bit-tech.net

EK Supremacy PreciseMount for delidded Ivy Bridge CPUs Review

Comments 1 to 25 of 27

Reply
Dave Lister 8th June 2013, 10:38 Quote
What a good idea, I wonder if they take into account variations in motherboard thickness though as that is something i'd still be worried about if i was looking at buying this product.
Stanley Tweedle 8th June 2013, 11:07 Quote
LOL!! You feature this at the same time I searched direct die and posted it here on the Haswell heat feature.

Great minds think alike.

60c down to 43 is amazing.

Intel really need to fire a few numpties and rethink this whole IHS crap.
V3ctor 8th June 2013, 11:23 Quote
I'm going to delid my 4770k next week, at 4.7Ghz with Prime96 it hits 100ºc.

Thats just absurd... The only good thing is, that in cpu benchmarks the 4770k at 4.5Ghz is at the same level as a 3770k@4.7-4.8Ghz, and 2600k@5-5.1Ghz
Combatus 8th June 2013, 11:28 Quote
Do let us know how you get on!
abezors 8th June 2013, 12:44 Quote
Such a niche product yet they only charge £4 for it. That's when you know they're as pissed with Intel as the rest of us
Stanley Tweedle 8th June 2013, 13:22 Quote
Quote:
Originally Posted by abezors
Such a niche product yet they only charge £4 for it. That's when you know they're as pissed with Intel as the rest of us

Well it's hard to charge £150 for a set of screws and washers. Microsoft and Apple undoubtedly would but not a waterblock maker.
V3ctor 8th June 2013, 14:12 Quote
My only problem with deliding is this...

https://pbs.twimg.com/media/BLr_dhkCIAAWdje.jpg:large

Next to the die, there is a row of transistors, if i'm not mistaken Coollaboratory Liquid Pro is conductive. I don't want to short circuit the cpu :D

I don't know if i should put nail polish on top of those transistors, but with temperatures so high I don't know if it will hold or it will make them hotter... :/
phuzz 8th June 2013, 14:43 Quote
Quote:
Originally Posted by V3ctor
Coollaboratory Liquid Pro is conductive. I don't want to short circuit the cpu :D
If the cooling liquid it getting anywhere near the CPU you're in trouble, IHS or no.
That's why you leak test, I'm bloody glad I do, had a couple of near misses, saved by leaktesting first..
S1W1 8th June 2013, 14:52 Quote
Quote:
Originally Posted by V3ctor
Next to the die, there is a row of transistors, if i'm not mistaken Coollaboratory Liquid Pro is conductive. I don't want to short circuit the cpu :D

I was just wondering the same thing - an all-metal thermal paste sounds risky for use under the IHS where there are exposed components.
Quote:
Originally Posted by phuzz
If the cooling liquid it getting anywhere near the CPU you're in trouble, IHS or no.

I think the Coollaboratory Liquid Pro is a thermal paste, not a coolant.
Stanley Tweedle 8th June 2013, 15:06 Quote
LOL. I actually found this conversation confusing. I didn't know people were using thermal paste as coolant in their liquid loop. You'd need a hell of a lot of it.
Combatus 8th June 2013, 15:37 Quote
Quote:
Originally Posted by V3ctor
My only problem with deliding is this...

https://pbs.twimg.com/media/BLr_dhkCIAAWdje.jpg:large

Next to the die, there is a row of transistors, if i'm not mistaken Coollaboratory Liquid Pro is conductive. I don't want to short circuit the cpu :D

I don't know if i should put nail polish on top of those transistors, but with temperatures so high I don't know if it will hold or it will make them hotter... :/

Oh blimey. Yes, would be a bit dodgy. I doubt covering them in some kind if electrical insulating liquid that hardens will impact on cooling. I'm sure they'd be okay but equally the liquid metal thermal pastes are by far the best ones to use here.
djzic 8th June 2013, 18:15 Quote
Dielectric grease like you use when you do LN2 or phase change cooling should do nicely?
Combatus 8th June 2013, 18:24 Quote
Yep or liquid electrical tape.
Bakes 8th June 2013, 18:30 Quote
Or nail varnish, if you just want something to hand.
Xploitedtitan 8th June 2013, 18:35 Quote
Somehow, I recalled Liquid Pro's to be pretty corrosive (attacking at least copper bases). Not sure I would put that on an unprotected CPU core.

Btw, would be nice to remind ppl that taking off the IHS cancels the warranty. ;)
Combatus 8th June 2013, 18:52 Quote
It's just aluminium that it had issues with - copper water blocks and the core should be fine. It will indeed void you're warranty but we're assuming you've already de-lidded your CPU here hence we haven't done a full guide :)
siliconfanatic 8th June 2013, 19:17 Quote
Ninja'd... As usual.
V3ctor 8th June 2013, 21:54 Quote
Maybe i'll test the Coollaboratory first between the IHS and the cooler... see if the temps get higher after a month or not...

