Jump to content


 


Register a free account to unlock additional features at BleepingComputer.com
Welcome to BleepingComputer, a free community where people like yourself come together to discuss and learn how to use their computers. Using the site is easy and fun. As a guest, you can browse and view the various discussions in the forums, but can not create a new topic or reply to an existing one unless you are logged in. Other benefits of registering an account are subscribing to topics and forums, creating a blog, and having no ads shown anywhere on the site.


Click here to Register a free account now! or read our Welcome Guide to learn how to use this site.

Photo

Computer shutting down when stressed


  • Please log in to reply
8 replies to this topic

#1 Scrawny

Scrawny

  • Members
  • 9 posts
  • OFFLINE
  •  
  • Local time:02:59 AM

Posted 05 February 2014 - 07:44 AM

Putting the computer under stress (e.g. Higher end video rendering or gaming) causes the computer to suddenly shut down. It doesn't slow down before shutting down, doesn't restart or bluescreen, just goes black. Restarting the computer immediately after this happens causes it to shut back down shortly after loading the desktop.
From what I've read, this seem to be caused by the CPU overheating. I'm running an FX-8320 at around 35c when doing nothing, and going as high as 65c (According to AI Suite II. HWMonitor never seems to show over 60c) when gaming. I'm using the default heatsink that came with the CPU, and have already tried replacing the heat compound.
 
To get to the main question, is the default heatsink on the 8320 so utterly worthless that it can't keep the CPU temperatures safe, or could this be caused by something else? I already replaced my PSU, thinking that that might be at fault, so I'd like to get a few opinions before I start spending cash on more replacement parts again and end up replacing the entire PC. Again.
 
GPU: GTX 780
CPU: AMD FX-8320
Mobo: Asus M5A97 LE R2.0
RAM: A single Kingston 8Gb HyperX Blu, 1600MHz DDR3
PSU: Corsair 750W, CS750M
 
All parts excluding the RAM are brand spankin' new.
If the heatsink clearly is the problem, any suggestions for a replacement would be more than welcome. Preferably a little over what I actually need just to be safe. I'm not really sure what to look for in the sinks other than reviews, so any thoughts there are appreciated too.
I do plan on doing some overclocking. Not looking for the absolute maximum, just going to take it up a little.


BC AdBot (Login to Remove)

 


#2 zingo156

zingo156

  • BC Advisor
  • 3,333 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:07:59 PM

Posted 05 February 2014 - 08:33 AM

I would recommend downloading and running prime95 to test the max cpu temperature and stability. Since you have already replaced the psu, it seems more probable that it is heat related. You can get prime 95 here: ftp://mersenne.org/gimps/p95v277.win64.zip

 

Watch your temperatures while running prime 95, if it runs stable for 30 minutes or longer, I doubt it is heat related to your CPU. This will use the cpu at 100% which is most likely more than any modern game will do since modern games use primarily the GPU.

 

If Prime 95 passes, then run furmark which will only test the gpu: http://www.geeks3d.com/20131118/furmark-1-12-0-download-gpu-vga-videocard-burn-in-stress-test-opengl-benchmark-utility-nvidia-geforce-quadro-amd-radeon/#download

 

Again watch your temps, if you have crashes during furmark I would suspect a GPU related issue. If you have crashes while running prime 95 or BSOD then you might assume ram or CPU.


Edited by zingo156, 05 February 2014 - 08:33 AM.

If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#3 Scrawny

Scrawny
  • Topic Starter

  • Members
  • 9 posts
  • OFFLINE
  •  
  • Local time:02:59 AM

Posted 05 February 2014 - 09:21 AM

I would recommend downloading and running prime95 to test the max cpu temperature and stability. Since you have already replaced the psu, it seems more probable that it is heat related. You can get prime 95 here: ftp://mersenne.org/gimps/p95v277.win64.zip

 

Watch your temperatures while running prime 95, if it runs stable for 30 minutes or longer, I doubt it is heat related to your CPU. This will use the cpu at 100% which is most likely more than any modern game will do since modern games use primarily the GPU.

 

If Prime 95 passes, then run furmark which will only test the gpu: http://www.geeks3d.com/20131118/furmark-1-12-0-download-gpu-vga-videocard-burn-in-stress-test-opengl-benchmark-utility-nvidia-geforce-quadro-amd-radeon/#download

 

Again watch your temps, if you have crashes during furmark I would suspect a GPU related issue. If you have crashes while running prime 95 or BSOD then you might assume ram or CPU.

 

I ran the In-place large FFTs test a few times. According to HWMonitor, the heat got up all the way to around 70c in less than a minute, before the PC blacked out. That seems pretty excessive even for a 100% test. Could a better heatsink really affect the heat enough to keep the CPU alive for 30 minutes under that pressure? Or could there be some other reason it's overheating so quickly?



#4 zingo156

zingo156

  • BC Advisor
  • 3,333 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:07:59 PM

Posted 05 February 2014 - 10:03 AM

Even with the stock cooler you should be able to use the cpu at 100%, there must be some other issue. You can try a different cooler, something like the evo 212 will help especially if you plan on overclocking.

