Ivy Bridge Temperatures Could Be Linked To TIM Inside Integrated Heatspreader: Report

[btarunr]

PC enthusiasts with Ivy Bridge engineering samples, and reviewers at large have come to the consensus that Ivy Bridge is a slightly warmer chip than it should be. An investigation by Overclockers.com revealed a possible contributing factor to that. Upon carefully removing the integrated heatspreader (IHS) of an Ivy Bridge Core processor (that nickel-plated copper plate on top of the processor which makes contact with the cooler), the investigator found common thermal paste between the CPU die and the IHS, and along the sides of the die.

In comparison, Intel used flux-less solder to bind the IHS to the die on previous-generation Sandy Bridge Core processors in the LGA1155 package. Attempting to remove IHS off a chip with flux-less solder won't end well, as it could rip the die off the package. On the other hand, the idea behind use of flux-less solder in CPU packages is to improve heat transfer between the die and the IHS. Using thermal paste to do the job results in slightly inferior heat transfer, but removing IHS is safer. One can be sure that making it safe for IHS removal couldn't have been the issue behind switching back to conventional thermal paste, as everything under the IHS isn't user-serviceable anyway, and off limits for them. Perhaps Intel kept extreme overclockers in mind.



View at TechPowerUp Main Site

 

[Fourstaff]

Perhaps Intel kept extreme overclockers in mind.

And shafted everyone else? Not a good trade I say :/

Someone change the TIM into a solder and give it a shot?

 

[TheoneandonlyMrK]

is this a joke? if not wow

Someone change the TIM into a solder and give it a shot?

forget that what will it do direct cooled

 

[NC37]

I wonder who at Intel thought this was a great idea. I would guess it would cut the costs of production. But by how much? The heat issues have made me reconsider a SB build when I would have been willing to pay more for IB.

 

[v12dock]

I am sure Intel engineers tested with and without the TIM

 

[sneekypeet]

Awesome so we are back to de-lidding procs. to get the most out of them again. Been a while since this could be done reliably. Wonder what they do like this on extreme cooling.

 

[erocker]

I wonder how it performs without the heatspreader on it.

 

[boogerlad]

I'm not even mad. This is great Intel! I love delidding my cpus. Now I don't need a torch to do so.

 

[cadaveca]

Perhaps Intel kept extreme overclockers in mind.

As if.

They did it so cooler manufacturers, their partners, can sell new coolers, designed to deal with this new "heat problem".





You'll see these new coolers in the coming months and at Computex.

I hope a few notice the die orientation in reference to the PCB, and note how the CPU sits in the socket, and that they align heat-pipe-based coolers in the right direction.





Pretty sure the extreme guys already know this anyway. Power consumed + CPU temps + pot temps = problems with IHS contact. Why do you think it was tried in the first place?

IF anything, Intel made sure to have a socket clamp design that is easily removed so that extreme guys don't have clearance issues with de-lidded CPUs and the retention mechanism.

 

[TheoneandonlyMrK]

perhaps intel couldnt apply the initial heat to fit normal Ihs without damageing their new tri gates or the chip on mass, it is a different process.

 

[douglatins]

Now i wait for a revision Lol

"The first answer is correct, but wrong at the same time – power density is greater, but it isn’t what is causing temperatures to be as much as 20 °C higher on Ivy Bridge compared to Sandy Bridge when overclocked."

OMFG WUT????

 

[qubit]

This smacks to me of Intel not designing IB with enthusiasts in mind. AnandTech's IB review includes a good explanation of Intel's shift away from targeting big, power-hungry PCs, to small portable computers. With this in mind, it makes sense, most likely because some piece of shit TIM is cheaper to make and less likely to damage the core during manufacture, resulting in trashed product.

Note that the Overclockers article explains that the difference in thermal conductivity between solder and TIM is absolutely huge:

Solder: 80 W/mK
POS TIM: 5 W/mK

Along with the marginal instructions per clock improvement (and almost zero gaming benefit) and now this, I'm glad that I bought my 2700K last November and have no desire to replace it anytime soon. If I were buying a laptop though, getting IB would be a no-brainer.

 

[_Zod_]

These are engineering samples, of course they are using tim. The production runs will be soldered. You have to take reviews of engineering samples with a grain of salt, wait for tests of production models.

