• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

i7-8700k runs insanely hot even at stock speeds ?

I think I need some help, same board, same CPU...

Can you help me?




2.JPG
 
Hi there man, I just had 1 month ago this problem exactly. No matter what I did to replace the thermal paste or to update bios, was always the same.
Only think what I did and helped and regain my normal temperatures, was to reload previous settings from Windows Troubleshooting from the bios settings.
Thats all, it really helped and made me smile again.
Good luck
 
Sebsta said:
Hi there man, I just had 1 month ago this problem exactly. No matter what I did to replace the thermal paste or to update bios, was always the same.
Only think what I did and helped and regain my normal temperatures, was to reload previous settings from Windows Troubleshooting from the bios settings.
Thats all, it really helped and made me smile again.
Good luck
Click to expand...

Where is it exactly?
 
Last edited:
Verbatim said:
Who has i7-8700k cpu ? Please all run intel burn test like me to compare 1:1 Settings are fully stock cpu/voltages and custom stress level "8000 mb" "Times to run 5" for intel burn test.
njYOKS.png



I have tried to lowering these voltages but temperatures are almost the same... So it's isn't helping to me.
Click to expand...
Only a moron would run ibt on there own cpu for extended periods, nice job!!! Cant understand whats wrong.....
 
jaggerwild said:
Only a moron would run ibt on there own cpu for extended periods, nice job!!! Cant understand whats wrong.....
Click to expand...

I mean, you should be able to run anything on a truly stable CPU, dude...
 
Don't forget to run furmark for 2 hours on your video card too!
 
johnspack said:
Don't forget to run furmark for 2 hours on your video card too!
Click to expand...

CPUs are not GPUs. GPUs have hard watt limits for a reason. CPUs don't.

I regularly use Prime95 Small FFTs and Intel IBT as stability tests for hours. I laugh at people who say their overclock is stable and then tell people they can't run this because there is something "wrong" with these programs...

...But that's just me, and I like to laugh.
 
EarthDog said:
CPUs........well......motherboards have hard watt limits. ;)
Click to expand...

Well yes, but then, your OC is bad and VRMs generally throttle you.

I meant in bios... whole different ball of wax, anyway.
 
Use ASUS RealBench H.264 test for 10 loops. If it passes, it's stable. I found it to be the most realistic test unlike dedicated burn tests. It puts heavy load on all cores and uses AVX instructions which uncovers instabilities quickly.
 
R-T-B said:
CPUs are not GPUs. GPUs have hard watt limits for a reason. CPUs don't.

I regularly use Prime95 Small FFTs and Intel IBT as stability tests for hours. I laugh at people who say their overclock is stable and then tell people they can't run this because there is something "wrong" with these programs...

...But that's just me, and I like to laugh.
Click to expand...
I laugh at people who run stress tests and think that if it passes it's stable. I could run 100 loops of IBT at max. preset but some games would crash it in 3 minutes. Good luck with torturing your CPU with stress tests for hours days and weeks. You proved absolutely nothing.
Sure, they're helpful, mainly to see if you're not hitting some sort of power limit when you OC and to do a quick recon of what your CPU is capable of, but time verifies if your OC is stable more than a stress test.
 
Last edited:
cucker tarlson said:
I laugh at people who run stress tests and think that if it passes it's stable. I could run 100 loops of IBT at max. preset but some games would crash it in 3 minutes. Good luck with torturing your CPU with stress tests for hours days and weeks. You proved absolutely nothing.
Sure, they're helpful, mainly to see if you're not hitting some sort of power limit when you OC and to do a quick recon of what your CPU is capable of, but time verifies if your OC is stable more than a stress test.
Click to expand...

IBT is honestly only a tool in the toolbox. But if it fails IBT or Prime95 it is certainly not stable. That was my point.
 
Bund said:
Where is it exactly?
Click to expand...

Restart now the pc, once it starts to show the screen with the ''American Megatrends'' and all the USB connections list, exactly than press Deletele or F2 or the same time a few times, so that it goes to the bios. Once there, just go and find in the top right side window the word keys ''Load Previous Data windows system'' or something like that. Once Apply, just wait to do its thing, is 1 min or 2 thats all.

If nothing changed after this, from my own experience, than there is a better option, of totally restoring the previous windows files, namely at the Windows Setting from the Start -> and after you just go to Recovery sector and click to Windows Reset. From here you will find your own best way, is easy ;) .
And let us know if it all worked for you, thank you.
 
cucker tarlson said:
A bit warm maybe cause his voltage at stock sits at 1.256v (1.23v VID), which is probably more than needed. I'd try set 1.20-1.21v manually to get maybe 1.23v during stress test.
Click to expand...

For 4.3GHz even less than 1.20V, i need 1.22V for 4.8GHz.
 
R-T-B said:
IBT is honestly only a tool in the toolbox. But if it fails IBT or Prime95 it is certainly not stable. That was my point.
Click to expand...
Yes but if it passes stress tests and freezes a game it's not stable either. Way I did it was: set the frequency you're aiming to hit, like 4200MHz. Find the voltage you need for it to be stable, but start low. Then when you have found that voltage, just play games and run stuff as usual, you might lower the game settings to push the CPU harder if you wish like I did. It's a good idea to compare your previous fps to see if the actual increase is worth it in the first place. When it runs stable for a week or two, bump the voltage ever so slightly, like I did from 1.28v to 1.29v, just to be sure. I ran IBT for 2 minutes only to see if it doesn't power throttle. No need for regular stress tests in p95, what is the point of that ?
 
off topic but maybe not.. i just bought a cheap I5 kaby lake 1050TI for another family member.. he dosnt game but the 1050ti was just in case.. :)

what i noticed was.. even though it has the sh-t intel cooler fitted.. how cool it runs under full load.. about 40 C max..

the second thing i noticed was how well the little 1050TI performed.. time spy turned it into a bit of a slide show but for casual lower resolution gaming it works fine..

i took the side off and fitted a small ssd for the boot drive.. while i was in there i removed the single case fan which was buzzing the super thin case material..

the only case fan it has now is the one in the top fitted psu.. under full gaming load it blows out cool air..

the moral of this story.. dunno but i recon there is one.. he he

trog
 
Correct either way. But id rather take the time up front than be plagued with freezing and instability when im trying to actually use my machine. Hence the need. ;)

What programs and how long depends on the uses for the machine and user. Which, obviously varies by person.
 
cucker tarlson said:
Or cause it quicker.
Click to expand...

I love how ever since we have 6c/12t CPUs from Intel, we are now resorting en masse to 'apologies' for failing stress tests by saying 'but it'll game fine for hours'... or 'but I never use AVX'...

Most of these comments stem from a lack of knowledge. Gaming will use AVX today and more as you go forward. You need to be AVX stable, you need to be able to pass a couple of IBT runs, and you need to be able to P95 without bluescreening. Do you need to be able to run that for 24 hours? No. But they are tools to gauge stability, and if they fail, you can't just ignore that. Its called being lazy ;)
 
EarthDog said:
Correct either way. But id rather take the time up front than be plagued with freezing and instability when im trying to actually use my machine. Hence the need. ;)
Click to expand...
Once you stabilize yor OC and you're not one of those people who must have their voltages borderline to what they've found stable there's no need. Plus testing with the same tools to find instability is futile, why do you regularly keep on testing with a tool that showed it's stable the last 10 times. Test with other ones.

Vayra86 said:
I love how ever since we have 6c/12t CPUs from Intel, we are now resorting en masse to 'apologies' for failing stress tests by saying 'but it'll game fine for hours'... or 'but I never use AVX'...

Most of these comments stem from a lack of knowledge. Gaming will use AVX today and more as you go forward. You need to be AVX stable, you need to be able to pass a couple of IBT runs, and you need to be able to P95 without bluescreening. Do you need to be able to run that for 24 hours? No. But they are tools to gauge stability, and if they fail, you can't just ignore that. Its called being lazy ;)
Click to expand...
You're missing the point.
I'm not saying when failing a stress test a CPU is stable, I'm saying passing them is not always an indication of stability. Not 100% times.
 
cucker tarlson said:
Once you stabilize yor OC and you're not one of those people who must have their voltages borderline to what they've found stable there's no need. Plus testing with the same tools to find instability is futile, why do you regularly keep on testing with a tool that showed it's stable the last 10 times. Test with other ones.
Click to expand...
Que??

Of course there isnt a need to run more programs once you are stable...i didnt say nor allude to otherwise.

Because its stable for my uses is why i would use the same application.

Im really not sure what you are on about here. Im simply explaining why there is a need to test for stability. Everyone has a different method and if it works for them, it works. It really is that simple.
 
EarthDog said:
Because its stable for my uses is why i would use the same application.
Click to expand...
This makes no sense on so many levels.
Let's say you passed 10 hrs of p95. Why then do p95 regularly afterwards, you just found out it's stable. And no stress test really reflects what our normal uses tend to be.
 
cucker tarlson said:
You're missing the point.
I'm not saying when failing a stress test a CPU is stable, I'm saying passing them is not always an indication of stability. Not 100% times.
Click to expand...

Correct, but many people perceive that statement as 'so why would I run P95 or IBT, its not a guarantee of stability either' which is just plain wrong :D
 
You must log in or register to reply here.
Back
Top