test of thermal pad on AMD

T

Trent©

New's to me, where did you find this information and please posts some links
backing this info up. AFAIK electronics love running cool and I've never
heard "They run best at medium to medium-high" anywhere.

Experience, my child!! lol

You car works the same way.

I try to shoot for the median temperature between the recommended high
and low temps.


Have a nice week...

Trent©

Follow Joan Rivers' example --- get pre-embalmed!
 
T

Trent©

Why? I can -SEE- the difference between the stock HS and one lapped without
one..

When installing a pad, why do you think that having a non-smooth
surface would be a BAD idea?


Have a nice week...

Trent©

Follow Joan Rivers' example --- get pre-embalmed!
 
J

John

You might try a Thermalright SLK947U,
$19.99 w/coupon code "SVC947"
May 17 thru 23
http://www.svc.com/thslsoco.html

Although, for optimal results you still might consider lapping it... a few
months back, maybe a year ago I posted about lapping an SLK900 till it
acheived same temps "dry" as with heatsink compound on it. The degree to
which I lapped it was a waste though, beyond a certain point the return
was diminishing towards nothing, especially since I used thermal compound
anyway on final fitting. On an o'c mobile Barton it gets about 20F over
ambient but it only has a voltage-reduced 80mm Panaflo M1A on it...
primary goal was highest o'c possible while still silent. At the time
the SLK947 was just hitting the market and if I'd the chance to buy again
I'd get the 947 due to higher m'board compatibility, as it's fins aren't
so wide. 947 is also better suited to P4 due to wider base, but that's
splitting hairs a bit, either is fine for P4 too.

Yeah Id never buy another heatsink unless I was guaranteed a 20 F drop
or even 10-15F too much of a hassle. What Intrigued me was the poster
was getting incredibly low readings with a stock unit. I think they
live in Canada , I live in Hawaii so the temps maybe very different
now.

The AMD heatsink fans are sooooo quiet compared to my usual fans that
any other units would have to be reasonably quiet AND cause a 20F
drop.

Last time I tried and this was my usual routine on all units since the
usual advice is to apply paste - I used 400 and 600 sandpaper and then
washed the surface with water and then rubbing alcohol and then put
paste on it. Because I didnt notice much of a difference I would end
up putting the heatsink on several times to see if I put it on right.

Another thing is anything sent here usually means another $20-25
tacked on to the price.

Ill see if they have the fan selling at local shops though and check
it out the next time I have to buy one.
 
K

kony

Experience, my child!! lol

You car works the same way.

I try to shoot for the median temperature between the recommended high
and low temps.

CPU will work fine at temps lower than mechanical parts or electrolytic
caps will.
 
S

Stacey

Trent© said:
Experience, my child!! lol

You car works the same way.

Not even the same sport. You're talking about a thermal engine vs
electronics. Again please post ONE link saying this is true or I call BS.
 
S

Stacey

Trent© said:
When installing a pad, why do you think that having a non-smooth
surface would be a BAD idea?


Because metal to die contact is ALWAYS a better conductor than any pad/paste
etc.
 
T

Trent©

CPU will work fine at temps lower than mechanical parts or electrolytic
caps will.

I never said anything about 'fine'.

And there IS a high temperatore...and a low temperature...where the
CPU will not work at all...or not for very long, at least. I think we
can all agree on that.

My determination of the optimum operating temperature is somewhere
between the high and low recommended by the manufacturer. Neither the
high nor the low is ever my goal.


Have a nice week...

Trent©

Follow Joan Rivers' example --- get pre-embalmed!
 
T

Trent©

Because metal to die contact is ALWAYS a better conductor than any pad/paste
etc.

You need to buy a microscope! lol


Have a nice week...

Trent©

Follow Joan Rivers' example --- get pre-embalmed!
 
T

Trent©

Not even the same sport. You're talking about a thermal engine vs
electronics.

What do you think it is that makes your car move?

Why do you think CPU manufacturers give a low and a high temperature
max setting? Why do you think some temperature in between might not
ALSO be important?

Why do you think standard fluorescent bulbs don't work well in cold
temperatures?
Again please post ONE link saying this is true or I call BS.

Just call it BS then...and let's move on.


Have a nice week...

Trent©

Follow Joan Rivers' example --- get pre-embalmed!
 
M

~misfit~

Trent© said:
What do you think it is that makes your car move?

