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

SonDa5 said:
If Intel wants to gimp a review to help protect SB-E performance then they could have gimped the performance with the TIM. To me it seems intentional. Intel is at the top of their game and they know very well how using TIM instead of fluxless solder would effect overall performance.
Click to expand...

If we can get the IHS successfully, get a cooler that makes great contact, put some good TIM between the core and the HS/F I think we might see some solid clocks after all if this is true.
 
beto said:
The thing I can't figure out is why Intel would go with something so inferior (thermal grease - if it is indeed vastly inferior to solder) when the performance of the chip would be so directly effected and thus the reception from consumers would likely be bad. I mean, it can't cost that much more if they were using solder on SB right?
Click to expand...



If TIM is indeed used on retail IB then consider it intentionally done to protect SB-E sales. Why buy the more expensive SB-E when you could buy IB that can over clock close to 6 GHZ on air? (Just for example)


Why gimp down a product? To protect another more expensive product.


That is one reason and the other is to save money.

I have no doubt that putting "secret sauce" TIM on the IHS will save alof of money over using fluxless solder.

I really hope Intel didn't use TIM on the retail chips.
 
Last edited:
SonDa5 said:
If TIM is indeed used on retail IB then consider it intentionally done to protect SB-E sales. Why buy the more expensive SB-E when you could buy SB-E that can over clock close to 6 GHZ on air? (Just for example)


Why gimp down a product? To protect another more expensive product.


That is one reason and the other is to save money.

I have no doubt that putting "secret sauce" TIM on the IHS will save alof of money over using fluxless solder.

I really hope Intel didn't use TIM on the retail chips.
Click to expand...

I really really doubt the TIM issue would stop Ivy from going to 6 ghz on air. Just a hunch. ;) But I do think it's interfering somewhat.
 
ShuttleLuv said:
If we can get the IHS successfully, get a cooler that makes great contact, put some good TIM between the core and the HS/F I think we might see some solid clocks after all if this is true.
Click to expand...



Yes but it may void the warranty and I doubt any heat sink builder is going to design a heat sink with the perfect spacing to correctly fit a IB cpu with the IHS removed. It is a small market and then there is the risk of the end user destroying their cpu if not properly seated.


A thin large piece of copper might work to replace IHS which could work with any regular S1155 heat sink.
 
My Hyper 212 is screw down so I'm positive it could be screwed down a bit and make good contact if jerry rigged.
 
I suspected this might be one of the two possible causes (IHS), but the enginerds on a very technical board believe the problem is more likely hot spot heat dissipation from the shrink. No doubt that the IHS could be improved, and that's been true since Intel moved (back to) an IHS.
 
SonDa5 said:
Yes but it may void the warranty and I doubt any heat sink builder is going to design a heat sink with the perfect spacing. It is a small market and then there is the risk of the end user destroying there cpu if not properly seated.


A thin large piece of copper might work to replace IHS which could work with any regular S1155 heat sink.
Click to expand...

Yes it would void the warranty but honestly, we're enthusiests. What the fuck is a warranty??? LOL. ;) I remember popping the IHS off of alot of intel and amd cpu's, good fun. :cool:
 
http://www.overclock.net/t/305443/ihs-removals-how-to-do-it-should-i-do-it-and-the-facts

Look at this old article and the quote from it:


(1.4) So those CPUs that are soldered to the IHS will have better temps than those that use TIM as a filler?

Generally yes, CPUs with TIM between the die and IHS have poor temps. Lets use the M0 stepping Allendale cores for example, once these hit 1.35v or higher they start having massive heat problems while many Conroe core CPUs using even more voltage at a higher speed with the same heatsink have temps over 10c cooler than you do. That is the difference between a soldered IHS and a non soldered IHS, the heat from the die gets trapped around the die rather than transfering straight into the IHS. Now for the list of CPUs that are soldered and those that aren't, let me remind you this list is constantly under updation as more CPUs are released or as corrections are made.
Click to expand...
 
I'm half inclined to think that people are trying to grasp at straws...
 
Did you read my link man? Sounds just like Ivy. Once you start hitting 1.35 you get skyrocketing temps. Sounds like it could be the TIM or a big part of it at least.
 
TIM would make a huge difference... 6 W/mK at best versus like 80 W/mK with the better solders. Maybe they got around 15 W/mK if they used some kind of thermal epoxy, but still... WTF was Intel thinking if this was true. It also makes the IHS completely useless...
 
RealityCrunch said:
Ohhh... Very tempted to buy IB and remove the IHS now. Love me some core contact cooling.
Click to expand...


Make sure you get the spreader with perfect contact to the HS/F though cause sometimes it looks like it's making contact but it's really not.
 
RealityCrunch said:
Ohhh... Very tempted to buy IB and remove the IHS now. Love me some core contact cooling.
Click to expand...



Wait to see if the retail IB cpus have TIM on them. Could be that only the engineering sample
and review IB cpus had TIM on them.


