4790k & Noctua NH-D14 Temp issues?

[ch3rn0byl97]

On my 4790k @ 4GHz (1.175v) And Noctua NH-D14 im sitting around 30-35 idle and under 98% load it spikes to 71C under 100% load for more than a moment i see it spike to 80-90, ive reseated repasted multiple times and im using Arctic silver 5. I plan on letting ti set before reseating again but im a little lost here, any ideas.

 

[Darkbreeze]

Something is wrong. I'd pull the CPU off and make 100% sure that the mounting bracket to backing plate fasteners are completely tight. It doesn't need to be hercules tight, but it needs to not be moving around, not even slightly, either.

I've seen cases where the wrong adapter was being used or the backing plate to mounting bracket fasteners were a little too long and needed to have plastic or fiber washers in between the backing plate and motherboard to lengthen the distance between the backing plate and the CPU cooler mounting adapter so that the fasteners will actually get tight rather than bottoming out in the backing plate threads.

Also, what method are you using to paste the CPU lid? How much TIM are you using? You don't need more than about two grains of cooked rice size TIM on there, and actually one really big rice grain sized amount is probably plenty. This is a different cooler, but the pasting methods are similar for pretty much all coolers, so it might give you an idea where things are going wrong. Or it might not be the problem at all. Something is definitely not right though.

You sure that's your actual CPU core voltage as set in the bios and that you don't have any desktop overclocking utilities that are changing the values?


http://www.tomshardware.com/faq/id-2520482/solving-temperature-issues-hyper-212-evo.html

 

[Karadjgne]

What are you using to get those loads? You have a Haswell cpu, pretty much everything new, and any Prime95 newer than v26.6 uses AVX instructions that can and will drive a Haswell to unrealistic amounts. This includes IBT, OCCP, Aida64, Linpack etc.

 

[ch3rn0byl97]

I was using powerMAX by CPUID, and it was using avx instructions, and i use what ive always used nothing but a small dab on the center of the cpu, That 71C temp was drawn when running the Stalker Call of Pripyat benchmark. Im using a THermal take view case, plenty of airflow. Im just lost everything is tight, this isn't my first rodeo with aftermarket cooling.

 

[Darkbreeze]

Nobody is saying it's your first rodeo, but even experienced enthusiasts and system builders make mistakes, or overlook things, all the time. So it's always worth double checking the basics. One of our LONG time builder/moderators recently put a system together and couldn't get the 6th SATA port to work. He knew, but overlooked, the fact that one of the X1 slots was stealing a PCI lane and couldn't be used with all of the devices, plus an NVME drive, and have a device in one of the extra slots. Basic, but easy to overlook. He was ready to RMA the board until he realized what the issue was.

So, not a fail on you, just saying that it happens.

I'll assume the heatsink isn't full of dust, and that you have not changed the profile for your CPU fan. Did this system previously have better cooling performance or are these new (To you anyhow) parts that you've assembled or are testing for the first time? Is the CPU cooler new to you? Sure there are no perforations in the heat pipes anywhere? This can happen. I've seen solder points fail or pinholes develop in heat pipes before and that will drastically reduce the cooling capabilities of the unit even it it's only in one pipe.

Aside from that, if you are hitting 71 degrees with the stock configuration, using a D14, and all your doing is gaming, then I agree something is wrong.

What are you using to get your thermal readings? Try CoreTemp and see if the readings are the same? Might just be your monitoring software telling lies.

 

[ch3rn0byl97]

Iver double checked things, the only think i can think of is the compound being used, i reached 68C just playing forza horizon 3. Im using HWMonitor, 0 dust in the cooler, it's looking minty. Like i said this has me baffled, ive never seen HWmonitor vary so much on cpu temp. I mean even on idle ill see it vary multiple degrees avidly.

 

[ch3rn0byl97]

Heck my gtx770 is running like 6 degrees C cooler than my prior case.

 

[Darkbreeze]

Stop using HWmonitor, and start using HWinfo or CoreTemp. At LEAST long enough to verify that what you're seeing in HWmonitor is accurate. For me, and a lot of other enthusiasts, about 60% of the time, HWmonitor is wrong or even hilariously wrong.

HWmonitor, Open hardware monitor, Realtemp, Speedfan, Windows utilities, CPU-Z and most of the bundled motherboard utilities are not always terribly accurate, invariably, and in some cases. Some are actually grossly inaccurate, especially with certain chipsets or specific sensors that for whatever reason they tend to not like or work well with. I've found HWinfo or CoreTemp to be the MOST accurate with the broadest range of chipsets and sensors. They are also almost religiously kept up to date.

CoreTemp is great for just CPU thermals including core temps or distance to TJmax on AMD platforms.

HWinfo is great for pretty much EVERYTHING, including CPU thermals, core loads, core temps, package temps, GPU sensors, HDD and SSD sensors, motherboard chipset and VRM sensor, all of it. Always select the "Sensors only" option when running HWinfo.