A volatile gas/vapour mixture exploding in a controlled environment where
the expansion of the gas as it explodes is converted into kinetic energy
(Movement) by directing it's force against the top of a piston, causing it
to be pushed down a cylinder. The piston is connected (by a 'connecting rod'
of all things, often shortened to con-rod) to a crankshaft, off centre from
the crankshaft's centre of rotation. The piston being forced down the
cylinder by the expanding gasses has the effect of turning the crankshaft,
thereby turning (albeit through a complicated array of things like a gearbox
and differential) the wheels and providing motive force.

However I don't see a fuel-induction system, pistons or wheels on my CPU.
Maybe I need a microscope? Or maybe LSD?
Why do you think CPU manufacturers give a low and a high temperature
max setting? Why do you think some temperature in between might not
ALSO be important?

Have you managed to find low temperature figures for CPU cores? Maybe you'll
find a recommended low for the environment it runs in. However, that is due
to the dew-point of the surrounding air and the desire to keep condensation
(water) away from the CPU. Tomshardware got a P4 up over 5Ghz using liquid
nitrogen cooling. (minus 190-odd °C) I know a guy who uses a phase-change
cooling system to keep his CPU around -2°C and he runs stably and
consistantly at a phenomenal overclock. If what you say is true then the two
examples above wouldn't make sense now would they?
Why do you think standard fluorescent bulbs don't work well in cold
temperatures?

It's a matter of physics. Fluorescent bulbs work by 'exciting' gas molecules
to the extent where they are *so* excited (Read:
flying-around-bashing-into-each-other) they give off light (but very little
heat). This is then enhanced by putting the gas in a glass container with
the interior coated with phosphors that react to, and amplify the light the
gas gives off (as a lot of the light given off by the gas is in the UV
range, the phosphors convert it to visible light). Ever heard of P over T
equaling V? V = Volume, T = Temperature and P = Pressure. If the temperature
goes up but the volume remains consistant then the pressure will increase.
Why you ask? Well, at higher temperatures the molecules are more excited,
more mobile. They fly around crashing into each other *and their container*
a lot more than at lower temps. This crashing into the walls of the
container more increases pressure.

From the above you should be able to see that, in cold temperatures, the gas
molecules may be so 'unexcited' that even passing an electric current
through the container may not be enough to get them moving enough to crash
and glow properly. Hell, you even get those little molecules cold enough and
they'll all give up flying around altogether and lie down and just wriggle.
Ever hear of liquid nitrogen and wonder why it's so cold? It's a gas that's
been chilled (or compressed, same effect) to the extent the molecules have
very little energy and lie down and become a liquid.

What happens inside a CPU on a sub-atomic level is far and away different to
what happens in a fuorescent light.

As with the car analogy, it is totally inapropriate. Your CPU *isn't* a
fluorescent light. You might as well say that eggs fry nicely at 130°C so
that should be a good temperature to run your CPU at. However, as well as
not being a fluorescent light or a car engine, your CPU isn't an egg.

You could also say that, if you swim in really cold water, your penis
shrinks. Therefore low temperatures are no good for CPUs. The thing is, your
CPU isn't a penis either.
Just call it BS then...and let's move on.

It's BS. As are a lot of your comments and advice. However I don't have the
time or the inclination to refute all of them, I'll just stick to one or two
here and there, just so people who read here know to take your advice with a
pinch of salt.

A little knowledge is a dangerous thing.
 
S

Stacey

Trent© said:
What do you think it is that makes your car move?

The heating of air trapped in the cylinder. When the engine it too cold the
gasoline/fuel doesn't vaporize very well and there are thermal losses of
the heated air (from the burning of a combustable material) to the cold
engine parts instead of expanding and creating pressure. I wasn't aware
that heated air expanding is what makes a CPU operate.
Why do you think CPU manufacturers give a low and a high temperature
max setting?

Please post a link showing a minimum CPU temp spec.
Why do you think some temperature in between might not
ALSO be important?

Given there is no low temp spec I've ever seen for a CPU, how can you
determine this? Nitrogen cooled CPUs seem to operate just fine!
Why do you think standard fluorescent bulbs don't work well in cold
temperatures?

So why wouldn't a CPU work well at the temp you cook a pizza at? Makes about
as good an analogy.

Just call it BS then...and let's move on.


Stop posting BS or post something to back it up.
 
S

Stacey

Trent© said:
I never said anything about 'fine'.

And there IS a high temperatore...and a low temperature...where the
CPU will not work at all...or not for very long, at least. I think we
can all agree on that.