 

[douglatins]

This smacks to me of Intel not designing IB with enthusiasts in mind. AnandTech's IB review includes a good explanation of Intel's shift away from targeting big, power-hungry PCs, to small portable computers. With this in mind, it makes sense, most likely because some piece of shit TIM is cheaper to make and less likely to damage the core during manufacture, resulting in trashed product.

Note that the Overclockers article explains that the difference in thermal conductivity between solder and TIM is absolutely huge:

Solder: 80 W/mK
POS TIM: 5 W/mK

Along with the marginal instructions per clock improvement (and almost zero gaming benefit) and now this, I'm glad that I bought my 2700K last November and have no desire to replace it anytime soon. If I were buying a laptop though, getting IB would be a no-brainer.

Yeah that change is HUGE and really unacceptable. It completely ruins overclocking.

At least maybe the 2700k could be cheaper now.

Annnd i don't think its because they are ES, i could get behind that if there were previous accounts of ES having TIM

 

[qwerty_lesh]

These are engineering samples, of course they are using tim. The production runs will be soldered. You have to take reviews of engineering samples with a grain of salt, wait for tests of production models.

well if what you say is true we'll soon find out when people assuming TIM is used wreck their IVB cpus delidding them. its only a matter of time until we know if the retail ones are tim or solder.

my money is on them all being TIM though.

 

[qubit]

Yeah that change is HUGE and really unacceptable. It completely ruins overclocking.

At least maybe the 2700k could be cheaper now.

Annnd i don't think its because they are ES, i could get behind that if there were previous accounts of ES having TIM

Yeah, don't it just?

Unfortunately, this issue means that the 2700K will stay UP in price, not come down, since savvy enthusiasts will prefer this to IB. I now really don't want to overdrive mine too hard, to ensure that it has a long service life.

 

[OneCool]

(that steel plate on top of the processor which makes contact with the cooler)

No.Its copper with some kind of powder coat or anodised coating.


If they removed the IHS and assumed this was the reason behind the higher temps why didnt they just put it back in and confirm it?

Anyway i guess it could be because these are ES.

 

[function69]

"Using thermal paste to do the job results in slightly inferior heat transfer" - well that's an understatement!

Also, everybody is talking about waiting for another stepping, but I'm not sure much can be done here, considering the 22nm tri-gate process itself is an inherent obstacle.

 

[eidairaman1]

simple fix for users who want to keep the cap on. metal thermal compound or TIMs

 

[phanbuey]

this is so easy to fix... pop the lid, and then OC the crap out of it.

 

[qubit]

this is so easy to fix... pop the lid, and then OC the crap out of it.

Yes, if you can manage to not physically wreck it one way or another, I'll bet you can. There are reports of these things hitting 6.5-7GHz just as they are, with extreme cooling. These don't have the cold bug, so can you imagine what a naked IB on LN2 could do?! I can't wait to see it and you can bet that famous Chinese overclocker (forget his name) will try this.

 

[phanbuey]

Yes, if you can manage to not physically wreck it one way or another, I'll bet you can. There are reports of these things hitting 6.5-7GHz just as they are, with extreme cooling. These don't have the cold bug, so can you imagine what a naked IB on LN2 could do?! I can't wait to see it and you can bet that famous Chinese overclocker (forget his name) will try this.

you cant really wreck it - there are no cache chips or transistors on the top... all you need is a razor to cut away rubber and gently pull off the IHS - i did this on my old e4300.

Didn't help with the OC, but that thing ran like that for 2 years.

 

[qubit]

you cant really wreck it - there are no cache chips or transistors on the top... all you need is a razor to cut away rubber and gently pull off the IHS - i did this on my old e4300.

Didn't help with the OC, but that thing ran like that for 2 years.

But how much cooler did run?

 

[MeanBruce]

These are engineering samples, of course they are using tim. The production runs will be soldered. You have to take reviews of engineering samples with a grain of salt, wait for tests of production models.

Will we see these tests next week as retail samples hit the street? Why do they even review ES samples to publish, fuckin stupid pissing me off.

 

[douglatins]

Yeah, don't it just?

Unfortunately, this issue means that the 2700K will stay UP in price, not come down, since savvy enthusiasts will prefer this to IB. I now really don't want to overdrive mine too hard, to ensure that it has a long service life.

I don't think we are in such numbers to do that. Most maisntream High end PC owners just buy them prebuilt or don't oc that much