CPU damaged due to high temperatures?

[HayZ]

You can RMA the CPU if you do think its been damage from heat.

If it was damaged while using a after market cooler then Intel can be fussy. However if you used the stock cooler, and stock settings then you can RMA.

 

Use CPU-Z to see if it throttles, or some other program.

I can't see the CPU still working "properly" if its degraded / damaged.

 

Sure you don't have a memory leak?

I would be happy to RMA if needed. 

Also, how would I see on CPU-z if its throttling?

I'm not sure what a memory leak is or how I'd know if I have one. =/

[stconquest]

I would be happy to RMA if needed. 

Also, how would I see on CPU-z if its throttling?

I'm not sure what a memory leak is or how I'd know if I have one. =/

 

Strees test it and watch the clock in CPU-Z.  Monitor temps too, just in case.

 

Aida64 will do:  http://www.aida64.com/downloads

 

or, Intel Burn Test:  http://www.techspot.com/downloads/4965-intelburntest.html

 

or, Prime 95...

 

This is not to run the stressed CPU for hours... just long enough to see if it dips in performance... maybe 2-5 minutes.

[HayZ]

It was a shot in the dark... trying the simplest things first.

 

You can run a simple stress test (CPU at 100%) and monitor the clock and temps.  If it is running at 100% clock during the whole test, then it is definitely not the CPU that is causing the performance issue.

 

@HayZ

I've already done multiple stress tests and temperatures stay at around 75-80^oC. Nothing out of the ordinary really. I'll do some more tomorrow I guess for longer and see what happens

[Lucenttiger]

Maybe your cooler isnt mounted properly. Try remounting it and making sure there is adequate thermal compound there to transfer the heat. If that doesnt work you can also look up DELIDDING CPUS, every CPU after sandybridge intel has put a thermal paste instead of sodder. That really kicks the heat up on some CPUs

[stconquest]

I've already done multiple stress tests and temperatures stay at around 75-80^oC. Nothing out of the ordinary really. I'll do some more tomorrow I guess for longer and see what happens

 

Well, if your clock speed is fine during the stress test (and there is no crashing), then it is not your CPU that is causing your problems.

[HayZ]

Maybe your cooler isnt mounted properly. Try remounting it and making sure there is adequate thermal compound there to transfer the heat. If that doesnt work you can also look up DELIDDING CPUS, every CPU after sandybridge intel has put a thermal paste instead of sodder. That really kicks the heat up on some CPUs

I actually checked today and re-installed it to make sure. I had a bit too much thermal paste so I cleaned it and reapplied with a pea size and its still the same..

Not really sure if I'd want to De-lid my CPU. Seems like it helps with temperatures or something but thats not the problem now.

[Ryoku]

I actually checked today and re-installed it to make sure. I had a bit too much thermal paste so I cleaned it and reapplied with a pea size and its still the same..

Not really sure if I'd want to De-lid my CPU. Seems like it helps with temperatures or something but thats not the problem now.

 

I wouldn't bother, that was mainly an issue with the 4670-4770k cpus, the haswell refresh cpus 4690-4790k are a lot more power efficient which in turn makes them more heat efficient.

[HayZ]

Well, if your clock speed is fine during the stress test (and there is no crashing), then it is not your CPU that is causing your problems.

Hmm, I honestly don't know what else... I'm really hope its not my CPU.

[0x1e]

I've already done multiple stress tests and temperatures stay at around 75-80^oC. Nothing out of the ordinary really. I'll do some more tomorrow I guess for longer and see what happens

Temps don't mean anything.

You can be throttling to 2Ghz and stay at 80C. As it was the throttle that allowed it to get to 80C, rather that 99C @ 4.4Ghz (random numbers being used)

 

CPU at the bottom left will show you your clock speed. Do your normal tasks in windowed mode. Leave CPU minimised as you will then be able to see clock speeds in the task bar.

 

A memory leak is when the program is not deleting old data from ram. So it uses new ram and leaves the old data sitting there. Is slowly eats away all your ram.

So you know if you have a memory leak if you see the amount of memory being used by a program constantly go up.

It could take 5 minutes to begin, or 5 hours.

 

 

Also, do NOT delid a 4790k. Were they not soldered? 

[HayZ]

Temps don't mean anything.

You can be throttling to 2Ghz and stay at 80C. As it was the throttle that allowed it to get to 80C, rather that 99C @ 4.4Ghz (random numbers being used)

 

CPU at the bottom left will show you your clock speed. Do your normal tasks in windowed mode. Leave CPU minimised as you will then be able to see clock speeds in the task bar.

 

A memory leak is when the program is not deleting old data from ram. So it uses new ram and leaves the old data sitting there. Is slowly eats away all your ram.

So you know if you have a memory leak if you see the amount of memory being used by a program constantly go up.

