I heard somewhere that if you remove the heatspreader on your chip

[Crescent13]

is there any truth to this? does removing the heatspreader really lower the temps enough to make a higher overclock?

 

[Duvie]

YMMV...Your Mileage May Vary....

For some it has worked and for others it made no difference to actually damaging the chip...

The IHS or integrated HeatSpreader was incorporated first and foremost to protect the fragile cores. Sometimes through process of manfacturing they dont make the best contact and hence why some chips may vary so much among themselves. I think in the most part the minimal advanatge in Celsius you may see does not overshadow the risk in damage you may incur...

 

[Leper Messiah]

yes. but only if you have a HSFU that will be able to soildly contact the core surface. You can also use a pre 1963 penny, sanded smooth if it doesn't. the solid copper should be better than the IHS.

 

[Shimmishim]

i have a naked chip.

i haven't tried going higher but i know for a fact that the chip runs 8C's cooler than before.

its risky though cuz you can end up killing the chip.

 

[xenolith]

Originally posted by: Leper Messiah
yes. but only if you have a HSFU that will be able to soildly contact the core surface. You can also use a pre 1963 penny, sanded smooth if it doesn't. the solid copper should be better than the IHS.

Why do you think they put the IHS on? Oh they say it's to protect the core...

... overclocking? Oh no... we don't do any of that here Mr. Ruiz.

 

[seanp789]

You can get marginally better temps if you connect the core directly to the HSF however the chances of cracking your core are increased conciderably. I would not recommend it unless you are will to buy a new core and are pretty comfortable with CPU hardware.

 

[Crescent13]

okay, thanks everybody.

 

[RichUK]

When i get my next CPU, (dont know whether to go for a dual core, X2 or opty s939), I will be looking to take off the IHS and put the right thickness of neoprene around the core, so to act as a stabilizer like the four circular pads on the socket A's, this is so when applying the heatsink or water block i dont put to much pressure on one of the sides and crush/crack the core . Then I will apply my soon to be water setup, and prepare myself for some immense Overclocking <<I Hope>>

I'm feeling brave, i just don't want to make a mistake when taking off the IHS, as i have read some people have razored to far and into the pin grid array, and chopped some vital tracers, i think i will practice on my Winnie first

 

[andrewbabcock]

I wouldn't reccomend it. You are better off water cooling or just getting a better air soulution. XP-90C with a zalman on top...

 

[redhatlinux]

And you think that you can get 'a pre 1963 penny of solid copper sanded smooth' so that it contacts the core and the base of the heatsink completelyand of course it will not corrode (can we say copper oxide anybody), and the pressure from the heatsink is within AMD specs. Gotta bridge to sell you.

 

[Leper Messiah]

Originally posted by: redhatlinux
And you think that you can get 'a pre 1963 penny of solid copper sanded smooth' so that it contacts the core and the base of the heatsink completelyand of course it will not corrode (can we say copper oxide anybody), and the pressure from the heatsink is within AMD specs. Gotta bridge to sell you.

um why not? Copper is copper. If the penny was going to corrode, so is the copper base of your HSFU when you put it on the aluminum IHS.

Granted, you concern about the cracking of the core is valid, but if you do something like RichUK said, you should be fine...

 

[Furen]

The IHS is not aluminum, but nickel-plated copper, which makes it resilient to oxidation, among other things. if you make a deep enough grove into it you should get to the copper. A penny would be worse than the IHS unless the IHS was not making proper contact with the CPU or heatsink.