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.

[DSche4]

This is my solution! No guaranty!

I no longer have time to mess with this. I bought a gigabyte motherboard and Intel chip that was supposed to work right out of the box. If this is your hobby... I can see how solving these issues can be amusing; but for me, they have been quite irritating. I probably should have RMA'd the chip, or went with an entirely different setup.

The situation started when I decided to upgrade my system to the latest I7 chip. Shortly after the build, the computer would shut off suddenly while under load with some games and rendering videos. After a little investigation, I found that the processor temperature was hitting 100 degrees C in these conditions. I searched for solutions online and this thread seemed to be the closest thing to a solution. I was never interested in overclocking or modifying any settings. Here are the things I did to solve the problem:

1. Changed the settings using Intel Extreme Tuning Utility to match the recommended setting earlier in the thread. ie. 88, 110 and 105 or whatever. RESULT: no improvement.

2. Reseated the cpu with Arctic Silver thermal paste using the stock cooler. RESULT: reduction in temperature and no shutdowns. temps still above recommended levels.

3. Intalled Noctua NH-U12S aftermarket cooler. RESULT: Temperatures rarely exceed 65 even under stress test conditions.

PROBLEM SOLVED FOR ME!

The following videos outline my work on solving this issue. Clearly, the Intel provided stock cooling fan is insufficient! I hate that I have had to invest this much time in this and hopefully my solution will allow others who do not wish to overclock their cpu to move passed this issue in a timely manner.

Applying Thermal Paste

http://youtu.be/SQR5WmYg3lU How To: Apply Arctic Silver CPU Thermal Paste i7 4790K - YouTube

Installing Noctua Fan

http://youtu.be/iB8FatWndDU How To: Install the Noctua NH-U12S CPU Tower Fan - YouTube

Good Luck!

Dan

[SRene]

My new cooler arrived (CoolerMaster Hyper 212X) it's not a supercooler but it made me get the temperatures down 23ºC.

Idle with the stock cooler with undervoltage vcore:

Idel with the new cooler (notice the clock):

AIDA64 3 minutes test with stock cooler (with TDP limited):

AIDA64 3 minutes test with new cooler (with TDP limited):

AIDA64 3 minutes test with new cooler and default bios options:

Prime95 with new cooler and limited TDP and undervoltage:

Prime95 with new cooler and default options:

With a room temperature of 29ºC, I believe it's ok now.

[SBuck1]

I've submitted a ticket with Gigabyte on this after checking what the IXTU program said were the "defaults" for my processor.

Not sure wtf is going on with that, but we'll see what they say. I should add that I haven't bothered limiting 3core to 43X and 4core to 42X because it's already limited that way in the bios. At full load the cpu only hits 4200MHz.

[RKimb]

I have a similar setup:

This is Gigabyte Z97X-UD3H with Bios ver F7.

I did the stress test.

CPU Total TDP value was 80W and the CPU temp got up to 85C, at which point I stopped the test (after 4.5 mins).

Given that this is 10 degrees higher than the Intel stress test reported earlier, I'm wondering what the explanation might be. The XTU program didn't offer me the option of setting the Core voltage. Maybe that's an issue. I could do it in the bios separately, there's a setting for VCORE (which I think it currently 1.2) - would lowering it be the answer?

[SBuck1]

Interesting. I can't get the Intel tool to report total TDP correctly, it just shows up as 1W no matter what. I think gigabyte needs to update some BIOSes.

[RKimb]

Interestingly, I seem to have solved my problem. Have you tried clearing the CMOS and loading the default optimized settings? I did this and my problems have been resolved. With Intel Turbo Boost turned off, the highest temperature I now get in normal operation is 66C. If it is on, there are occasional peaks at 80C, which I assume is within the Turbo Boost specs. I haven't tried to overclock - was previously getting problems without overclocking.

I did the IXTU stress test, with Turbo Boost off, and passed, with maximum temperatures of 69C and no temperature throttling.

I did the benchmark and got a score of 958, with a maximum temperature of 76C.

There seems to be something wrong with the BIOS settings some people have, especially with Gigabyte MBs. I don't know whether it's the retailer who sets up the BIOS, or whether they just take what Gigabyte provide without checking. But clearing the CMOS and re-loading was the solution for me - perhaps that should be the first thing to do before seeking a replacement CPU.

[RKimb]

I have since received a reply from Gigabyte Support that suggests clearing the CMOS.

I suppose the real question is why, on a new machine, the BIOS settings were not optimal in the first place.

[aanwa1]

I confirm that, clearing CMOS on my gigabyte z97x gaming 5 f5 bios did the trick and now i get 75-77c max with turbo boost off using prime 95 small FFT torture test !!

