Thermal sensor issue i7-7700k?

[CK7]

I have a brand new build; EVERYTHING NEW. i7-7700k is running at stock speeds. I have the RAM set to XMP for DDR4-2666. Motherboard is Asus Maximus IX Hero Z270.

I have found that the i7-7700k reports a momentary (a second or less) temperature spike +25 > 35 degrees Celsius anytime a program is opened, a webpage is opened, a background app runs etc. The temperature blip cascades through the cores in random order; not the same every time. This causes my heatsink fan to constantly cycle up and down. Temperatures otherwise report as steady, normal increases. Peak temperature under Prime95 blend test is 71 degrees Celsius.

Attempted solutions:

I have re-installed my heatsink and thermal paste with no change.

I have tried to manually set my fan speed in the bios. The only setting that avoids this issue is setting the temperature / fan at a constant (and loud) 80-100%. I've tried PWM and DC mode.

I have found a few user reports elsewhere on the web, all reasoning that it's just the way it is. I don't accept that. Opening a folder or browser should not spike temps +30 degrees. Not only is the fan cycling annoying, it puts undue stress on my fan; possibly shortening its lifespan.

What's the answer, if any? RMA?

[idata]

ADRBL2 Temperatures on my 7700k system jump from low 30s (say 32-33) to 61C on the hottest core, a differential of just under 30C. Temps on the 5820k system jump from the same lows to a high of 52C on the hottest core, call it 20C. When put into context this way I think the 7700k doesn't look so bad. That 5820k chip has a TDP rating of 140W @ stock but it is actually running @ 280W (calculated using this formula: OC Wattage = TDP * ( OC MHz / Stock MHz) * ( OC Vcore / Stock Vcore )^2) = In the case of my 7700k it has a 137W TDP or about half the power of the 5820k but also in a die which is almost certainly less than half the total area, given that Haswell-E has a die size of 356mm^2 (and Kaby Lake's die size is unknown but certainly less than 200mm^2). Thermal density AKA thermal mass has a role to play here and its effects are evident in the differences exhibited by these chips in terms of thermal characteristics. Objects of larger mass are more resistant to changes in temperature, hence why the 5820k's temp delta is lower than that of the much smaller 7700k.

[AS19]

Fantastic. What are you trying to prove here? Care to explain how all of that correlates with the chip's life duration? We all know that if you up the temperature on a chip, you shorten its life. Intel has said that in the past. So are you telling me 30C spikes, every second or two, are good for the chip?

[idata]

ADRBL2 There is no correlation between the lifespan of the chip and these transient temperature spikes. Any correlation you or other users make is assumed. As ra5040 said, it is only sustained high temperatures close to Tj Max which can affect the lifespan of a CPU. Average temperature is what matters here. What you descibe as 1 second spikes are more likely to be on the order of milliseconds in terms of duration. Temperature spikes measured in thousandths of a second aren't terribly meaningful as they pertain to reliability or circuit degradation. The fact that the observed temperatures *immediately* cool back down to "normal" aka "idle" are a clue to this property of silicon-based microprocessors. If the temperature can change that quickly in either direction, it means there's not a large amount of total energy being consumed, only when taking snapshots of a moment in time does it appear that way.

Put it another way: Intel collects the equivalent of millions of hours of test data with these chips under various working conditions, including full load. OCCT/Linx are variations upon the Linpack software library which Intel engineers use to test all their chips. With millions of hours of stress testing data I am sure the results they gathered are as statistically meaningful as they need to be in order to determine whether or not the chips can hold up under whatever workload users can throw at them. The chips are going to be fine. The only thing for you, the user, to worry about is the variation in fan speed - which can be solved by adjusting your fan curve either through your motherboard's UEFI or a 3rd party fan controller.

[idata]

A reply to this alleged issue from a stand up guy from another forum which helps explain my previous post better:

"Temp spikes in OCCT are completely normal because if you look at the CPU Package Power in HWInfo Sensors you can see it jumping around, it does this on my 6700K - it's completely normal, it's the test (OCCT). I'm not sure why the OCCT graphs are included, I don't see it as any kind of meaningful information, for reason just described - the basis for the complaints isn't based on OCCT is it?

EDIT: just ran OCCT for 5 mins to double check. Individual core temperatures fluctuate by a large amount up to just over 20 degC difference (temperature fluctuations of any given core, fluctuations within itself - not compared to other cores). CPU Package temperatures fluctuate in region of 10 degC.EDIT 2: just ran Prime95(v26.6) for 5 mins as a comparison to OCCT above. Temperature fluctuations were a LOT lower: 6 degC temperature fluctuation of any given core, and 4 degree temperature fluctuation of CPU Package.That's a big difference between the two tests, OCCT creates massive temperature fluctuations, so if the complaints are just based on seeing fluctuations in OCCT, then I don't think the complaints are founded.

* As to the Intel guy who replied in this thread: he's only a community manager and obviously does not speak for Intel itself, even though it will be seen as such for equal obvious reasons. He shouldn't have said it.

* What strikes me as a local pandemic on this forum is that only here there really seem to be these reports and then only a few, with many other just jumping onto the bandwagon as well. Just saying...pandemic...

[idata]

No offense dude, but if the issue were as simple to solve as not using OCCT I don't think there would be a 600 reply thread regarding it.

