CPU i7-4790k rises to 80 as soon as i stress test it

Toggle sidebar Toggle sidebar
C

chas337

Honorable
Jan 6, 2015
16
0
10,510
i just brought a system off of someone and on my old system i had a nepton 140 and even under stress test my i5-4690k only reached 69c but for this its sky-rocketing is there something im missing? also the cpuvcore on hardware monitor shows max 1.48V is that right? as thats seems HIGH. i just wanted to know if this dude sent me a dodgy system thanks (its in push only at the moment)
 
Solution
10tacle
Yeah do NOT run that CPU with that high of a Vcore. You are killing that CPU. Temps actually are not that bad considering that's such a high voltage, WELL above the maximum recommended long term safe use of 1.35v use even under the best cooling solutions.

Return the CPU configuration settings in BIOS to default/auto and see where your baseline temps are in a stress test. Your stock CPU voltage should be somewhere between 1.1v-1.2v at 4.4GHz turbo for a 4790K under load. If your voltage is higher than that at stock, you may have a damaged chip. But again, those temps are really not that bad even on closed loop water considering the voltage.
Sort by date Sort by votes
10tacle

10tacle

Splendid
Dec 6, 2008
4,941
116
29,140
Yeah do NOT run that CPU with that high of a Vcore. You are killing that CPU. Temps actually are not that bad considering that's such a high voltage, WELL above the maximum recommended long term safe use of 1.35v use even under the best cooling solutions.

Return the CPU configuration settings in BIOS to default/auto and see where your baseline temps are in a stress test. Your stock CPU voltage should be somewhere between 1.1v-1.2v at 4.4GHz turbo for a 4790K under load. If your voltage is higher than that at stock, you may have a damaged chip. But again, those temps are really not that bad even on closed loop water considering the voltage.
 
Upvote 0 Downvote
Solution
C

chas337

Honorable
Jan 6, 2015
16
0
10,510



ok thanks currently have it at 1.22 and temps are now only hitting 58c under 100% load thanks a ton. now is there a way to see if this buffoon has damaged the chip and weather i should try and return it to him and get my money back (may be difficult)
 
Upvote 0 Downvote
10tacle

10tacle

Splendid
Dec 6, 2008
4,941
116
29,140
About the only way I would say is do like I said. Forget overclocking return the BIOS settings to default and auto. Let the motherboard control the Vcore on auto setting...don't manually enter in a Vcore setting. Then run a stress test and see where your 4.4GHz turbo boost speed Vcore use is. Anywhere between 1.00v-1.20v is acceptable, with 1.12v-1.16v being the most common average range (according to reports of others). If your use is over 1.20v at stock speed, you may have had some damage already (damaged chips require more voltage to work than one not damaged). It doesn't mean it won't function fine, but it does mean its life will be shortened, especially if you overclock it.

Next you can try a mild overclock to say 4.6GHz on turbo. You should be somewhere in the range of 1.20v-1.30v needed for that. Anything higher may further validate there was some damage done. Over time, overclocking with voltage increases reduces the life of the chip anyway, but it is gradual (called CPU degradation). If you are within those tolerances of voltage in both tests, stock and overclock, I wouldn't worry about it. And it's really hard if not impossible to prove someone abused voltage on an overclock, so yeah, you likely wouldn't have much luck returning it anyway. Only way to make a reasonable assumption is to see where that chip's voltage use falls in line compared to others.

One side note: as an afterthought, even if you are slightly out of those voltage ranges mentioned, it may simply just mean that chip was not a good one from the factory (lower quality silicon, whatever). Some people get lucky and get a good chip that uses lower voltages than others (I got lucky with my i5 4690K).

Good luck!
 
Upvote 0 Downvote
C

chas337

Honorable
Jan 6, 2015
16
0
10,510
ok thanks now im having yet another problem! ok so i have got this build off of a friend and i tired enabling SLI but it didnt allow it. both cards show up on bios and afterburner etc. but when i go into nvidia control panel it doesn't show up. just brought a brand new SLI bridge and nope still no luck. im getting a D4 error on my mobo and im guessing that that is the problem and after searching i still dont know how to fix it.
i put the top card in x16 and the second in x8 just how the manual shows mobo is ga-z97x-soc force

thanks
 
Upvote 0 Downvote
10tacle

10tacle

Splendid
Dec 6, 2008
4,941
116
29,140
Okay I'm not totally clear. When you say you go into Nvidia control panel and "it doesn't show up" do you mean your second card is not being seen there or that you aren't seeing SLI showing as enabled? I've had to actually swap cards in an SLI build. For whatever reason, it "woke up" that second PCIe slot for Nvidia to see.

The fact both the motherboard BIOS and Afterburner see both cards leads me to believe it's an Nvidia controller/software problem. Take the GPUs out and swap them around. But before you do that, take out the CMOS battery (make sure PC is unplugged from your PSU of course) and let it reset for a minute. This should clear that D4 error which is a general PCI allocation error from what I've quickly checked.

If you do all that and still have the same problem (Nvidia control panel is NOT seeing the second GPU), then try de-installing the Nvidia drivers completely with a program like CCleaner (to clear out anything Nvidia-related that got into the Windows root registry that a normal de-install may not get all of). Then remove the lower GPU, reboot, re-install the Nvidia drivers, then install the second card and connect. If you do all that and still have the same problem, I'm not sure what the issue is. I assume you have the latest BIOS for that motherboard too.
 
Upvote 0 Downvote
C

chas337

Honorable
Jan 6, 2015
16
0
10,510



thanks for all the help it took me reinstalling everything and putting in the cards 1 by 1 to get it to work
 
Upvote 0 Downvote
10tacle

10tacle

Splendid
Dec 6, 2008
4,941
116
29,140



Cool! I was wondering what had ever happened. SLI can be temperamental and have kinks that need to get ironed out. Lots of trial and error between messing with the hardware and software. In one of my new SLI builds from years ago, I was getting intermittent SLI disconnects - the second card would just suddenly not be detected randomly - and nothing would clear it. Others were reporting the same problem. Turns out the motherboard maker had to issue an updated BIOS to fix the problem for certain GPU types that was never discovered at the factory. You just never know.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom