On 21 Jul 99, at 21:46, Lucas Wiman wrote:

> (Off topic, when I computed that value in Landon's Calc program, my 
computer
> PII/233 started making a weird humming noise.  The noise stopped 
when the
> calculation finished.  Very strange indeed...)

Perhaps it was humming because it was happy ;-)

Coincidentally, on 21 Jul 99, at 19:46, burlington john wrote:

>   I run mprime (or prime95 ntprime) on my laptop.
> 
>   When I start the programm on my notebook it makes a strange mechanical
>   pulsing sound like an old damaged wheel or like an ill cricket.
>   The sound doesn't come out of the speakers and is not the fan
>   from the cpu.

Very recently I acquired a new Toshiba Satellite CDS4070 (Celeron 
Mobile 366 MHz), this is the first system I've had with Win 98 on it. 
(Actually I could have run Win 95 instead but decided to take the 
plunge) The power management is different to the other systems I've 
had - you can choose to either slow the CPU clock when it starts to 
get warm, or start the case fan first (then slow the CPU clock if 
it's still getting too hot). Obviously the sensible thing to do is to 
let the fan run if the system is on AC power, but it might make sense 
to let the CPU slow down if the system is running on its battery. Win 
98 lets you configure the behaviour in these states seperately.

The relevant fact about this is that the case fan is an axial fan set 
into the "back panel" of the case, which is only about an inch high. 
The fan is therefore quite small & needs to run quite fast to shift 
any significant volume of air. The "jet noise" is unexpectedly loud - 
certainly louder than a typical desktop system, but nowhere near the 
volume of my Alpha workstation. And it's _definitely_ fan, the HDD is 
almost silent, and in any case the jet of warm air can be felt.

At "room temperature" (68F-75F, 20C-24C) the system is silent when 
idle, but the fan starts about 2 secs after starting Prime95, 
stopping a few seconds after Prime95 is stopped. I think the reason 
is that the FPU is normally inactive, when the FPU turns on because 
it's needed the CPU needs assistance with cooling.

Neither John nor Lucas indicates which OS they're running, but I 
think most modern systems (even desktops) running an ACPI power 
management aware OS may behave in a similar fashion.

Note that, even if you have fans running continuously, starting 
Prime95 (or any other program which drives the FPU hard) causes a CPU 
temperature rise of a few degrees, presumably due to the extra power 
draw when the FPU is active. Though on desktop systems with large 
heavy heat sinks, the temperature change is slow, i.e. of the order 
of minutes rather than seconds.

>   When I press some areas on my notebook the sound disappers for a
>   few seconds but then it comes back.

Definitely sounds like fan to me. Maybe a loose wire or something is 
just touching the back of the fan blades? Distorting the case a bit 
by pressing on it moves the obstruction just far enough to be out of 
contact.
> 
>   The funniest thing is when I start other cpu stressing programms
>   while (m)prime is running the sound is changing. Each programm
>   plus (mp)prime changes the sound in a special way so I can recognize
>   exactly on the noise which programms are currently runnig.

Maybe the system runs the fan at different speeds depending on the 
heat output. Usually when you run something else as well as Prime95 
(or mprime - they're really the same program but packaged for 
different operating systems) the FPU is less busy than it would be if 
Prime95 was running on its own. So the fan might run slower, causing 
a different noise.

The only other thing I can think of is that a piece of wiring - 
probably in the display - or a lump of foil EMF shielding resonating 
in sympathy with something driven by software. I don't think this is 
very likely, but such things are known. If this is what the problem 
is, then turning off the display whilst Prime95 is running might fix 
it, or the problem might not exist if Prime95 was running on a 
different exponent with a different FFT size.
> 
>   I'm goning to get crazy listening that noise all the time.

Well, you could try configuring slow CPU speed instead of starting 
the fan, if that's what the trouble is. Or run factoring assignments 
instead - those don't use the FPU as much. Otherwise, either drop 
Prime95 8-( or go crazy, like the rest of us 8-)


Regards
Brian Beesley
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ      -- http://www.tasam.com/~lrwiman/FAQ-mers

Reply via email to