Saturday, May 12th 2012

TIM is Behind Ivy Bridge Temperatures After All

It's proven: the thermal interface material (TIM) used by Intel, inside the integrated heatspreader (IHS) of its Core "Ivy Bridge" processors are behind its higher than expected load temperatures. This assertion was first made in late-April by an Overclockers.com report, and was recently put to test by Japanese tech portal PC Watch, in which an investigator carefully removed the IHS of a Core i7-3770K processor, removed the included TIM and binding grease, and replaced them with a pair of aftermarket performance TIMs, such as OCZ Freeze and Coolaboratory Liquid Pro.

PC Watch findings show that swapping the TIM, if done right, can shave stock clock (3.5 GHz, Auto voltage) temperatures by as much as 18% (lowest temperatures by the Coolaboratory TIM), and 4.00 GHz @ 1.2V temperatures by as much as 23% (again, lowest temperatures on the Coolaboratory TIM). The change in TIM was also change the overclockability of the chip, which was then able to sustain higher core voltages to facilitate higher core clock speeds. The report concluded that Intel's decision to use thermal paste inside the IHS of its Ivy Bridge chips, instead of fluxless solder, poses a very real impact on temperatures and overclockability.

Source: PC Watch
Add your own comment

219 Comments on TIM is Behind Ivy Bridge Temperatures After All

#1
n-ster
badtaylorx said:
oh realy....has intel or amd EVER warranted physical damage???
your point? I don't cover fans in the warranty I give when I sell my custom builds, but I still put a fan grill if needed so that exposed fans are protected from wandering fingers, pens etc

Why woulld I want the customer to think I product is weak and not built well? Why would I want my customers to break their new toy and get frustrated when told it isn't covered?

your point is idiotic
Posted on Reply
#2
Wile E
Power User
badtaylorx said:
troll...nice try
i know you're a modderator but http://lmgtfy.com/?q=troll

this is on topic and fairly unemotional.....
not a troll

oh well i suppose you can lead a horse to watter
How many cpus have you run delidded?
Posted on Reply
#3
badtaylorx
one (c2d e6400)

and no the point was not idiotic. the point was that cpu companies DO NOT put an "internal heat spreader" on a die to protect the die from customers crushing them. they'd be called a protective shim or something like that. to call somebody an idiot because of ignorance.....now thats troll'n!!!

ive been reading through some ihs american patent aps today and one thing that comes up often is the term "hotspots"...certain parts of the die get much hotter than others and there is a need to have a medium that evens (spreads) the heat out.
Posted on Reply
#4
theoneandonlymrk
badtaylorx said:
but only uses a shim arround their (cooler) gpu chip???
because hardly anyone bar the manufacturer touches the shroud and hence gpu but , the cpu is almost allways handled in some way and built by man is built by man and Amd use a bigger IHS as i said as a save a chip measure as light tim and HS contact on said bare chip with load would pop it ,but with said IHS it wont pop( most of time it will shutdown) in fact read my last post again would yah i mentioned low refit chips like NBs and Southbridges Gdamn it:p:confused:

badtaylorx said:
ive been reading through some ihs american patent aps today and one thing that comes up often is the term "hotspots"...certain parts of the die get much hotter than others and there is a need to have a medium that evens (spreads) the heat out.
as does a well fitted HS on a bare chip

Why do you think its called an Integrated HEAT sink, its integrated and just like a mounted HS its a Heatsink by your reckoning i should maybe get a solid water block with side ports and then mount a third fan heatsink on top ot the WB and IHS to increase again my thermal POwers

3s better then two right, two's better then one and maybe ill fit some wheels on the outside of my cars wheels so ig goes faster , cos itll grip better:slap:
Posted on Reply
#5
badtaylorx
how would adding another hsf on top of the waterblock NOT help as long as it has access to ambient air???

it's good there is oneandonly of you....you keep slipping into this land of make believe stuff..

