mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Hardware (https://www.mersenneforum.org/forumdisplay.php?f=9)
-   -   Computer Fan Slows Down during Stress/Torture test (https://www.mersenneforum.org/showthread.php?t=2118)

mickylaren 2004-02-20 03:51

Computer Fan Slows Down during Stress/Torture test
 
Hi everyone...I'm new to the forum. I recently downloaded Prime95 after reading an article in Discover magazine about the discovery of the largest prime number so far. Anyway, I was fooling around with the options in Prime95, and I tried out the stress test for a few minutes. I noticed that while running the test, the noise my computer's fan makes (Dell Dimension 8200, 2.26GHZ, 512MB RDRAM) changes. It sounds like while the test is running the fan slows down (not significantly, but enough that I can hear the difference). Is this because the test uses more electricity? Is this a problem?

Thanks, everyone, for your help with this!

-Micky :grin:

TauCeti 2004-02-20 08:14

[QUOTE=mickylaren]It sounds like while the test is running the fan slows down (not significantly, but enough that I can hear the difference). Is this because the test uses more electricity? Is this a problem?
[/QUOTE]

I have experienced the same behaviour with my machines. Even with a high-quality PSU i can 'hear' prime95 kicking in ;)

But my guess is that the effect is only a few rpm if any at all. If you have some software installed to monitor the voltages (or use a multimeter), you should see a small drop.

Tau

LoKI.GuZ 2004-02-20 13:15

I also notice the same thing... On my machine i am *sure* it is due to a bad combination of fan/PSU. Also you have to consider the heavy impact prime95 has on the system: it's IMO the best stress program available. i see the temp on my Palomino 2000+ rise 10 degrees celsius as I use prime95...

frmky 2004-02-21 07:32

[QUOTE=mickylaren]I noticed that while running the test, the noise my computer's fan makes (Dell Dimension 8200, 2.26GHZ, 512MB RDRAM) changes. It sounds like while the test is running the fan slows down (not significantly, but enough that I can hear the difference). [/QUOTE]

Are you sure it's not the fan turning on or speeding up? For example, the cpu fan on my Opteron turns completely off when the cpu is idle. When I start something running, it kicks on after a few seconds and slowly ramps up to full speed over a couple of minutes. When I shut the process down, the fan slowly ramps down then turns off over a few minutes. I would be surprised if the newer Intel processors didn't do the same thing.

Greg

TauCeti 2004-02-21 10:21

[QUOTE=frmky]Are you sure it's not the fan turning on or speeding up?[/QUOTE]

If you fans speeds are controlled by temperatures and your system is under low load before you start prime95, your observation fits.

But if alls your system fans are already running at full speed (e.g. manually set to do so) and _then_ prime95 kicks in, the increased power draw often leads to a small decrease in valtage causing the effect on the fans.

adpowers 2004-02-24 00:38

Ah ha, but how many of you can hear little squeaks coming from your computer when Prime95 runs? My old PIII 450 used to squeak when running Prime95 on Windows 98. It doesn't do it with Mandrake for some reason. It was the weirdest thing, and it was definately related to Prime95. I never figured out what it was and I was the only one able to hear it.

adpowers

TauCeti 2004-02-24 09:41

[QUOTE=adpowers]Ah ha, but how many of you can hear little squeaks coming from your computer when Prime95 runs? My old PIII 450 used to squeak when running Prime95 on Windows 98. It doesn't do it with Mandrake for some reason. It was the weirdest thing, and it was definately related to Prime95. I never figured out what it was and I was the only one able to hear it.[/QUOTE]

No idea. Perhaps some harmonics generated by the onboard switched power supply? AFAIK Windows 98 uses larger time slices in the kernel scheduler compared to linux. Maybe that in combination with prime95 produces some oscillations with the onboard voltage controllers? Just guessing.

... reminds me of the good old times when i first played the BASIC-game "space invaders" on a CBM PET2001. That box had no sound output at all and debugging was no easy task. AFAIR it had a 6502 with 1MHZ.

You could use a normal AM-radio and put it right next to the computer. If you tuned the AM-radio to the right frequency, you could _hear_ what the computer was doing. Every combination of BASIC commands produced unique sounds. With some training you could actually hear what kind of basic commands where running at any given time.

Every program had it's own soundprint. Funny times. Long gone.

Mutu 2004-02-28 17:37

P4's take their juice from the 12V line (the same one which powers fans).

The fan quietening will be your 12V line dropping under the extra load, so long as it remains stable during the torture test then there's no reason to worry.


All times are UTC. The time now is 01:54.

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