[DS3]

The spikes are there, its a fact since fans ramp up and down constantly.

So not sure how one monitor or another got something to do with it.

[idata]

@Dimasub OCCT isn't a monitoring program, it is a stress-testing program. i.e. the very thing which is causing the temperatures to increase.

[DS3]

The case here is that you dont really need to stress to see the temp spikes.

The most simple actions such as opening your Chrome causes a one second spike of up to 20 degrees which causes the fans to double its rpm for only one second...

Currently i've configured my fans to work at 43% (the minimum I can) till 50 degrees celcius.

But thats just messed up that I even need to do such inefficient configuration when its all a matter of few hours of coding by Intel/Asus to create a mechanism to compensate for those spikes.

[idata]

It's obviously not just occt, considering I don't think occt was even mentioned once in the dozens of pages of this thread until I mentioned it and I wouldn't be surprised if most people posting haven't even heard of occt.

Numerous people are reporting spikes under all sorts of loads and even with little or no load on the cpu at all. But even under a heavy stress test spikes nearing 100c with adequate cooling and low voltage is ridiculous, and again, not to mention the numerous complaints of high temps under much more realistic loads.

[idata]

Have you see this?: http://www.tomshardware.com/charts/thermal-compound-charts/-3-CPU-Water-Cooling-High-Pressure,3363.html Charts, benchmarks Thermal Compound Charts, [3] CPU Water Cooling, High Pressure

It's rather surprising (well, in fact it's very surprising).

And this article is worth checking out: http://www.tomshardware.com/reviews/thermal-paste-performance-benchmark,3616-21.html It's Not As Easy As Picking A Favorite... - Thermal Paste Comparison, Part Two: 39 Products Get Tested

as is this one: http://www.tomshardware.com/reviews/thermal-paste-heat-sink-heat-spreader,3600.html Thermal Paste Comparison, Part One: Applying Grease And More

Robert

[clala1]

Hello to all me also synonymous problem.

in game Bf1 or for Honor my core package temperature and 48 ° maxis my h115 is up max 1600 Rpm .but a simple internet page raises the core package up to 57 ° in 1 second

mi config

CASE: cooler master haf x

CPU: i7 7700k

CM: Z270 MAXIMUS IX HERO BIOS 801

REFROIDISSEUR CPU: AIO corsair h115 AND TIM IS Ggrizzly conductonaut

RAM: Corsair Vengeance LPX DDR4 2 x 8 Go 3000 MHz CAS 15

CARTE G: EVGA GTX 980 Ti SC+ Gaming ACX 2.0+

PSU: 1000W Be Quie Dark Power Pro 10 80PLUS Gold

[clala1]

hello all,sorry for my english im french

im buy msi z70 gamig m7,and i7700k in February 2017,I have a big problem since 1 month when I starts the computer is bootloop befor the bios,im contact suport msi,After several exchanges he gave me a new bios,after 5 days the pc is stop secure,the MB and CPU is dead,my resseler test and confirm,He repaid me

[CJaun]

I just found out about this thread.

Noticed temp and core spiking since I bought the machine, I found it strange it would spike just moving my mouse around.

https://youtu.be/k1ABTlH7DSY ROG Maximus IX Hero Problem - YouTube

[SShah28]

Intel CPUs defiantly need a heat sink redesign. before delid it did not matter what cooler i put on top of this cpu, liquid, air they were all the same. actually it was a waste of money to put a liquid cooler on it. after applying lequid Ultra to the die my H100i AIO get so toasty I cant even contain the heat. I am thinking of going costume loop 360mm at least.I have never seen radiator actually getting so toasty. this cpu can actually boil water in the waterblock. This is so awesome, for the first time i feel my liquid cooler is actually doing something more than a regular stock heat sink and fan.Not to mention i got at least 200 MGhz more in a very dirty overclock.

[BBob7]

Agreed.

Now I feel that I should have gone with Ryzen instead of the 7700k. Sure, it may not be a great overclocker, but at least you don't hit an artificial thermal wall because they gimped the heatspreader installation.

[LCh]

Delid, Delid and ones more Delid !!! How did the great Lenin...

[AKios]

So we found the problem and its the Thermal Paste which is garbage on a CPU of 350$.

Intel should warranty Delid'd CPUs in the warranty after this. Its all Intels fault non consumer should loose the warranty of its product because the Intel cannot apply a premium paste on its products.

Im not hating im just point facts here and we should still have warranty after Delid.

Hope you are all with me

Thanks

[idata]

Well this is just what I have been thinking all the time...Intel should give us new chips with better Thermal paste OR they should delid our cpu without losing warranty!!

That is what I want!

And this is why i havent started my delid prosess because i think this is not my problem its Intel fault that their 400 euro prosessor is using crappy paste and customers are having so much trouble because of that!

[AKios]

Indeed,

Good that pointing the EURO price because we citizens of EU we are paying way more of course thats not their problem but its well mentioned.

Intels should fix all that by saying "We now provide full warranty of DELID'd CPUs because this thermal issue its OUR fault"

But which company will admit that they send a "faulty" cpu out in the market?

[LSøre]

I would call it false advertisement.

Thats at least a big deal en the EU where I live.

On the intel product pages it says.

"Let you custom tune for even higher performance"

[LCh]

I think it will be great to go to the Intel's office and "give" them as mach bubble gum as you can to help them to make there cpu.