No "we" don't agree to this, there is no low temp where it won't work at all
that I've ever seen documented. Here's one running great at -37C

http://www.overclockers.com/tips422/

Here's what toms hardware said about cooling "We learned firsthand how to
sufficiently cool the Intel P4 to -196°C to bring it past the 5 GHz mark."

http://www4.tomshardware.com/cpu/20031230/
My determination of the optimum operating temperature is somewhere
between the high and low recommended by the manufacturer. Neither the
high nor the low is ever my goal.


Then please enlighten us to what this low temp specification is. I've never
seen it and would like to know what it is.
 
T

Trent©

No "we" don't agree to this, there is no low temp where it won't work at all

....or not for very long
that I've ever seen documented. Here's one running great at -37C

http://www.overclockers.com/tips422/

Here's what toms hardware said about cooling "We learned firsthand how to
sufficiently cool the Intel P4 to -196°C to bring it past the 5 GHz mark."

http://www4.tomshardware.com/cpu/20031230/

Sorry...I thought you were talkin' about the real world. If he got a
machine to boot up at -196° C., then I stand corrected. And that's
probably the temperature you should run your machine at...'cause
that'll keep it nice an' cool, all right! lol

If your goal is just to be correct...and for me to be wrong...you win.
I stand corrected.

But, on the other hand...yer references are simply absurd...at least
in most of the world.
Then please enlighten us to what this low temp specification is. I've never
seen it and would like to know what it is.

Low temperature is -196° C...I'm not sure about the high temperature.
What the hell IS the temperature on the Sun...any idea? That would
probably be the high side.

Then try to reach a temperature somewhere in between! lol

I quit...you win. lol


Have a nice week...

Trent©

Follow Joan Rivers' example --- get pre-embalmed!
 
T

Trent©

The heating of air trapped in the cylinder.

No...that's not it. Heat can't push those cylinders.
Stop posting BS or post something to back it up.

Feel free to jump in whenever you see it. I kinda like this! lol


Have a nice week...

Trent©

Follow Joan Rivers' example --- get pre-embalmed!
 
M

~misfit~

Trent© said:
...or not for very long

Please explain that comment. With references to back up your explaination.
Sorry...I thought you were talkin' about the real world.

You've never heard of super-cooled CPUs before? Makes one wonder if you know
where the real world is.
If he got a
machine to boot up at -196° C., then I stand corrected. And that's
probably the temperature you should run your machine at...'cause
that'll keep it nice an' cool, all right! lol

Sarcasm is the lowest form of wit.
If your goal is just to be correct...and for me to be wrong...you win.
I stand corrected.

As you should, you were wrong.
But, on the other hand...yer references are simply absurd...at least
in most of the world.

Maybe in your world, not in the world the rest of us live in. I know of a
few people who use phase-change coolers (refrigeration units) to get more
speed out of their CPUs. It's an undisputed fact that CPUs perform better at
lower temperatures, last longer and are capable of more speed.
Low temperature is -196° C...I'm not sure about the high temperature.
What the hell IS the temperature on the Sun...any idea? That would
probably be the high side.

Ha ha. Sarcasm is *so* funny.

I noticed you didn't address my rebuttal of your statements.

Stick to what you know Trent.
 
T

Trent©

Clueless about more than just computers it seems....

http://inventors.about.com/library/inventors/blinternalcombustion.htm


"The combustion of the fuel results in the generation of heat, and the hot
gases that are in the cylinder are then at a higher pressure than the
fuel-air mixture and so drive the piston back down."

'Gases...drive...'

Heat can't push those cylinders.

Thank you for confirming that. But common sense should have told you
that.

Heat is but a catalyst.


Have a nice week...

Trent©

Follow Joan Rivers' example --- get pre-embalmed!
 
S

Stacey

Trent© said:
'Gases...drive...'

Heat can't push those cylinders.


No Trent, the expansion of the gasses -from being heated- is what creates
the pressure. So yes Trent is IS the heat that pushes the cylinders and why
a cold engine isn't as efficient as a warm one is.. A cold engine absorbs
some of this heat from combustion so less is avalible to heat up the air
(gasses) in the cylinder. It's also why cold air induction makes power,
cold air is more dense and so there is more air in the cylinder to heat up
and expand.

If not, lets hear your reason why a cold engine isn't as efficient? This
should be interesting! :)
 
C

CBFalconer

Stacey said:
Trent© wrote:
.... snip ...

No Trent, the expansion of the gasses -from being heated- is what
creates the pressure. So yes Trent is IS the heat that pushes the
cylinders and why a cold engine isn't as efficient as a warm one
.... snip ...

The whole thing comes under the general heading of heat engines,
and has been fairly well understood since the mid 19th century or
earlier. Look up Carnot cycle.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top