How best proceed with overheating i7-4790K?

[REnso1]

I have an i7-4790K in a GA-Z97MX Gaming 5 mobo latest F4 BIOS.

I used a Noctua NH-L12 cooler rated at 95W for cooling the processor rated at 88W in a Lian Li PC V354 with 4 fans, 2 in 2 out, case closed and case open, ambient is 27-30°C.

Temperatures in BIOS and memtest86+ were high so I decided to try stress testing and in Prime95 small FFT cores 1&2 overheated to 100°C using Core Temp.

I tried reseating the heatsink and renewing the NT-H1 TIM and opening the case but it made no difference. I have a photo of the contact pattern here.

When I tested using the OCCT benchmark I was unable to complete a test due to the processor overheating so I underclocked the processor to 3.6 GHz, disabled turbo and manually set vCore to 1.1v.

With an underclocked processor I was able to get a heating and cooling curve using the OCCT auto capture, to enable me to study the problem.

Even when underclocked the processor was reaching high temperatures, rapid fluctuations in temperature with work load suggest a bottleneck in the thermal pathway. When I tested with the intel retail cooler which came with the CPU the cooling was much less effective than the NH-L12 (even when underclocked taking just over a minute of OCCT to reach the 85°C cut off point see below) indicating the NH-L12 was doing a good job of removing heat, which meant the processor was making the heat or the source of the bottleneck.

I have discussed it http://forums.hexus.net/cpus/327593-4790k-overheating-nh-l12.html elsewhere. Advice was to contact Intel due to an absence of information relating to my retailer's testing procedures. I have asked about these but am still waiting for a reply.

So my question is how should I proceed from here? Does this qualify for an RMA? If so is it possible to negotiate this with Intel direct or do I have to go through my retailer?

I have done my best to make sure I am not doing anything wrong and I would be grateful for any pointers to any mistakes I may be making.

[Allan_J_Intel1]

Thanks for joining the processor community.

I understand you have an overheating issue with your processor model i7-4790k.

If your processor has reached such high temperatures, this means that could be a hardware problem, BIOS issue or even improper installation of the heatsink.

At this point, I recommend checking some tips about processor overheating at this URL: http://www.intel.com/support/processors/sb/CS-029427.htm Processors — What do I do if my computer is overheating?

If you are still experiencing same problem, you may need to contact Intel support via chat at http://www.intel.com/chat Contact Support and choose processor area that way you could check warranty options for your processor.

Allan.

[REnso1]

Thanks for your reply Allan. Sound advice. Though I have already taken time to check these potential causes its always worth thinking about anything I might have missed. The BIOS is latest F4 version and I reseated the two heatsinks involved several times each. This appears to be a genuine hardware fault but it is currently unknown whether it is in the Mobo or the CPU. Hopefully we can find that out.

I have an update to add which is that after starting an online chat with my retailer I was able to perform a test using AIDA64 and sent them the output below which resulted in their recommendation that I send the CPU and motherboard to them for testing. You can just about read below that the throttling was at 19% with Intel stock HS/F with the mobo outside the case on a mobo tray. Core Temp recorded 3 of the cores reaching 100°C, expected temps were around 60°C. I reseated the Intel HS three times and it made no difference so CPU and mobo will be going to them on Monday. Hopefully they will replicate and confirm the fault so I won't get charged for the carriage and assessment ! Then hopefully they might find some answers and if they do I will relay them here.

[Allan_J_Intel1]

Thanks for the information.

I have informed this overheating problem to our engineering department and they would to fix this matter as soon as possible.

They would like to know if possible, check on your BIOS screen for Microcode information. If there is any BIOS screen that shows the processor microcode, please post some pictures of it.

Thanks

Allan.

[REnso1]

Thanks for replying and relaying the problem to the engineering dept.

Unfortunately I sent the CPU and motherboard to my retailer (SCAN) today so I cannot check the microcode identifier myself but I can give you the CPU package ID code which might enable some kind of diagnosis.

package

SR219 MALAY L418C133 1718

I have not changed the microcode so it is presumably the microcode it shipped with, though there was a problem with crossed pins in the socket initially (rectified by replacing the motherboard), I don't know if that could have had an effect. I have alerted SCAN to this thread and the request for microcode information so will relay any if it becomes available.