It could take 5 minutes to begin, or 5 hours.

 

 

Also, do NOT delid a 4790k. Were they not soldered? 

I haven't Delidded anything, don't worry. 

Also, where can i see the memory of a certain program? And what would be a good program to test it on?

[stconquest]

I haven't Delidded anything, don't worry. 

Also, where can i see the memory of a certain program? And what would be a good program to test it on?

 

Thermals are fine on the i7 4790K... That is one of the improvements made to the Haswell Refresh (Devil's Canyon) CPUs.  It is why they are clocked higher than the Haswell CPUs.

[Coollok100]

Heat degradation can be an issue on certain CPUs, with them requiring more voltage to run at the same level of stability over time. However, Haswell CPUs are relatively resistent to this and while definitely on the warm side, 90C isn't enough to cause cause serious issues over this time period. I'd expect that what you're experiencing is some kind of software issue. For monitoring resource usage I'd recommend simply using Task Manager, or maybe NZXT's CAM. 

[Me1z]

I don't think it's actually possible to damage a processor through high temperatures alone. I'd look at whatever storage device you have or perhaps faulty drivers.

 

Scary CPU Story Time:

About 5 years ago, back when I was building my first computer, I was incredibly impatient and was assembling everything a little bit at a time as I could afford parts. It turns out the last part I ordered was my heatsink, and it wasn't due to arrive for about a week. Everything else was put together and it just so happened I got my hands on a newly released copy of Portal 2. In my infinite wisdom, you can easily guess what I decided to do. Not only did I boot up my computer and install Windows with no heatsink, I played through the entire campaign of Portal 2 with it in that state. Every 10 minutes or so I would pause the game and use my IR thermometer to monitor the temperatures, and I can't believe the thing didn't explode. At some points it got so hot the thermometer couldn't even read the temperature. It survived an entire weekend of that torture and now, 5 years later it still runs perfectly. Trust me when I say heat will not kill these things, it's just not physically possible.

[0x1e]

Also, where can i see the memory of a certain program? And what would be a good program to test it on?

 

Windows task manger will show you. Sort it by memory used.

[HayZ]

I don't think it's actually possible to damage a processor through high temperatures alone. I'd look at whatever storage device you have or perhaps faulty drivers.

 

Scary CPU Story Time:

About 5 years ago, back when I was building my first computer, I was incredibly impatient and was assembling everything a little bit at a time as I could afford parts. It turns out the last part I ordered was my heatsink, and it wasn't due to arrive for about a week. Everything else was put together and it just so happened I got my hands on a newly released copy of Portal 2. In my infinite wisdom, you can easily guess what I decided to do. Not only did I boot up my computer and install Windows with no heatsink, I played through the entire campaign of Portal 2 with it in that state. Every 10 minutes or so I would pause the game and use my IR thermometer to monitor the temperatures, and I can't believe the thing didn't explode. At some points it got so hot the thermometer couldn't even read the temperature. It survived an entire weekend of that torture and now, 5 years later it still runs perfectly. Trust me when I say heat will not kill these things, it's just not physically possible.

Ok, so if not the CPU.. how would I test drivers or storage devices for any problems? =/ I don't usually get problems, so testing these kinda things gets a bit confusing for me.

[stconquest]

Ok, so if not the CPU.. how would I test drivers or storage devices for any problems? =/ I don't usually get problems, so testing these kinda things gets a bit confusing for me.

 

Have you stressed you GPU?  To check temps?  Your problem is a FPS issue, right?

 

I have to ask, is your GPU dusty?

 

It is not your RAM.  Bad RAM will crash your system, not cause a drop in a game's FPS.

 

A loss of performance like that sounds like a heat issue (dust does this), or the most likely scenario... software.

 

How old is your Hard Drive?  SSD?  Run a test on them... not really sure but Crystal Disk, I think, tests drive speed.  I have never used it.

[StareX]

Where an i7 4690k can die is when you put too much voltage onto it for say like 1.8V when you shouldn't.

 

Your CPU isn't dead, it's probably just bottlenecked by something in your PC. So it's not dead, replacing coolers isn't what kills a CPU.

 

And if your CPU stayed 90C along time it doesn't matter, you would've noticed it before as it would've unclocked itself and recalibrate its Voltage as I believe your voltage control is on Auto ..

 

Which comes to another point it's 1.2V so it's not worst. CPU Dying = Crashes, BSOD, whatever .. you know.

 

It could really be your GPU to be honest or Hard Drive. or Windows itself.

 

 

But I believe whenever you're gaming alone, not streaming you're not having issues, now do you? Uhuh, now lets assume it's Software issues for now, ok?

 

 

P.S. What I recommend is run MSI Afterburner On-Screen-Display method of FPS + GPU Usage + GPU Clock ratio + GPU Temps..

