|
Calculators
To help take some of the repetitive drudgery out of testing, I have written a couple of calculators to help me figure heat output for my processors and °C/W for my heat sinks.
Last update -- April 16, 2003
Please let me know if your current processor is not listed.
Calculate Your Overclocked Wattage
Fahrenheit to Celsius Temperature Conversion
Calculate Your °C/W
This calculator uses 5 values to compute the
power consumption of your overclocked processor. The equation is:
Overclocked Watts = Default Watts * (Overclocked
Mhz \ Default Mhz) * (Overclocked Vcore \ Default Vcore)². To
make life easier, I included some recent processors with the default
speed, Vcore and watts. Because the stepping of a processor often
changes Vcore and watts info, make sure you check to see that the numbers
are correct for your particular version.
The Intel Celeron and Pentium III information may be different than what you are used to seeing. I decided against using Intel's "Processor Thermal Design Power" specifications (because they're for "reference", not actual values) . In fact, in their notes, they allude to the fact that they are just reference numbers.
(Page 47 of 24526407.pdf)
"2. Thermal Design Power (TDP) represents
the maximum amount of power the thermal solution is required to dissipate.
The thermal solution should be designed to dissipate the TDP
power without exceeding the maximum Tjunction specification.
3. TDP does not represent the power delivery
and voltage regulation requirements for the processor. Refer to the tables for voltage regulation and electrical specifications. "
So, as the notes said, I referred to Table 5 (Celeron) and 7 (PIII).
Specs for the Celeron were taken from Intel Data Sheet 243658.pdf - Table 5, pages 26 & 27 where watts = volts * amps.
Specs for the Celeron (0.13) were taken from Intel Data Sheet 29859604.pdf - Table 7, page 26 where watts = volts * amps.
Specs for the Pentium 3 (.18 micron) are from Intel Data Sheet 24526407.pdf - Table 7, pages 28 to 31 where watts = volts * amps.
Specs for the Pentium 3 (.13 micron) are from Intel Data Sheet 24965704.pdf - Table 7, page 27 where watts = volts * amps.
Specs for the Pentium 4 (.18 micron) were originally taken from Intel Data Sheet 24988701 - Table 6, page 20; however, that PDF seems to have disappeared. This is the most recent Intel PDF file: Intel Data Sheet 24919805.pdf - Table 5, page 19 where watts = volts * amps.
Specs for the Pentium 4 (.13 micron) were originally taken from Intel Data Sheet 29864301 - Table 7, page 23; however, that PDF seems to have also disappeared. This is the most recent Intel PDF file: Intel Data Sheet 29864307.pdf - Table 2-6, page 25 where watts = volts * amps.
NOTE: If you find that the Intel Data Sheets have changed (again), please let me know and I will make the appropriate changes.
Specs for the AMD Thoroughbred are from AMD White Papers and Tech Docs 25175.pdf where (maximum) watts are used.
Pentium 4 NOTE: The maximum wattage for the P4 will probably never be reached. The P4 uses a Thermal Control Circuit (TCC)
which has a speed throttling feature that may be set to "automatic" or "on demand" through the system BIOS. In "automatic" mode, the processor's clock speed
is reduced by enabling a duty cycle (50% on - 50% off) when the core temperature approaches the pre-set thermal limit. In "on demand" mode, the duty cycle can be programmed from 12.5% on/ 87.5% off, to 87.5% on/12.5% off in 12.5% increments. In addition to throttling the processor's speed, in the event of a catastrophic cooling failure, the processor will automatically shut down when the silicon has reached a temperature of approximately 135 °C.
If your processor is not listed, you can
find the information you need from Intel,
AMD
or for quick reference, on Chris Hare's electrical
specifications page.
Once you have computed your processor's
watts, measure the ambient temperature of the intake air for the heat sink
fan and highest "hot side" temperature of the processor or heat sink.
To get an idea of sensor placement, take
a look at Stephen Hoar's articles
Taking
Temperatures and Hot
Air at BurningIssues.net.
Remember that WHERE you measure the hot side will have a direct impact
on your °C/W.
If you use the temperature from an Intel
chip with internal diode, the °C/W will show higher than measuring
the hot side from a sensor mounted in a hole drilled in the base of the
heat sink. The extra thermal junction between the chip and the heat
sink and the material of the chip itself make the difference.
For the TBirds, the difference in possible
high temperatures is a bit more difficult to judge. Depending on
the placement and the accuracy of the thermosensor (which is usually located
under the socket), the high temperature could be lower or higher (usually
lower) than a sensor drilled into the heat sink base. I wish that
one of the hardware sites who handle many motherboards, would do a comparison.
Using the same case, processor, heat sink and temperature software
- document the temperatures indicated by the different motherboards.
Stress Test
In looking for HIPWR30, I came across a
post by the author of burnp6, Robert Redelmeier, where he talks about
how much stress is actually being placed on the processor by his program
- and I quote from the post dated 09/27/1999:
"I am developing an Open Source program to test CPU stability (cooling)
with P6 optimized assembler to maximize current draw, hence heat production.
This way heatsinks [passive or active], interface materials, motherboard
power regulators and power supplies can be stressed in a fully assembled
system.
I am fairly satisfied with my beta program `cpuburn` (`burnP6`)
http://users.ev1.net/~redelm.
I measure about 88% of the datasheet power draw through an inductive
pickup.
I was curious if the 88% figure was still
accurate and checked out his site. I found that he is now recommending
the use of a program called runprio
to change the priority of running burnp6 to high. He claims an extra
5°C is generated in this mode of operation. I wrote Robert and
asked what he thought the percent of power draw was when using runprio
in conjunction with burnp6. He answered that it was probably not
more than 90%.
When using runprio and BurnP6 from the
command line (:\> runprio -x high burnp6.exe) burnp6 gets 96% of
the processor's time instead of 72% at normal priority. However,
I have had a problem running the priority on high and trying to get Motherboard
Monitor to log the temperatures. Instead of logging the temps every
6 seconds as usual, MBM will only log about once every couple of minutes.
For my purpose of trying to track temperatures, this won't do. I
now run burnp6 by itself and when computing °C/W, I adjust the
watts to 88% of the total overclocked wattage. I think that this
should be accurate, but will not know for sure until I get my new heat
sink tester completed. I hope you find these scripts useful. If
you find any mistakes in the values, please let me know.
Fahrenheit
to Celsius Temperature Conversion:
Enter a number
in either field,
then click outside the text box, or press
the tab key.
Celsius:
Calculate
Your °C/W
Computer case air temp °C
Processor high temp °C
Processor wattage
°C/W Value
°C/W Analysis
The last item you will need is a program
to "stress" your processor. I have used many programs, including
games, for this purpose. I am now of the opinion that the best stress
tester from the standpoint of keeping a constant high temperature and being
able to reproduce the results in test after test is burnp6
(CPUBurn). I should add that Intel's program HIPWR30.exe may be as
good or better, but I have been unable to get a hold of a copy.
If I just need to make sure that a processor is stable at a certain speed,
I run burnp6 through runprio.