Also I am not sure that core contact is better than IHS with proper fluxless solder.
 
Yeah, I'll be waiting for confirmation. It can be slightly better, though never enough to warrant removing a soldered IHS.
 
SonDa5 said:
If TIM is indeed used on retail IB then consider it intentionally done to protect SB-E sales. Why buy the more expensive SB-E when you could buy IB that can over clock close to 6 GHZ on air? (Just for example)
Click to expand...

Only problem with that logic is that Sandy Bridge itself is killing SB-E sales. For almost all users, SB is every bit as fast as SB-E.
 
Forceman said:
Only problem with that logic is that Sandy Bridge itself is killing SB-E sales. For almost all users, SB is every bit as fast as SB-E.
Click to expand...



IB could be fixing that "problem". I imagine IB will be around longer than SB.
 
Tsumi said:
I'm half inclined to think that people are trying to grasp at straws...
Click to expand...

I had the same thought, but then again we have seen odder things done in the industry. It will be interesting to see some of the more 'extreme' review sites over the next weeks take a stab at this issue and see a) if it is a retail vs ES difference, and b) whether they can come up with a direct-die hs to verify how much of an effect there is from the tim. <cough> looking at you [H] - show us some ivy bridge chips hacked apart!
 
I wonder what the cost difference is in the two methods. You wouldn't think they'd bother over a couple of cents a chip.
 
So 4.5ghz on 3770k is doable on air, even if uses TIM??

I personally don't need 5ghz. Using a hyper 212 plus right now. Bought a 2700k on Monday planning on exchanging it for a 3770k this coming Sunday. Guess I'll wait a few more days to hear reports on the TIM. The 15 day exchange/return policy expires on the 2700k eight days after IB release which I think gives me enough of a window to make a decision on this TIM issue. That said, I would like the IB chip for the additional features and am willing to sacrifice a couple hundred mhz for it.
 
LittleMeEgo said:
So 4.5ghz on 3770k is doable on air, even if uses TIM??

I personally don't need 5ghz. Using a hyper 212 plus right now. Bought a 2700k on Monday planning on exchanging it for a 3770k this coming Sunday. Guess I'll wait a few more days to hear reports on the TIM. The 15 day exchange/return policy expires on the 2700k eight days after IB release which I think gives me enough of a window to make a decision on this TIM issue. That said, I would like the IB chip for the additional features and am willing to sacrifice a couple hundred mhz for it.
Click to expand...

Looks like most review sites were able to get to 4.4/4.5 without raising the voltage much (if at all), so that should be okay. It gets bad pretty quickly after that though, at least so far.
 
Forceman said:
I wonder what the cost difference is in the two methods. You wouldn't think they'd bother over a couple of cents a chip.
Click to expand...

I read somewhere that the fluxless solder uses some expensive materials to include a very thin layer of gold. Definitely not cheap but works great.
 
You know, for those of us who don't plan on going to 17.7THz on IB, this isn't going to be much of a problem. I'm not going to overclock much using voltage, and I'm neurotic about cooling to the point of choosing my cases and fans based on fluid dynamic testing, so I think the TIM isn't going to be much of an issue for me and other people who might only take it up to 4.5GHz or so. If you're going to try to open up gateways to other dimensions with your CPU, then go Sandy Bridge.

edit: That being said, if it turns out that retail IBs are full of ceramic paste, I'll personally be taking the spreader off mine, removing all the paste, and filling the bastard with AS5.
 
Last edited:
Red Machine D said:
You know, for those of us who don't plan on going to 17.7THz on IB, this isn't going to be much of a problem. I'm not going to overclock much using voltage, and I'm neurotic about cooling to the point of choosing my cases and fans based on fluid dynamic testing, so I think the TIM isn't going to be much of an issue for me and other people who might only take it up to 4.5GHz or so. If you're going to try to open up gateways to other dimensions with your CPU, then go Sandy Bridge.

edit: That being said, if it turns out that retail IBs are full of ceramic paste, I'll personally be taking the spreader off mine, removing all the paste, and filling the bastard with AS5.
Click to expand...

At least use something better than that outdated crap. Like IC Diamond :p
 
Tsumi said:
At least use something better than that outdated crap. Like IC Diamond :p
Click to expand...

You mean to tell me that the industry standard has been replaced in the three years I've been soaking in complacency with the computer I built? LIES AND BLADERDASH!

I'm assuming that IC Diamond is the same concept, just conducts heat better, aye?
 
AS5 might have been the industry standard back in the day, but it has been surpassed.

The most popular of which is the MX-2. As I recall, it has better conductivity than AS5, costs less, and has no cure time. Not to mention it is completely nonconductive, which is not the case with AS5.