 

Do you currently have any over clocks? If you do I would remove those and retry.

 

If you do not overclock and it is overheating, there must be some issue with the cooler/fan or maybe it is not making contact well enough. Do you have the fan connected to the cpu fan spot on the mainboard? Is the fan speeding up when the cpu heats up? You should notice it speeding up.

 

Out of curiosity: what case do you have? What is the ambient room temperature where the computer is located? Is it in an enclosed area that can heat up?

 

Just for your information:

Max safe core temp recommended by AMD is 62 degrees celsius.
maximum socket temperature is 70 degrees celsius.

 

EDIT: Because your cpu is overheating so quickly (less than a minute) I suspect your cooler is not making contact with the cpu correctly...


Edited by zingo156, 05 February 2014 - 10:31 AM.

If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#5 Scrawny

Scrawny
  • Topic Starter

  • Members
  • 9 posts
  • OFFLINE
  •  
  • Local time:02:59 AM

Posted 05 February 2014 - 10:34 AM

Even with the stock cooler you should be able to use the cpu at 100%, there must be some other issue. You can try a different cooler, something like the evo 212 will help especially if you plan on overclocking.

 

Do you currently have any over clocks? If you do I would remove those and retry.

 

If you do not overclock and it is overheating, there must be some issue with the cooler/fan or maybe it is not making contact well enough. Do you have the fan connected to the cpu fan spot on the mainboard? Is the fan speeding up when the cpu heats up? You should notice it speeding up.

 

Out of curiosity: what case do you have? What is the ambient room temperature where the computer is located? Is it in an enclosed area that can heat up?

 

Just for your information:

Max safe core temp recommended by AMD is 62 degrees celsius.
maximum socket temperature is 70 degrees celsius.

 

EDIT: Because your cpu is overheating so quickly (less than a minute) I suspect your cooler is not making contact with the cpu correctly...

 

I have a Thermaltake Chaser MK-I, it's pretty darn well endowed when it comes to fans. It's a little crowded inside, though, but keeping a hand close to all the fans, and inside the case in general, it seems fairly cool.

I guess I'll try messing with the cooler a bit. Might as well try reapplying the thermal paste again, just to be sure.



#6 zingo156

zingo156

  • BC Advisor
  • 3,333 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:07:59 PM

Posted 05 February 2014 - 10:39 AM

When it comes to thermal compound, a little goes a long way. Too much can cause problems but you do want enough for good contact. Because it is overheating so fast it really leads me to believe there is no contact between the cpu cooler and the cpu or not enough. Even a cooler without a fan should keep a cpu cool enough to run prime 95 for longer than a minute.

 

EDIT: you can physically touch the cooler and see if it is getting hot, if it does not feel 70C when the processor is you can be certain it is not transferring heat and thus not making contact with the cpu itself.


Edited by zingo156, 05 February 2014 - 10:40 AM.

If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#7 Scrawny

Scrawny
  • Topic Starter

  • Members
  • 9 posts
  • OFFLINE
  •  
  • Local time:02:59 AM

Posted 05 February 2014 - 11:28 AM

Well, I redid the whole thermal paste shabang. Took it down a little, to the point where I can actually see through the layer (was previously a very thin layer, but not thin enough to see through). Ran the Prime95 test for two minutes without blacking out and only got to around 61c, so that's an improvement. Whether it was the paste or something else that was wrong, who knows. I'm a little scared to leave the test running for half an hour. Will run it for a bit longer now and see what happens.

 

EDIT: Boom, over ten minutes and the temperature's been at 63c for quite a while. I hereby deem it stable.


Edited by Scrawny, 05 February 2014 - 11:44 AM.


#8 zingo156

zingo156

  • BC Advisor
  • 3,333 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:07:59 PM

Posted 05 February 2014 - 11:53 AM

Congrats! I do believe it was an issue with the cooler making contact with the cpu, it could be too much thermal compound or that the cooler just wasn't fitting correctly. I would say that 63c is ok but a bit hot for a stock cooler under 100% load. If you still want to over clock I would recommend an after market similar to the EVO 212, there are better coolers but that is a good example of a decent after market cooler which would allow for some O.C. Best of luck!


If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#9 jonuk76

jonuk76

  • Members
  • 2,178 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Wales, UK
  • Local time:01:59 AM

Posted 05 February 2014 - 12:07 PM

Yep too much thermal paste is bad.  It's just there to fill tiny gaps between the heatsink base and the CPU.  The best application procedure is to put a small amount in the middle of the CPU as per this.  Don't attempt to spread it around or anything.

 

After market coolers can do a much better job than the ones bundled with CPU's.  Some of them are quite large, but it looks like that case will take up to a 180mm high one, which is huge, so you could go with something like the Coolermaster 212 Evo which is good value, and well reviewed.

 

Also it's really important to (as best as possible) keep the wiring in the case tidy and out of the way of airflow.  If you have room tie them under the motherboard tray or bundle them into an unused drive bay if not.  Poor airflow through the case will raise temperatures substantially.  From this it looks like it's a case where routing cables behind the motherboard tray should be possible.

 

thermaltake_chaser-mk-i_filled.jpg


7sbvuf-6.png





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users