[REnso1]

Update : My retailer Scan tested my CPU and the motherboard together and found an an apparent overheating fault resulting in throttling when they are paired.

They also tested an I7 4770K with the same motherboard and it also appears to overheat, suggesting a motherboard issue. The test used was AIDA64 as above.

The motherboard is a Gigabyte Z97MX-Gaming 5 and it has the latest F4 bios which was installed by Gigabyte themselves (I previously sent it off to them for testing for other reasons).

The nature of the fault is currently unknown.

Product description GByte Z97MX-Gaming 5 MoBo

Fault found Yes

Scan diagnosis overheating CPUs. i7-4790k and i7-4770 both throttle under AIDA stab test.

[REnso1]

I have received the new mobo and old CPU back from Scan and the CPU is still overheating just as it ever was.

I was at a loss to understand how this could happen so I experimented with the memory settings as these were one of the things which may have been different at Scan's end, the results were significant,

-With memory at 1333 and 1.5v the AIDA test could be run with stock Intel HS and no thermal throttling, CPU temps reaching a maximum of 94°C.

-With memory at 1600 and 1.5v the AIDA test could not be run with stock Intel HS without thermal throttling. 17%

-With memory at 1333 and 1.65v the AIDA test could not be run with stock Intel HS without thermal throttling. 3%

-With memory at 2400 and 1.65v (XMP) the AIDA test could not be run with stock Intel HS without thermal throttling. 17%

Since Intel's http://ark.intel.com/products/80807/Intel-Core-i7-4790K-Processor-8M-Cache-up-to-4_40-GHz# @specifications memory specification for the CPU is 1333/1600, technically speaking, it still seems to be overheating.

I have been able to inspect the CPU in BIOS for ID and update revision numbers.

ID# : 000306C3

Update revision: 00000019

Does that need updating? Or should this CPU be replaced?

[cpalm3]

I'm having the same overheating issues with the i7-4790k using a Gigabyte motherboard GA-Z97X-SLI. Is there a problem with these processors with Gigabyte motherboards or just the processor itself?

[REnso1]

I guess that is the other option chilipalm, which should be considered. I was just hoping someone might already know what the problem is but it looks like I have acquired either a one off perfect storm of faulty hardware or a cutting edge compatibility problem which has not been recognised and fixed yet for one reason and another.

The previous mobo also made a 4770K overheat at SCAN's test center so they replaced it yet the fault remains. So either the CPU by coincidence also has a fault or there is a generic repeatable fault with the mobo and BIOS with the high end CPUs or a compatibility issue which requires both mobo and CPU to be updated. Logic suggests its more likely to be a mobo fault if it also makes the long established 4770K model overheat repeatably but I guess we have to rule out other possibilities to get to that conclusion.

So I will have to contact SCAN tomorrow and ask them if they tested the CPU on another mobo and whether they tested the replacement mobo with the CPU after the BIOS was updated, then we need to establish whether this is a generic fault with all these mobos in which case I need to take it to Gigabyte support.

EDIT Update, since posting this I have contacted Gigabyte support to ask if they are aware of this overheating as a generic issue and whether they are working on a fix. I am waiting for their response.

[cpalm3]

Thanks for the update boolybooly. An update for me, I updated my BIOS to F5 for my mobo and found cpu usage has decreased a little bit resulting in temps lowered by maybe 10-15 degrees celcius. However, should cpu usage increase beyond say 35%, temps increase past 80 going into 90s degrees celsius. I see that we are the guinea pigs on this new processor. Very poor testing standards by Intel and Gigabyte.

[LPala6]

Same problem with ASUS z97-A!

[BKauf]

I also have a gigabyte board, and seem to be having as many issues as these guys.

I have a 4790k on a Gigabyte z97x-UD5 with a corsair H100i as a top exhaust in a NZXT h440, using the stock thermal compound on the cooler. I had even higher temps when I used the intel stock cooler. All drivers/bios are up to date.

My peak temps are reaching 100C across all 4 cores, and apparently throttling down my speed to stay there vCore never breaks 1.17 either, and throttling is between 0 and ~40%

[REnso1]

That is interesting Broxh, you would expect better results with a hydro cooler like the H100i. Have you contacted Gigabyte support about it? I am still waiting for their reply but they usually take a couple of days.

