the temperatures of my processors and the box, as well as the processor fan speeds (thermistor controlled speed, but the thermistor is in a dumb place at the air intake to the fan instead of the exhaust from the processor wind tunnels). Anyhow, this results in all 4 processors (count \'em, 2) running at 99% to 100% load all the time.

My machine has been running (Linux) 26 days since the last reboot (and that only due to a power failure longer than my UPS can withstand), so it is safe to say that the processors have stabilized as to temperature. They are currently running at 54°C and 52.5°C, and the box inside is running at 41°C. It has been doing this for a couple of years now, so if there are overheating problems, they have not yet shown themselves. The processors get hotter in the summer, but not over 60°C because the inside temperature of the box goes up somewhat (no air conditioning here in New Jersey), though the processor fans speed up.

There is a 660 Watt Enermax EG851AX-VH(W)FM power supply in there to deal with 6 hard drives, the two 90 watt processors, and all the other stuff.



ID: 18630 · Report as offensive     Reply Quote
Profile old_user35334

Send message
Joined: 8 Jan 05
Posts: 13
Credit: 115,849
RAC: 0
Message 22442 - Posted: 28 Apr 2006, 2:02:53 UTC

here\'s an update - been running with the new PSU for 11 months

no probs so far

case temp 26C
cpu temp 51C


psu fan 1340 rpm
cpu fan 2790 rpm
+ fixed fan

have to vacuum the cpu heatsink about every 4 months - or temp goes up from 47 C to 55C due to dust clogging the fins (as you can see I\'m half way through the cycle)

think that having an over size name brand psu helps a lot


ID: 22442 · Report as offensive     Reply Quote
racinjimy

Send message
Joined: 19 Apr 05
Posts: 53
Credit: 6,325,436
RAC: 0
Message 22682 - Posted: 10 May 2006, 1:39:20 UTC - in response to Message 22442.  



I think that having an over size name brand psu helps a lot



yes it does, cheap PSU\'s are the devil..............
ID: 22682 · Report as offensive     Reply Quote

Questions and Answers : Preferences : About my processor @ 100%

©2024 climateprediction.net