As for IC Diamond, it is actually fairly different from AS5. As I read it, it basically uses microscopic diamonds to cut into the surfaces when they are pressed together, creating an essentially completely gapless seal. It is notoriously difficult to separate the heatsink and CPU after they are attached, not to mention costing significantly more. The best today is Indigo Extreme. But yes, the basic concepts behind TIM is still the same, but technology has significantly improved.

Here is a nice review of TIMs on the market. Not complete yet, still a work in progress, but it gives you an idea of where AS5 stands in relation with the competition. Quite frankly, the only reason why it still sells is because people are stuck in the mindset that AS5 is still the best.
 
Tsumi said:
The most popular of which is the MX-2. As I recall, it has better conductivity than AS5, costs less, and has no cure time. Not to mention it is completely nonconductive, which is not the case with AS5.

As for IC Diamond, it is actually fairly different from AS5. As I read it, it basically uses microscopic diamonds to cut into the surfaces when they are pressed together, creating an essentially completely gapless seal. It is notoriously difficult to separate the heatsink and CPU after they are attached, not to mention costing significantly more. The best today is Indigo Extreme. But yes, the basic concepts behind TIM is still the same, but technology has significantly improved.
Click to expand...

Tell me about this Indigo Extreme. I'm lookin around the net about it, and it looks the shit, but I'm having a hard time wrapping my head around how you actually apply it.
 
Red Machine D said:
Tell me about this Indigo Extreme. I'm lookin around the net about it, and it looks the shit, but I'm having a hard time wrapping my head around how you actually apply it.
Click to expand...


It goes on like double sided tape. You stick it to your cpu, and then put your heatsink on it. Then you boot up with your heatsink fan off to cook it onto your cpu. Its safe for intel because they have thermal protection that keeps em from burning out.
 
This is not true, the paste instead of the solder isn't causing the high temperatures.
korrrhonen, or better known as SF3DOC, the Finnish extreme overclocker delidded his IB..
http://www.sf3d.fi/forum/viewtopic.php?p=410#p410
http://murobbs.plaza.fi/1708625462-post736.html

In his own forum with the two pictures:
"I changed the paste between the IHS and die, but the IHS HAS to be in place, or else the brittle processor pcb does not make proper contact with the socket pins. This operation helped a little, now it doesn't throttle so easily near 4.9Ghz. I got +50Mhz, so there is NO BENEFIT. So, DO NOT destroy your cpu's."

And in MuroBBS he posted:
"Like you might have guess, the IHS removal or changing TIM didn't help.
*link to his forum*
The overheating near 5Ghz is just simply because of that architecture, it is not because the IHS and die would have a bad contact. The temperature jumped to 105*C just like before.
So all it needs is extreme cooling and presto the clock scaling continues."
 
The discussion is on TIM vs solder. All the people did in your's was change the TIM which will obviously not help much, if at all. So I don't see how you can say this discussion is not true.
 
Anybody know if the 2011 chips i7 3820K, 3930 ect are soldered or timmed ?
 
I wonder if you just remove the socket locking mechanism and then just use the heatsink or water block to clamp it down. Right now the retention bracket requires the tabs on the IHS to maintain proper pressure.

This is why I'm skipping 1155 and just goin direct to Haswell. Hopefully the process will be mature by then, and hopefully AMD will be able to pose a threat so Intel will be forced to put out something awesome. right now, Intel can lollygag and pussyfoot around all they want. It doesn't matter.
 
Reminds me of the days of trying to mount HSF's on AMD Thunderbird cpus and hoping to god that I wasn't about to crack the core.
 
thesecond said:
It goes on like double sided tape. You stick it to your cpu, and then put your heatsink on it. Then you boot up with your heatsink fan off to cook it onto your cpu. Its safe for intel because they have thermal protection that keeps em from burning out.
Click to expand...

Install heatsink, hit with torch? :D
 
If you are going to remove the IHS, why not solder it back on instead of just replacing the thermal paste?

1) Take off IHS
2) Heat IHS with a propane torch
3) Apply solder to IHS while keeping it hot with the propane torch
4) Remove heat of the propane torch and quickly align the Ivy Bridge CPU on the IHS
5) Solder will cool fusing the IHS and the die together

Might want to create a jig for aligning the IHS and the CPU, but it should be doable. Who wants to be the first to try? :D
 
siliconnerd said:
If you are going to remove the IHS, why not solder it back on instead of just replacing the thermal paste?

1) Take off IHS
2) Heat IHS with a propane torch
3) Apply solder to IHS while keeping it hot with the propane torch
4) Remove heat of the propane torch and quickly align the Ivy Bridge CPU on the IHS
5) Solder will cool fusing the IHS and the die together

Might want to create a jig for aligning the IHS and the CPU, but it should be doable. Who wants to be the first to try? :D
Click to expand...
Wouldnt it be safer to just bake it in a toaster oven for a certain period of time? I am thinking maybe Julia Child has a recipe for Silicon au Gratin that might be easily followed.
 
You must log in or register to reply here.
Back
Top