what you're describing is not too far from a coolermaster v10 though(worst purchase ever btw)

just a guess oneandonly.....you think 94octane gas burns hotter and faster than 87 dont you....
Posted on Reply
#6
theoneandonlymrk
badtaylorx said:
what you're describing is not too far from a coolermaster v10 though(worst purchase ever btw)
i appreciate you auto answering your self , its worse because the fan HSF would possibly pull heat into the water in a poorly vented ,rad front case but wouldnt really cool any better anyway and wtf , i was being sarcastic:p:D

there is only one of me :cool: , their are many like me and a many better then me but thats not you on the topic of thermodynamics apparently, and im an engineer not a phycisist or anything all that technical, just good old hands on tried it type inteligence with enough maths to keep my brain working, if you can carry away more heat you remove more heat and the discrepencies you speak of between an enthusiast fitting his own lid correctly verse the same plus intel fitting one inbetween a bit shit dont exist.

badtaylorx said:
just a guess oneandonly.....you think 94octane gas burns hotter and faster than 87 dont you....
whats this mock nameing me shit lmao

i think you should read this then as it clearly states im no chemist and since the only contact i have with fuel is putting the overpriced shit in my car why would i care which is better i put the soddin cheapest in obviously ,, helo engineer not phycisist

im fully aware of hotspots onchip btw ,did you know intel and ibm are messing with on die peltiers , micro peltiers in silicone
Posted on Reply
#7
badtaylorx
you brought cars into this. not me

no.....article??? thats interesting
Posted on Reply
#8
theoneandonlymrk
badtaylorx said:
you brought cars into this. not me

no.....article??? thats interesting
dude the cars bit was sarcasm but fair enough, i Read about that ages ago so sorry no link , il try my googlefoo an update if lucky.:)

http://www.technologyreview.com/computing/22016/

im not hunting all night but theres one
Posted on Reply
#11
cadaveca
My name is Dave
erocker said:
Interesting... So what are the temperature differences? Higher, lower?
I'll post screenshots tonight. I mentionedin this thread or another I thought there were reported wrong, but haven't ahd a chip to comfirm. WIll have 3570k in a couple of hours. I am nearly 99% sure temps are reported higher than they are, and many people that think it's running hot, are jsut using the wrong software.
Posted on Reply
#12
Wrigleyvillain
PTFO or GTFO
Relying on any software for such inherently sucks...
Posted on Reply
#13
n-ster
damn if its software I was baited into buying SB-E :D I don't exactly regret it though
Posted on Reply
#14
RAJOD
That is another way of looking at it. Direct contact with the core is the best. But OMG so many people are going to crack off the corner of the chip. That was an issue in the past and one of the reasons they went with a soldered IHS.
Saved many cpus.



DualAmdMP said:
I think you are wrong about Intel. I'm actually glad that they used TIM instead of solder on the core. You need to try to put a waterblock directly on the core (once in your life) and you will be surprised how well it conducts heat into water compare to IHS + waterblock. XS forum has a lot of info on this topic.

I has few Athlon X2 (65nm/90nm) running with waterblock directly on the Core. No matter how much voltage i pumped into the CPU, the temperatures were always great!

I hope AMD doesthe same thing with their new FX line.

If you don't care about watercooling the CPU for max overclock or low temperature, then just ignore me:D
Posted on Reply
#15
cadaveca
My name is Dave
cadaveca said:
I'll post screenshots tonight. I mentionedin this thread or another I thought there were reported wrong, but haven't ahd a chip to comfirm. WIll have 3570k in a couple of hours. I am nearly 99% sure temps are reported higher than they are, and many people that think it's running hot, are jsut using the wrong software.
So...





Actual voltage is 1.128 V (via digital multi-meter).

Relatively speaking, temps could be considered hot, but given clocks, it doesn't seem so. Same thing as SB for me, but with lower voltage. FAR lower voltage. The i7 3820 is running similar temps @ 4.5 GHz, but with 1.35 V.


Can't say there's much reason to complain...?
Posted on Reply
#16
Wile E
Power User
Now delid it and run it again to put the argument to bed once and for all. lol.
Posted on Reply
#17
cadaveca
My name is Dave
Wile E said:
Now delid it and run it again to put the argument to bed once and for all. lol.
Meh. What argument? It's undeniable temps seem high, but whther thats software, the TIM, or what, doesn't really matter to me when temps are at those levels.

Sure, you wanna OC to the max, pull the lid, slap on some coolaboratory pro, and get what you want...


If ya want a warranty, live with it? Not gonna matter what's in between in that case, is it? If you wanna clock within warranty, spend tha extra $25, and forget about it? That'll be MY option...


;)

:roll:
Posted on Reply
#18
Wile E
Power User
No, I mean the guy saying that removing the lid results in HIGHER temps than putting it back on with good tim.