In cases where it is relevant and you are seeking help, then in order to help you, it's often necessary to SEE what's going on, in the event one of us can pick something out that seems out of place, or other indicators that just can't be communicated via a text only post. In these cases, posting an image of the HWinfo sensors or something else can be extremely helpful. That may not be the case in YOUR thread, but if it is then the information at the following link will show you how to do that:

*How to post images in Tom's hardware forums



Run HWinfo and look at system voltages and other sensor readings.

Monitoring temperatures, core speeds, voltages, clock ratios and other reported sensor data can often help to pick out an issue right off the bat. HWinfo is a good way to get that data and in my experience tends to be more accurate than some of the other utilities available. CPU-Z, GPU-Z and Core Temp all have their uses but HWinfo tends to have it all laid out in a more convenient fashion so you can usually see what one sensor is reporting while looking at another instead of having to flip through various tabs that have specific groupings.

After installation, run the utility and when asked, choose "sensors only". The other window options have some use but in most cases everything you need will be located in the sensors window. If you're taking screenshots to post for troubleshooting, it will most likely require taking three screenshots and scrolling down the sensors window between screenshots in order to capture them all.

It is most helpful if you can take a series of HWinfo screenshots at idle, after a cold boot to the desktop. Open HWinfo and wait for all of the Windows startup processes to complete. Usually about four or five minutes should be plenty. Take screenshots of all the HWinfo sensors.

Next, run something demanding like Prime95 version 26.6 or Heaven benchmark. Take another set of screenshots while either of those is running so we can see what the hardware is doing while under a load.

*Download HWinfo


For temperature monitoring only, I feel Core Temp is the most accurate and also offers a quick visual reference for core speed, load and CPU voltage:

*Download Core Temp

 

[ch3rn0byl97]

This is only at 4GHz 1.185V, Howmonitor is spot on to coretemp, and those max temps were reached powermax (thus its pulled up)

 

[Karadjgne]

Fan speeds?

 

[zebarjadi.raouf]

There's nothing wrong with the cooler installation or the software. It's the CPU. Notice the temp difference between the 1st and 4th core (26c). Mine did that too. You have three solutions:

1. Delid. Will outright fix it if done correctly. Temp difference between my cores is now 5c + 20c overall drop (note it's a 5y old CPU, not brand new like the ones on youtube with 5c drop). I did it personally with a razor but I recommend you get it done by a professional.

2. Underclock or undervolt.

3. Time to upgrade to a newer CPU.

 

[Darkbreeze]

Wrong. Differences between CPU core temperatures are normal and present on 99% of consumer processors. This is not the problem. You should try to not unsubstantiated statements like this with no basis in fact.

Fact is, a 4 degree difference between the lowest and highest temperature per core is fairly small or at worst, average. Many CPUs have as much as a ten degree difference. More than a ten degree difference per core, well, that might indicate an issue with the internal TIM, but at ten degrees or less that would be considered a normal variance AND could even be more relevant to a bad or uneven paste job, not necessarily a poor internal TIM problem.

 

[Karadjgne]

Um, darkbreeze, what the prior poster was referring to was the 100% utilization of cores yielding max temps. Core #0 read 82°C, same as the package temp, but temps consistently got lower per core until Core #3 read 56°C, a total difference of 26°C at a supposed 100% usage. That's a mite larger than the 10° common variance. Considering that, Core #0 & #1 will usually see the highest percentage of use from games, which will also yield the highest temps.

That said, that consistent drop in temp directly lowering per core, even under supposed 100% usage is what I find suspect. So far op has used AVX instructions, unknown quality software to get those temps. My suggestion is for Op to run a known variable, Prime95 version 26.6 SMALL fft, a proven 100% usage across all cores, and no AVX or other non-functional instructions for at least 1/2 hour and log those temps. If results do show anything close to the same as Op's previous testing, it's pretty much assured that it's a really bad TIM application and delid could be an option or a bad paste application or bad mounting of the cooler, resulting in paste squashed from one side and bunched at the other.

 

[Darkbreeze]

You are right. I appologize. I was looking at the current core temps, not the max core temps column. I hate the HWmonitor layout. My bad on that.

However, I agree, that is not a program that provides a consistent steady state load, and I said earlier that for thermal testing Prime95 version 26.6 should be used, as we all well know.

There are few people who have invested as much time and energy in the area of Intel thermal specifications, as Computronix has, and this is what he has to say about that. Though discussions with him and a few others, also based on the thermal testing instructions of various other sources, I don't think 1/2 hour is necessary for testing thermal compliance though. I've never seen any generation of AMD or Intel processor continue to rise in temperature after 15 minutes of Prime testing, so in my opinion, and that's all it is, 15 minutes has generally been shown to be the longest amount of time that I've seen any FFT length create a thermal increase in Prime Small FFT testing.

I can think of several reasons why x264 encoding or AVX / AVX2 / FMA3 apps won't work as a unilateral metric for thermal testing.

(1) A steady-state workload gives steady-state temperatures; encoding does not.

(2) Simplicity in methodology; most users would find encoding apps unfamiliar and cumbersome to accomplish a simple task.

(3) Most users such as gamers never run any apps which use AVX or FMA, so adaptive or manual voltage aside, it makes no sense to downgrade your overclock to accommodate those loads and temps unless you KNOW you will be making significant use of AVX/FMA/AVX2.

