The on-line specs for the various flavours of chips given their current
consumption per Mhz. Since the power is ALL becoming heat anyway it
simply becomes a matter of whose heat sink has the lowest thermal
resistance.

Put a bigger heatsink on ANY chip and it will run cooler, IF the
input power (and air flow) remains the same.
All you need to do is compare the current consumption between AMD
and Intel for similar MIPS and then you'll have a quantitative
measure of efficiency.

I assume that the overclocking web sites or discussion groups would
have some hard numbers since these gamers squeeze every clock cycle
they can out of their system, and that means running the chips as
cold as they make them e.g. liquid cooling, TECs, massive copper
heatsinks, etc.

Russell McMahon wrote:
>
> In private correspondence, re claims about AMD versus Intel PC CPU operating
> temperatures,  Rod wrote
>
> > Have trouble believing there is much difference between I and AMD ...
>
> It's garbage.
>
> (Someone had written)
> >> ... , but did you know that with the fan/heatsink off an AMD
> >> Palomino runs at 5680F and an AMD Athlon 1400 it run at 6980F! If the fan
> >> fails that leaves only the heatsink to dissipate all that heat.

Which is why modern motherboards have CPU fan RPM & temperature monitoring,
and turn down/off the CPU when it gets too hot.
Unfortunately too many motherboards -forget- to monitor the chipset
fan, and when it fails (anyone checked their Asus V7V266 MB fan's lately?),
the chipset is usually toast.

> >> In comparison, the Intel Pentium 4 2Ghz runs at 840F and the Intel
> Pentium
> >> III 1Ghz runs at 1000F with the fan/heatsink off.
>
> ALL modern processors draw extremely high currents at low voltage.
> Haven't looked up requirements but probably 20A plus plus at 1.5 to 3 volts.
>
> Even 20 watts (below any of above) on an unheatsunk chip will fry it vvv
> quickly.
>
> It may be that Intel's cpu's have a little more thermal mass than AMDs which
> may allow them to survive for a few more seconds totally unheatsunk. (But an
> eg  Celeron 1700 looks pretty minimal in any sort of mass to me).
>
> Just possibly Intel's heatsinks are superior and without a fan they run
> cooler.
> But even then the temperatures they claim are bunk.
> 84 F = 27C

If the ambient air is at 0C or lower, the numbers are 'real'.
You have to read the fine print to find out what the ambient test
temperature is. And they probably test with a box the has a bloody
hurricane blowing through it so ambient STAYS at "ambient".

> The heatsink on the Celeron 1700 I installed a few days ago runs annoyingly
> hot with VERY large Intel heatsink and fan. I'd say around 50C !

You just need to move more air through the case <G>.

> Anyone here have any REAL figures?
>
>         Russell McMahon

Many gamers have thermistor based CPU temperature displays on their
cases. You could always visit a LAN party and see what the boxes
do it real life. These displays are cheap and non-computer versions
are readily available at RadioShaft should you want to make your
own measurements.

Robert

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu