Could my 4790k be causing my random overheating problems?

[cytg111]

It's not my cooler.

I will say it again...my temps are fine for days on end then suddenly they won't be after a cold boot.

When I sleep my PC and wake it again the temps return to normal...I will film this behaviour next time if I'm still not being believed.

They then stay normal for a few more days then go crazy again after 5 cold boots or so.

This happens with the stock cooler too.

I've had the PC built outside its case to check a proper seat and it's absolutely fine.

Volts read what thy should whether stock, clocked, idle or load.

I don't want to be persuaded that it's my cooler I just want an idea of whether my cpu could be faulty.

Ps it's not my cooler or the seating of said cooler.

I don't know anyone who has a haswell set up and I don't know whether I would want to risk their kit anyway.

I'm going to rma the board unless someone can give a good argument as to how it could be the CPU.

And while temps are shooting through the roof, windows still reports 800Mhz activity? Can you measure, at the socket, how many watts your rig pulls before vs. during? Im thinking that since it heats up it gotta be burning more joules, finding the muncher is key. (and given the info available, i'd put money on the CPU).
edit : even if you were the victim of a rootkit I've never heard of one spoofing speedstepping..

 

[McGraw]

Why don't you start again, List your full system spec's were not mind readers. Also you come in here asking for help then yell at people who reply? You'll not get much useful help till you do list your full system, along with a I don't know it all attitude! Clearly the cause of the issue is you, RMA is not a fix...............

Hi. I wasn't yelling and I didn't really come on here asking for help actually, just a quick vote on whether people thought it could be my cpu causing my problem or whether that would be impossible.

Unfortunately despite explaining everything in crystal clear detail I have still been asked multiple questions that I had given the information for in my original post .

Please, just for my point of view, carefully read my firs post then read the questions I was asked I was in a bad mood last night so apologies if I came across badly but you don't need my system specs. I already said the only unknowns were whether it was the board or the cpu.

Ps I discovered tonight that if my system boots to this unstable state and AIDA instantly throttles, simply stopping then starting the stability test again gets rid of the problem.

 

[cytg111]

Hi. I wasn't yelling and I didn't really come on here asking for help actually, just a quick vote on whether people thought it could be my cpu causing my problem or whether that would be impossible.

Unfortunately despite explaining everything in crystal clear detail I have still been asked multiple questions that I had given the information for in my original post .

Please, just for my point of view, carefully read my firs post then read the questions I was asked I was in a bad mood last night so apologies if I came across badly but you don't need my system specs. I already said the only unknowns were whether it was the board or the cpu.

Ps I discovered tonight that if my system boots to this unstable state and AIDA instantly throttles, simply stopping then starting the stability test again gets rid of the problem.

youtube or it didnt happen

 

[McGraw]

Ok next time I will youtube.

 

[Termie]

OK, if stopping and starting a stability test fixes it, it's not the cooler.

Now, if that's true, there are only two things that can cause overheating. Clock speed or voltage. In this case, it appears it's neither. Which is of course impossible. For huge spikes like this, it's most likely voltage, so either you have a faulty CPU or motherboard, and in fact there are two faults - the voltage regulation and the voltage reporting. Very unlikely scenario, but it seems that's where we're at.

Have you ever heavily overclocked this system? The only way it's the CPU is if it's physically damaged, and it most certainly didn't come this way from the factory.

By the way, just for the OP's benefit, it's really better if you don't limit the information you post here if you're looking for technical help. For instance, you never once provided the reported voltages. That would be helpful. If you simply wanted us all to flip a coin on your poll, well, that's a bit of a waste of time.

 

[McGraw]

Ok well thank you for your reply. It's very difficult to list all the voltages concerned as I tried a lot of things to sort it out and never documented the voltages.

I can tell you that in the times I was able to check, the voltages and fan speed as stated originally did not change from before the problem was cured by a sleep cycle to after it had sorted it.

According to all the temp and voltage measuring software I have, there is no difference between the unstable and stable states.

I wasn't trying to limit the technical info, the exact opposite was my intention in fact.