(4) Standardization; Prime95 has been around since 1996; many users are familiar with it. It is TRIED and TRUE.

For the minority of users who routinely run AVX/FMA apps, then P95 v28.5 or later can be useful for tweaking the BIOS for thermal and stability testing on THOSE types of systems only. For others, it is not recommended.


regardless of platform or architecture, Prime95 v26.6 works equally well across ALL platforms. Steady-state is the key. How can anyone extrapolate accurate core temperatures from workloads that fluctuate like a bad day on the stock market? They can't. That's why steady state is necessary for testing of thermal compliance and for baseline stability verification.

I'm aware of 5 utilities with steady-state workloads. In order of load level they are:

(1) Prime95 v26.6 - Small FFT's (Important. NOT Blend or Large FFT)
(2) HeavyLoad - Stress CPU
(3) FurMark - CPU Burner
(4) Intel Processor Diagnostic Tool - CPU Load
(5) AIDA64 - Tools - System Stability Test - Stress CPU

AIDA64's Stress CPU fails to load any overclocked / overvolted CPU to get anywhere TDP, and is therefore useless, except for giving naive users a sense of false security because their temps are so low.

HeavyLoad is the closest alternative. Temps and watts are within 3% of Small FFT's.

-Computronix

So if those core temp results at 100% steady state still look like they do in the screencap above after doing that, then I'd have to agree there is something wrong somewhere. I still think I'd be looking at some other possible issues first before saying for certain it was the CPU itself though.

Might also want to run a couple hours of Realbench to see if there are any evident errors.

 

[zebarjadi.raouf]

Wrong

Whoa, calm down. You sounded like my biology teacher.

Considering that, Core #0 & #1 will usually see the highest percentage of use from games, which will also yield the highest temps.

I thought they were stress testing temps, not games. If it was games, then you're right. It might be something else.

There's also a fourth solution I forgot to mention regarding TIM problem (if it is). Since his 1st core is running hot, he could disable it manually with affinity adjustment using task manager.

BTW, those voltages seem quite high (1.185v, VID is Vcore on HWMonitor). they should be near 1.100v or less. Turbo is disabled too.

 

[Karadjgne]

VID is not vcore don't get them confused, common mistake.
VID (Voltage Identifier) = Voltage Level "request" by the CPU to the motherboard's VR (voltage regulator) to supply it, this is initialized by the CPU and can change accordingly if the CPU is in power saving mode like C1E/EIST features, and also each CPU has an unique max VID internally that was set at factory level while they're running at full load. Turning off power saving features like C1E/EIST will overide and disable those feature and the cpu's VID will be permanently set at their max.

VCore = Actual voltage "delivered/supplied" by the mobo to the cpu, this could be automatic from the cpu as the function of C1E/EIST features ... or ... it was manually set and override by user like in OCing, and this "manually set" vcore could be higher/lower than the cpu's VID at mobo with oc-ing capability.

About the voltage reading, VID is just a state or information that is in the CPU, while vcore is the actual voltage, and as usual, software based reading on vcore is not very accurate, expecting +/-10 to 20% off for normal condition.

Many games will run 1-2 cores hard, and the other cores somewhat get a break. It's most normally cores #0 & #1 on my pc, can't see it being much different on Op's. With his test results showing both those cores as running the hottest by a good margin, any gaming he does will be upto 26°C hotter than if the game landed on core #3.

Running vcore voltages vary per cpu, even on the same family. My 3570k runs stock at 1.18v, my 3770k runs 1.2v. There's been multiple reports of i7-7700k running at 1.4v under stock settings, and require that much to maintain stability. And could still be +/- 10-20% off

 

[zebarjadi.raouf]

VID is not vcore don't get them confused, common mistake.

I know. Good explanation BTW. I'm saying VID on HWMonitor is the equivalent of VCore from BIOS or Core Voltage from CPU-Z. My VCore on HWMonitor is 1.72 which would probably fry the CPU.

Many games will run 1-2 cores hard, and the other cores somewhat get a break.

Many CPUs have 1-2 cores that run hotter usually the first core being the culprit. I suggested if he has TIM problem, he could change the affinity.

 

[Darkbreeze]

I still believe there are other possible options for fault. This might be a fairly old CPU cooler, and it might not be. It's probably worth taking a look with a magnifying glass to see if there are pinholes in any of the heat pipes or in any of the solder points on the heat pipes. I've seen this happen before on both CPU coolers and copper pipes in residential/commercial plumbing applications where there may have been a contaminant in the solder weld and over time has become a pinhole, or where a less than perfect solder job was employed to begin with.

Also, it would only take a very minor "ding" up against something in some cases to do the same thing. If any of those things were to happen and one of the heat pipes lost it's internal coolant there could easily be symptoms similar to what we're seeing with the wild variation in temps.

If after a close inspection of the cooler you can find nothing wrong, and no other solutions seem probable aside from the CPU, then it's likely this CPU has issues, but it would be pretty uncommon. CPUs that were fine off the shelf rarely fail down the road unless they've been abused somehow. Possible, but not as likely as other reasons.