 

and then install HWInfo , run a CPU all cores temps check, total CPU usage and CPU Clock ratio ON SCREEN DISPLAY.

 

and send us some SSs .. yeah? when you're streaming.. and when you're not.

[HayZ]

Have you stressed you GPU?  To check temps?  Your problem is a FPS issue, right?

 

I have to ask, is your GPU dusty?

 

It is not your RAM.  Bad RAM will crash your system, not cause a drop in a game's FPS.

 

A loss of performance like that sounds like a heat issue (dust does this), or the most likely scenario... software.

 

How old is your Hard Drive?  SSD?  Run a test on them... not really sure but Crystal Disk, I think, tests drive speed.  I have never used it.

Its not so much an FPS issue. Yes, FPS is affected but the whole PC performance is affected too. 

 

Would it be smart to just format all drives and reinstall windows? 

 

Meanwhile, I shall run a stress test on my CPU and GPU together and see what happens..

[HayZ]

Where an i7 4690k can die is when you put too much voltage onto it for say like 1.8V when you shouldn't.

 

Your CPU isn't dead, it's probably just bottlenecked by something in your PC. So it's not dead, replacing coolers isn't what kills a CPU.

 

And if your CPU stayed 90C along time it doesn't matter, you would've noticed it before as it would've unclocked itself and recalibrate its Voltage as I believe your voltage control is on Auto ..

 

Which comes to another point it's 1.2V so it's not worst. CPU Dying = Crashes, BSOD, whatever .. you know.

 

It could really be your GPU to be honest or Hard Drive. or Windows itself.

 

 

But I believe whenever you're gaming alone, not streaming you're not having issues, now do you? Uhuh, now lets assume it's Software issues for now, ok?

 

 

P.S. What I recommend is run MSI Afterburner On-Screen-Display method of FPS + GPU Usage + GPU Clock ratio + GPU Temps..

 

and then install HWInfo , run a CPU all cores temps check, total CPU usage and CPU Clock ratio ON SCREEN DISPLAY.

 

and send us some SSs .. yeah? when you're streaming.. and when you're not.

Thing is, I'm not sure if it happens whilst not streaming because I never play for long enough when not streaming for it to cause a problem. 

 

Might have to try it though..

[stconquest]

Its not so much an FPS issue. Yes, FPS is affected but the whole PC performance is affected too. 

 

Would it be smart to just format all drives and reinstall windows? 

 

Meanwhile, I shall run a stress test on my CPU and GPU together and see what happens..

 

I tend to do this when I am mildly frustrated with something I can't pinpoint... It eliminates software issues.

[HayZ]

I tend to do this when I am mildly frustrated with something I can't pinpoint... It eliminates software issues.

http://imgur.com/a/zJFBK

^This is a link to images of a stress test i just did on my GPU and CPU at ~%100 usage.

 

GPU Temp stopped at ~80^oC 

 

everything seemed pretty normal from what I could see. 

 

Might just have to format and reinstall windows..

[Lotus]

http://imgur.com/a/zJFBK

^This is a link to images of a stress test i just did on my GPU and CPU at ~%100 usage.

 

GPU Temp stopped at ~80^oC 

 

everything seemed pretty normal from what I could see. 

 

Might just have to format and reinstall windows..

 

 

On that stress test you're hitting 96°C which is NOT normal and means you're likely thermal throttling, which shouldn't happen with your cooler. Something is wrong, and 99 times out of 100, it's how you've installed the cooler.

[HayZ]

On that stress test you're hitting 96°C which is NOT normal and means you're likely thermal throttling, which shouldn't happen with your cooler. Something is wrong, and 99 times out of 100, it's how you've installed the cooler.

Its OC'ed to around 4.4GHz.

At default settings I was getting around 70-80^oC

 

Maybe I need to underclock? Not sure =/

[HayZ]

On that stress test you're hitting 96°C which is NOT normal and means you're likely thermal throttling, which shouldn't happen with your cooler. Something is wrong, and 99 times out of 100, it's how you've installed the cooler.

I've installed it twice already. 

 

Used a pea size dollop of paste too

 

Followed the handbook on how to set it up too, not sure what could have gone wrong.

[stconquest]

I've installed it twice already. 

 

Used a pea size dollop of paste too

 

Followed the handbook on how to set it up too, not sure what could have gone wrong.

 

Go back to stock speed.

 

Your max CPU temps:  94, 94, 96, 96... nobody wants this kind of max temp.     Even the 80 degree GPU is too much for me, but is safe.  What kind of case do you have and you fan configuration?

 

For the CPU, what are your idle temps?  They should be under 40-45 (you might live in a hot climate).  AiO pumps sometimes fail... it happens.  Monitor the pump speed with something like Speedfan.  Make sure it is steady, it should not fluctuate.