[kpago]

guys when cmos reset xmp reset 2.if run i say agian with xmp disable all is good !!and with turbo 4,4 open.if you use xmp again heat 96 96 c.its not solution.and why we buy a proccesor and we must play with watts and ambers???must intel or the mbs factorys find solutions and run at least at 4,4 ghz.

[SPark48]

Well, if CMOS resets are resolving some of the issues, there are a couple of things to consider:

1) The issue is with the motherboard settings rather than the processor.

2) The issue can be resolved with a BIOS update and/or setting changes, rather than it being something wrong with the processor hardware.

I personally think that the issue is both - there is something wrong with the processor hardware AND the motherboard settings need to be adjusted to make this processor run correctly. I am also having the same problems as everyone else with an ASUS Z97-Pro, so it isn't limited to just Gigabyte motherboards. I am on the latest BIOS for the MB, yet at stock settings, temps are ridiculously high. As I stated before, adjusting the watt and amp to get better results is a start, but not the final solution that we should be satisfied with; especially considering we're talking about high temps running at stock speeds on an unlocked processor. I understand everyone is frustrated (as am I), but considering these developments, putting the entire blame squarely on Intel is a little premature at this point.

[aanwa1]

thats true, it seems that the problem is within memory frequency even when i changed my memory to 1866 with XMP disabled i got the same overheating like before.

[HFigu1]

Hello,

Motherboard: GA-Z87X-OC I'm getting CPU frequency test failed. Slow performances as well. I tried to follow the screenshots provided.

[HFigu1]

Are any of you getting the CPU frequency fail? I was directed here for a different post. I do not have the overheating problem.

[SPark48]

I don't know about anyone else, but my (overheating) CPU does not fail the Intel PDT test, which is the reason Intel did not want to grant an RMA at first.

However, your post and image has me curious... So far, all the MBs I have tried are Z97s and the 4790K is overheating on all of them. Can anyone who is keeping up with this thread confirm or deny if they are having overheating issues with a Z87 motherboard?

[HFigu1]

I cleared cmos and booted up the machine to run the stress test. Here are my results. I do still fail the CPU frequency test!

[SPark48]

jfig04,

Since I pretty much failed EVERY other test, but passed the Intel PDT and Intel refused RMA, maybe it will work the other way around for you. Even though everything else looks fine, since you're failing Intel PDT, maybe they will do RMA. Also, even if you're not having any problems now, failing the PDT can mean you may have problems in the future. I highly suggest contacting their support desk.

Good luck!

One other thing, do you happen to know your processor's lot number? Thanks!

[HFigu1]

My lot number is L4268956

[SElwa]

Just an update, as you know I replaced my CPU finally, I used Ken-intel suggested settings, but I increased VCore to 1.150v (previously was 1.109v@4.2Ghz) and set all cores multipliers to 44 X then used AIDA64 Extreme to do 2 stress tests, here are the results:

1- Stress test type (CPU Stress only), Duration:16 min, @4.4GHz (stable) using the stock cooler, Max temp: 71c to 73c (old CPU had been 100c always).

2- Stress test type (CPU Stress + FPU Stress), Duration:30 min, @4.4GHz (throttled to 4.2Ghz due to thermal and current throttling) using the stock cooler, Max temp: 91c to 97c (some times it goes to 100c in one core then gets back to 95c then 97c but the remaining cores never exceeds 91c-99c).

I think that is good using Stock cooler, you could call me crazy, but I think when I use aftermarket cooler the results will be brilliant taking into account that the normal stress test temperature was max 73c @4.4GHz stable and also the idle temp is 29c-33c (the old CPU was 37c-41c) and also I noticed something in the new CPU, it requires less VCore than the old CPU, the old CPU was semi-stable @1.199v but the new one could operate @4.2Ghz at VCore of 1.109v and 4.4GHz stable at VCore of 1.150v

What do you think guys?

My board is GA-Z97X-UD3H (rev. 1.0) - Bios 7

Memory XMP profile enabled (16G ram @1866)

My new CPU Batch# : L434C341 and S/N: 3E433036A2639

My old CPU Batch# : L4****** (can't remember) and S/N: 2L422235A4539

I hope that information helps.

[kpago]

i tryed all nothing same overheating.my batch number is L437c095 if someone know something.ty

[kpago]

i tryed 110W , 255A turbo was at 3,4 ghz only!!the temratcure was 55c.to low ghz for this cpu

[REnso1]

pagos that is interesting, since you have the same motherboard (GA Z97MX Gaming 5) and get the same result with both limiters set. At least it shows it is not just my set up. So hopefully Gigabyte will be able to reproduce it and maybe find a fix.

I am assuming the Intel recommended settings work OK on other mobos.

What I don't know is whether this issue has any bearing on overheating. Since high current generates heat I am wondering if it might just contribute to overheating if it is not being managed correctly in the BIOS.