But my only problem is that row of transistors... :/
bullseye 8th June 2013, 22:20 Quote
Quote:
Originally Posted by V3ctor
My only problem with deliding is this...

https://pbs.twimg.com/media/BLr_dhkCIAAWdje.jpg:large

Next to the die, there is a row of transistors, if i'm not mistaken Coollaboratory Liquid Pro is conductive. I don't want to short circuit the cpu :D ... :/

This may explain why Intel used the poor conductive thermal paste.
Waynio 9th June 2013, 00:32 Quote
Big temperature drop but also a big drop by Intel for letting the QC go AWOL forcing enthusiasts to void there warranties if they want a cooler running CPU.

I think Intel should be offering to have the effected CPU's fixed because if people try to de-lid & break it they are screwed, there must be some who have broke there CPU trying to do this. :(
John_T 10th June 2013, 14:01 Quote
Quote:
Originally Posted by Waynio
Big temperature drop but also a big drop by Intel for letting the QC go AWOL forcing enthusiasts to void there warranties if they want a cooler running CPU.

Have to agree with that. You'd expect a few degrees drop in temp by enthusiasts de-lidding and changing the contact method, but fourteen degrees? Fourteen? That's outrageous!

It's like a few years back in the GPU industry, you spend a couple of hundred quid on the latest GPU - but then the thing seemed to be cooled by a 99p heatsink and fan. The mind boggles.
Chris_Waddle 10th June 2013, 15:37 Quote
I was so tempted to get this system when I saw it a month or so ago. The thing was, my de-lidded system was already running 20degs lower under load. I'd already knackered one 3770k when de-lidding, so I wasn't going to risk another if it didn't work as well as the claiims.

If I was only just about to de-lid then I think I would give these a try. I don't really need another temp drop, but for £4, I'd use them.
DystopianDream 10th June 2013, 17:38 Quote
FYI the temperature drop due to de-lidding, or put it another way, the poor temps before delidding, have very little to do with
Quote:
the fact that the thermal paste seems to be severely lacking in thermal performance

This thread and specifically this post show that the main reason why delidding works is due to greatly reducing the gap between the CPU and the heat spreader. The Intel TIM itself is fine.
Quote:
Conclusion: The Intel stock CPU TIM is not the reason Ivy Bridge's run hot, and replacing the Intel stock CPU TIM is not the reason a delidded Ivy Bridge runs so much cooler - the benefits of delidding are entirely due to the resultant reduction in gap height between the CPU silicon die and the underside of the IHS.

and from this post where he discusses no heat-spreader
Quote:
The take home message there is replacing the CPU TIM doesn't really provide any benefit. Reducing the gap between the CPU and the IHS does. And removing the IHS entirely doesn't really provide much benefit either.

If anyone is considering de-lidding or wants to find out more, that thread is the best resource I have found, Idontcare had explored the topic exhaustively
Combatus 10th June 2013, 19:08 Quote
Some very interesting testing there!
Stanley Tweedle 11th June 2013, 13:49 Quote
Quote:
Originally Posted by DystopianDream
FYI the temperature drop due to de-lidding, or put it another way, the poor temps before delidding, have very little to do with
Quote:
the fact that the thermal paste seems to be severely lacking in thermal performance

This thread and specifically this post show that the main reason why delidding works is due to greatly reducing the gap between the CPU and the heat spreader. The Intel TIM itself is fine.
Quote:
Conclusion: The Intel stock CPU TIM is not the reason Ivy Bridge's run hot, and replacing the Intel stock CPU TIM is not the reason a delidded Ivy Bridge runs so much cooler - the benefits of delidding are entirely due to the resultant reduction in gap height between the CPU silicon die and the underside of the IHS.

and from this post where he discusses no heat-spreader
Quote:
The take home message there is replacing the CPU TIM doesn't really provide any benefit. Reducing the gap between the CPU and the IHS does. And removing the IHS entirely doesn't really provide much benefit either.

If anyone is considering de-lidding or wants to find out more, that thread is the best resource I have found, Idontcare had explored the topic exhaustively

Which basically equates to: Intel should redesign the package to get rid of the IHS. Have a top cover that doesn't cover the cpu die but fits around it and is level with it to help reduce chances of crushing. When intel release a K cpu it is with the knowledge that it will be overclocked. Fitting a metal cap over it to reduce cooling efficiency is therefore utterly stupid.
Log in

You are not logged in, please login with your forum account below. If you don't already have an account please register to start contributing.



Discuss in the forums