Oh come on, take one for the team. lol
Posted on Reply
#19
eidairaman1
Wile E said:
No, I mean the guy saying that removing the lid results in HIGHER temps than putting it back on with good tim.

Oh come on, take one for the team. lol
Of Course only time Higher temps occur is if the Die doesnt make full contact with the Heatsink or left the TIM/Thermal Compound off.

I look at it this way

Regular Joes will never remove the IHS- Works fine for them

Ones who are overclockers- remove the IHS and put a better TIM on and just use the IHS cuz they feel they need it to ensure they dont damage the CPU

Extreme overclockers- Understand the full risk of direct HS contact and crushing of core- have money to back it up
Posted on Reply
#20
MetalRacer
Xzibit said:
If this is true. This is an interesting twist to the Temp issue.

http://www.hardocp.com/news/2012/05/18/pop_top_on_your_i73770k_for_better_temps63

Guess Ivy-bridge doesnt report temps like in the past according to ASUS. Monitoring software needs to be compatible.
Before Ivy Bridge was released I ask uncleweb if RealTemp was reading temps correctly and this was his reply: http://www.techpowerup.com/forums/showpost.php?p=2601104&postcount=993
Posted on Reply
#21
cadaveca
My name is Dave
MetalRacer said:
Before Ivy Bridge was released I ask uncleweb if RealTemp was reading temps correctly and this was his reply: http://www.techpowerup.com/forums/showpost.php?p=2601104&postcount=993
There are differences between CoreTemp, Realtemp, and AIDA, as "stated by ASUS":




Who is right? I dunno. What I do know is that the reported 11c temp on the 4th core is impossible, and AIDA does not report that low, although overall, AIDA reports lower temps, and as the temps increase, that difference of max temps widens...

Wile E said:
No, I mean the guy saying that removing the lid results in HIGHER temps than putting it back on with good tim.

Oh come on, take one for the team. lol
You and I both know that how the TIM is applied, what TIM is used, and a myriad of other factors can each play their role. Because removing the IHS requires removing the retention bracket, and I'll be using this chip for reviews, I will not remove the IHS, but, if the other chip I am expecting shows up, I will definitely pop the top on that one, and see what's what. I cannot do so until i have a backup chip.
Posted on Reply
#22
Wile E
Power User
cadaveca said:
There are differences between CoreTemp, Realtemp, and AIDA, as "stated by ASUS":

http://www.techpowerup.com/forums/attachment.php?attachmentid=47176&stc=1&d=1337529624


Who is right? I dunno. What I do know is that the reported 11c temp on the 4th core is impossible, and AIDA does not report that low, although overall, AIDA reports lower temps, and as the temps increase, that difference of max temps widens...



You and I both know that how the TIM is applied, what TIM is used, and a myriad of other factors can each play their role. Because removing the IHS requires removing the retention bracket, and I'll be using this chip for reviews, I will not remove the IHS, but, if the other chip I am expecting shows up, I will definitely pop the top on that one, and see what's what. I cannot do so until i have a backup chip.
I look forward to results if you do get that backup chip. I want to see some more examples of how much it helps (or doesn't).
Posted on Reply
#23
Laurijan
I am currently running a i5-3570K with a Noctua NH-D14 cooler at stock bios settings.
I get about 25C-32C idle and 57C-61C with prime95 max heat test.
Is that good for an Ivy?

Edit: I didnt pop the IHS
Posted on Reply
#24
beck24
The problem with IVY is that it as relatively poor over clocker, not that its bad at stock speeds.
Posted on Reply
#25
cadaveca
My name is Dave
beck24 said:
The problem with IVY is that it as relatively poor over clocker, not that its bad at stock speeds.
I dunno, i hit 4.6 GHz with jsut under 1.2 V, and the CPU pulls ~ 100 W.

That is a 33% increase in speed(from 3.5 GHz), for a 100% increase in current drawn.


Hrm...:wtf:...my CPU pulls just 50 W at stock...I guess you are very much right. However, at the same time, I'm using just under 1.2 V for that 4.6 GHz, when most are using jsut under 1.3 V that I have seen.

That's when comparing to SB, which ran 4.6 GHz @ 125 W, and increase of ~50% from stock(which was ~ 80 W), for the same roughly 33% increase in clocks...


:confused:
Posted on Reply
Add your own comment