Your overheating looks comparable to the kind of problems I am getting, but there is no knowing if they have the same cause at this point. We really need to see if the fault is reproducible on other systems using the same components. My replacement mobo overheating in the same way as the original might seem to indicate that but the CPU is the same one, so it might also have been faulty in the first place and that needs to be ruled out. Also its just possible I got two faulty mobos in a row. I have yet to get confirmation from SCAN that the CPU was tested independantly but the RMA report states it had no fault which suggests it was. However I have no info on whether the replacement mobo was tested as SCAN CS are offline, I phoned but apparently they are busy after the bank holiday etc. I have a http://forums.scan.co.uk/showthread.php?3341-4790K-overheating-with-NH-L12&p=8433&viewfull=1# post8433 support thread on their site so I will wait for them to get get back to me via that.

In the meantime I have done some more testing with a better cooler on the replacement mobo. I wrote it up on http://forums.hexus.net/cpus/327593-4790k-overheating-nh-l12-3.html# post3360041 Hexus.net as they are affiliated with the retailer SCAN. Quote below. In short the Noctua cools better, proving that it is working properly, but the CPU still overheats to the point it shuts down the system, which isn't so good.

I tried testing with the Noctua NH-L12 again to see if better cooling would make a difference. I ran the tests at stock + XMP settings with 32Gb of ram (DDR2400, 1.65v).

Result in Prime95 was protective thermal shut down of the entire system within 5 seconds of starting the small FFT test. Core Temp showed all four cores showing yellow and about 96°C in the polling interval prior to shutdown, but the polling interval was longer during stress testing and the temperature rose so rapidly its likely that it was higher at shut down. I was so surprised I repeated this test several times with the same result each time.

Result in OCCT was that the test was automatically shut down due to reaching the thermal limit of 85°C within 80 seconds of starting the stress part of the test.

Result in AIDAx64 was that the test went ahead without thermal throttling, results showing that the cooler was taking at least 14°C off the stock cooler. However Core Temp issued thermal warnings and the max temp reached 86°C on 3 out of four cores.

Previously tests using AIDAx64 used the Intel retail HS/F because this is the reference test which SCAN recommended and use themselves. The expected temps with retail HS/F are 60°-70°C. While better cooling allowed the test to be run without throttling, this setup still overheated to much higher temps than expected even with the Intel HS/F and this does not change the fact that this setup fails to run the test without throttling with the Intel cooler reference set up thus failing SCANs test procedure.

So it does not change the fact that there is an overheating fault and the real world impact of this problem is evident from the Prime95 and OCCT tests. Even with a better cooler than Intel's there is an unavoidable risk of thermal shutdown in the middle of running an application or game which cannot be considered fit for purpose.

So I think its absolutely necessary for me to track down this fault and fix it so the board+cpu are running at normal temperatures to start with. It may be a wider problem than my set up, judging from the reports of other people .... So I feel it is appropriate to suggest it would help if anyone experiencing a problem with overheating could contact their mobo manufacturer technical support to ask for help and make them aware of the issue.

[BKauf]

I am going to test the cooler first, because I have been reading that the H100i doesn't always make a great contact with the processor without some extra washers behind the bracket. Will update whenever my thermal paste comes in and I can reattach it

[Allan_J_Intel1]

Thanks for the information; I will forward it to our engineering team.

Please keep in mind that the voltage spec for the embedded memory controller is 1.5v +/- 5% so the results when running memory at 1.65v are not all that surprising.

Allan.

[BKauf]

I have remounted the cooler, with new thermal paste and dropped load temps under prime95 to 70-85 depending on radiator fan speeds. Idle is still a little high at 35-40C with a room temp of 25

I am pretty sure I just mounted it poorly the first install, but it could be the new thermal paste.

[TIM_S_Intel]

what is the voltage set too?

You in Manual mode or Adaptive mode?

[REnso1]

Good point Allan, for the tests below using the Noctua, RAM is set to DDR1600 and 1.5v, timings are set to 11,13,13, 31 by the mobo.

tmsimmon wrote:

what is the voltage set too?

