System temperature - Power Consumption

When I turn on my PC, my BIOS reports the following values:

System temperature = 39C / 102F
CPU1 temperature = 57C / 134F
CPU2 temperature = 63C / 145F

Vcore1 = 1.56V
+2.5V = 2.48V
+3.3V = 3.28V
+5V = 4.91V
+12V = 11.92V
-12V = -12.03V
-5V = - 4.94V
VBAT(V)= 3.45V
5VSB(V)= 4.82V

Are these temperatures OK? At what level are temperatures considered
"too high"?
I guess that the longer the PC remains on, the temperature could rise.
But what about the power consumption? Would the reported values change
as I use the PC? And are the values I quote above satisfactory? I
particularly worry about the
"-12V" value, which is -12.03 and therefore higher (in absolute terms)
than 12. If this value is indeed too high, what are the implications
of doing nothing? Alternatively, what could I do to reduce it?

I C wrote:
When I turn on my PC, my BIOS reports the following values:

System temperature = 39C / 102F
CPU1 temperature = 57C / 134F
CPU2 temperature = 63C / 145F

Vcore1 = 1.56V
+2.5V = 2.48V
+3.3V = 3.28V
+5V = 4.91V
+12V = 11.92V
-12V = -12.03V
-5V = - 4.94V
VBAT(V)= 3.45V
5VSB(V)= 4.82V


Are those CPU's, AMD or Intel? And how fast are both of those processors?

On 8 Jun 2004 15:31:48 -0700, (I C) wrote:

When I turn on my PC, my BIOS reports the following values:

System temperature = 39C / 102F
CPU1 temperature = 57C / 134F
CPU2 temperature = 63C / 145F

Vcore1 = 1.56V
+2.5V = 2.48V
+3.3V = 3.28V
+5V = 4.91V
+12V = 11.92V
-12V = -12.03V
-5V = - 4.94V
VBAT(V)= 3.45V
5VSB(V)= 4.82V

Are these temperatures OK? At what level are temperatures considered
"too high"?
I guess that the longer the PC remains on, the temperature could rise.
But what about the power consumption? Would the reported values change
as I use the PC? And are the values I quote above satisfactory? I
particularly worry about the
"-12V" value, which is -12.03 and therefore higher (in absolute terms)
than 12. If this value is indeed too high, what are the implications
of doing nothing? Alternatively, what could I do to reduce it?

All of your voltages look fine, assuming the CPU has a default voltage of
1.50-1.55... you made no mention of which CPU.

The CPU temp, at 63C upon power-on in bios, is borderline, most often low
enough but in some cases could affect system stability. At 63C there is
no significant risk of permanent damage. If your ambient (room) temps
are not expected to get any higher and the CPU is a lower-speed model in
it's particular family then 63C (presumably getting hotter under full load
conditions) may be stable, but it's fairly clear that either your heatsink
or case cooling is sub-par for your CPU. If CPU temp was a half-dozen
degrees higher I'd suggest it's _necessary_ to take further action to
reduce the temp.

You could take appropriate measures to alter chassis airflow (a good idea
since other components may also be running hot), remove heatsink and lap
it if it's rough on the bottom then reinstall it with a fresh coat of
thermal compound, replace the heatsink, or just wait and see if it stays
stable at that temp.

In general, the "if it ain't broke don't fix it" principle applies. If
system is running OK then there is no pressing need to change anything...
just be sure that if the case airflow is the problem, that other parts
(like video card or hard drive(s)) aren't overheating.

"I C" wrote in message
om...
When I turn on my PC, my BIOS reports the following values:

System temperature = 39C / 102F
CPU1 temperature = 57C / 134F
CPU2 temperature = 63C / 145F

Vcore1 = 1.56V
+2.5V = 2.48V
+3.3V = 3.28V
+5V = 4.91V
+12V = 11.92V
-12V = -12.03V
-5V = - 4.94V
VBAT(V)= 3.45V
5VSB(V)= 4.82V


PC voltage tolerances are specified at plus or minus 5%, that's
3.135 - 3.465 for the +3.3V line,
4.75 - 5.25V for the +5V line,
11.4 - 12.6 for the +12V line.
The -12V line is only used for serial ports and they will tolerate anything
between -3 and -15V. The -5V line isn't used for anything anymore. Vbat is
wrong but that's nothing unusual, I've seen many BIOSes report it
incorrectly.
Motherboard sensors themselves are not high accuracy devices, the A/D
converter itself is only 8 bit meaning 0.8% error at mid scale. The best
overall accuracy of the sensors would be about +-2%.

I wouldn't like my CPUs to exceed 60C as I found out from experience that
above about 65C the likelihood of unexplained random crashes increases
exponentially. My P4-2.8 o/c'd to 3.5GHz currently runs at 47C with room at
30C (rather hot in here).

Thank everyone for their responses.

My two CPUs are AMD MP2800+

If I understood correctly, the main problem of excessive temperature
is frequent crashes. Is it possible that the processor may be
destroyed? What are other possible consequences of excessive
temperature?

I removed the side-panel of my case, but the temperatures have
remained at the same level. I guess this means that the chassis
airflow is fine, so if there is something wrong this has to do with
the heatsinks (which came with the processors as a retail package). Or
it could be because today is a very hot day.





(I C) wrote in message . com...
When I turn on my PC, my BIOS reports the following values:

System temperature = 39C / 102F
CPU1 temperature = 57C / 134F
CPU2 temperature = 63C / 145F

Vcore1 = 1.56V
+2.5V = 2.48V
+3.3V = 3.28V
+5V = 4.91V
+12V = 11.92V
-12V = -12.03V
-5V = - 4.94V
VBAT(V)= 3.45V
5VSB(V)= 4.82V

Are these temperatures OK? At what level are temperatures considered
"too high"?
I guess that the longer the PC remains on, the temperature could rise.
But what about the power consumption? Would the reported values change
as I use the PC? And are the values I quote above satisfactory? I
particularly worry about the
"-12V" value, which is -12.03 and therefore higher (in absolute terms)
than 12. If this value is indeed too high, what are the implications
of doing nothing? Alternatively, what could I do to reduce it?

On 9 Jun 2004 15:31:30 -0700, (I C) wrote:

Thank everyone for their responses.

My two CPUs are AMD MP2800+

If I understood correctly, the main problem of excessive temperature
is frequent crashes. Is it possible that the processor may be
destroyed? What are other possible consequences of excessive
temperature?

There should be no damage at those temps, but crash or just an error is
possible. By error, I don't _necessarily_ mean that the operating system
shows this error, but that an error occurs... you would hope that the
error causes an OS crash rather than foul important data.



I removed the side-panel of my case, but the temperatures have
remained at the same level. I guess this means that the chassis
airflow is fine, so if there is something wrong this has to do with
the heatsinks (which came with the processors as a retail package). Or
it could be because today is a very hot day.

Typically it would be both. The heatsinks are only adequate, and higher
ambient temp certainly does result in similar if not exact same temp rise
of CPU. If system must be used in high temp environement then get better
heatsinks, but the human using the system might benefit more from an
air-conditioner instead.