I do now seriously consider the fact my cpu could be faulty.

It has never been seriously overclocked but it was running at 1.41v when I first set it up as the gigabyte v1 bios is not fit for purpose. Maybe installing windows at these volts damaged by cpu? This is gigabyte's fault though not mine as I had no way to update the bios without getting my pc up and running first.

The most I have overclocked it is to 4.6 at about 1.26v but it was a bit too hot so I now run at 4.5.

I know it seems like I'm missing something here but I promise I'm not. I've been upgrading and building pcs for nearly 20 years albeit it that this is my first desktop for a while and I've not had intel since my PII 450.

Ps my cpu doesn't miss a beat once it's been cycled out of the unstable state. If it's damaged I think I'm very lucky to have a workaround.

 

[Termie]

1.41v over an extended period of time would have permanently damaged the processor. 1.26v would not. So if you ran it for a while at the ultra high voltage before you caught it, it's likely the processor. But RMA'ing for that reason is unlikely to get you very far.

 

[crashtech]

If you have not bothered to learn how to manually set the voltage in the BIOS, now would be the time.

 

[McGraw]

If you have not bothered to learn how to manually set the voltage in the BIOS, now would be the time.

No I knew how to do this. I just ran it stock as I installed windows and never expected the board would auto volt to 1.41.

As soon as I could I update the bios and it returned to around 1.212 at stock.

 

[DrMrLordX]

Do let us know when you get some numbers from the Kill-a-Watt. In the meantime, another thing you should look at is possibly your motherboard's VCCIN setting.

If I recall correctly, VCCIN determines how much total power the board can possibly feed to the CPU package. Since most voltage control has been moved onto the CPU, this is the only setting where the board itself controls voltage.

So if you were to figure out the "normal" operating voltages of every part of your CPU when not in balls-out AVX2 mode, add them together, and then limit VCCIN to that voltage, you could stop the board from being capable of overvolting the CPU. The downside here, of course, is that this idea might not play well with LLC, and naturally you're going to have some problems with Prime95 28.5 since normal Haswell behavior is to bump up the vcore a little while dealing with heavy AVX2 workloads.

Or, you could just set a relatively low VCCIN like 1.65v and see if your system will start at all under those settings.

If the overheat continues under those conditions, you can PROBABLY assume that the board is ignoring its own VCCIN settings in the UEFI and still feeding too much voltage to the FIVRs. Direct measurement would be better, but this way all you have to do is punch some keys and stuff.

 

[McGraw]

1.41v over an extended period of time would have permanently damaged the processor. 1.26v would not. So if you ran it for a while at the ultra high voltage before you caught it, it's likely the processor. But RMA'ing for that reason is unlikely to get you very far.

Thanks. It was for about an hour as I installed windows and checked volts/temps.

If my cpu is damaged from this then surely gigabyte have a case to answer as optimised defaults shouldn't do this.

 

[crashtech]

No I knew how to do this. I just ran it stock as I installed windows and never expected the board would auto volt to 1.41.

As soon as I could I update the bios and it returned to around 1.212 at stock.

May I ask if you have manually locked the voltage to see if it helps with this problem you are having? If you are sure the problem has nothing to do with heat removal, it must necessarily be from heat production, and the only way to produce enough heat to overwhelm a heat sink is to apply excess voltage. (Well, and excess frequency, but you can't get that without a lot of voltage, too)

 

[McGraw]

Ok thanks again for all your help.

The voltage is currently locked at 1.224 and it creeps to 1.248 in AIDA.

However, I think I have found the problem. Using hwmonitor I've kept my eye on the fan speed and it seems fan 1 (there is a fan 0) is much lower than it should be when the problem is evident so I think the motherboard is sometimes not controlling the cpu fan (h60 pump in this case) properly.

Providing I can replicate the problem again with the stock cooler I can now confidently rma the board.

If I can't replicate the problem I will eat humble pie and say it was the pump and rma the h60.

I did try about thirty sleep cycles before without a hitch so I still believe the mobo is glitching on every fifth cold boot or so.