You in Manual mode or Adaptive mode?

It is adaptive with offset zero according to CPUz and BIOS. http://justpaste.it/gyfj CPUz text report on justpasteit.

BIOS settings are all AUTO but show the following values in BIOS.

CPU VCORE 1.191v,

CPU VRIN External Override 1.8v

CPU Graphics 1.0v

CPU Ring Voltage 1.050v

CPU System Agent 0.000v (ditto all offsets and I/O Analog and Digital, also zero)

I checked CPUz output when actively running OCCT and it stated "Core Voltage" was 1.213v.

I then checked the OCCT output and strangely it logged "CPU VCORE" as 1.63v at the same time that CPUz was reporting 1.213v.

I dont understand how this is happening and not entirely sure what it means. On my other PC (QX9650) these two readings correspond with each other. Can anyone explain why these two readings would be different on a Haswell i7 4790K?

These are the OCCT temps for a 5 min test at the same DDR1600 1.5v settings, cooled by the Noctua NH-L12 (the Intel HS/F would not run this without thermal shut down due to exceeding 85°C). Temps peak at 78-79°C even with the Noctua.

Running Prime95 small FFT with "core temp" running showed three cores maximising at 100°C before being shut down manually to prevent system shutdown.

So there is still apparently serious overheating despite more than adequate cooling from the Noctua which improves on the performance of the Intel stock cooler. (wrt Broxh's experience, in my case the Intel cooler was tested and then reseated once just to be sure prior to fitting the Noctua, both contact patterns were good for the Intel and performance consistent so it provides a good comparison to indicate the Noctua is working within spec.)

I dont understand the OCCT "CPU VCORE" reading 1.63v.

EDIT I tried setting the BIOS power draw limiter to 86w (from auto with displayed value 88w) and it made a huge difference to top temps but caused a lot of CPU frequency changes when under load, varying from 4 to 4.4 GHz but more often less than 4.4 GHz. Transient power draws reached up to 220w in Core Temp monitor even at this setting. Should that be happening ?

[idata]

I have an AsRock Z97 Extreme6 and am experiencing the same temp variations. Up to 99C under stress with no turbo, or overclocking. 35C at idle, and immediate jump to 47C if starting most anything small. Any suggestions? At what temp should I be concerned?

Were should I go to find out a solution to the problem?

[REnso1]

I have been working on this all day TAS3086 and the best move I made was to set the power limiter in the (Gigabyte F4 for Z97MX-Gaming-5) BIOS to 86W. The second best move was to set the vCore to "normal" setting for adaptive function and then set a negative offset of 0.060v.

What this does is to hobble the turbo and force power regulators not to pour power into the CPU and drops the default power levels a bit.

But its definitely not satisfactory. You can see from this comparison benching Prime95 small FFT test (DDR1600, 1.5v), on the left there is a negative offset but no power limit is set and on the right a power limit of 86W is active. Compare the Watts drawn and they drop from 142W > 89.9W and the temps drop from 95°C to 72°C but compare the CPU frequency and turbo is not active on the right. So the conclusion is that for some reason this CPU is pumping out too much heat for the frequency it is reaching and the processing it is doing because it is using too much power to do it.

[REnso1]

Update : I contacted Gigabyte support to ask whether the overheating was a generic problem and they replied saying they could not replicate it. They offered the following screeny as evidence but there are indications this is not at default settings. I am going to provide them with necessary directions to replicate and try to find out what settings they are using and see how that plays out.

In addition SCAN have confirmed that during the RMA process the CPU was tested separately using the Intel Processor Diagnostics tool which it passed but the CPU mobo combination was not tested using the replacement mobo because they do not normally test replacement items. We have discussed a second RMA with more thorough testing conditional on my replicating their test bench and verifying that the processor is overheating under these conditions including standard replicable BIOS settings. I will first try to discuss the standard for BIOS settings with Gigabyte.

With such a flexible BIOS it is easy to underclock a CPU to remain within a thermal envelope even when package cooling is flawed. Its hard to establish what the baseline settings should be for a fair test, but that is what I must do.

Should a 4790K be able to run 4 cores at 4.4 GHz without overheating? Any input on that welcome, thanks.