Sign in with
Sign up | Sign in
Your question
Solved

HELP: Random System Crashes (no BSOD)

Last response: in Windows 7
Share
September 19, 2011 6:59:11 PM

Hey All,

I am experiencing some disturbing shut downs on my pc and would really appreciate some guidance. This has been plaguing me for about 6 months now and I can't take it anymore...

Just to be clear, these crashes do not provide a BSOD and therefore no error code for me to research and correct on my own.

In the last two days, this has happened 4-5 times, and most often while watching streaming video online (but not always). Without warning, everything will lock up (sometimes this is accompanied by a terrible sound emitted from the speakers - presumably looping audio from the video i was watching). The computer then reboots itself without my involvement.

Before anyone asks, my drivers are UP TO DATE, and my BIOS is up to date as well.

So far, I have asked around and most people point to my graphics card overheating, or a very dirty (dusty) pc as a likely cause. Firstly, my computer is generally kept very clean, and I am monitoring my GPU temps and they are usually WELL within normal ranges (frankly it runs very cool). I have called EVGA for support, and they told me they had not heard of this problem before. They had me run a bunch of gpu stress tests, all of which provided no clue (they were clean) as to what could be wrong. Someone else told me to make sure my RAM was properly seated. I did check that and I can't see any seating issues.

Any ideas? PLEASE help me! :fou: 

Thank you in advance.

OS. Windows 7 HP 64Bit
CPU. Core i5 750 2.66
Motherboard. Gigabyte GA-H57M-USB3
Memory. Mushkin Enhanced Blackline Frostbyte 6GB 3X2GB PC3-12800 DDR
Graphics Card. NVIDIA GeForce GTX 460
Monitor. Displays ACER LCD 24" 2MS H243H BMID
PSU. Corsair TX750W 750W ATX 12V 60A 24PIN ATX Power Supply
Case. Antec Three Hundred Gaming Case ATX 3/0/(6) 2xUSB Audio
Cooling. Arctic Cooling Freezer 7 Pro REV.2 CPU Heatsink Cooler LGA13
Hard Drive. Western Digital Caviar Green (WD10EARS) 1000GB (1TB) SATA 3 Gb/s 64MB (OEM)

Best solution

September 19, 2011 7:04:42 PM

It could be a heat issue with the CPU, you said you monitored GPU temps, what were your CPU temps? The max CPU temp is alot lower than the max GPU temp, so it could be overheating the CPU and shutting off the computer especially because videos and such use the CPU more so than the GPU unless you have GPU acceleration enabled.

Other things it may be:
Could be a RAM issue, check your ram with memtest86+ or another RAM tester to make sure one stick of RAM isnt bad. Anything below a perfect score on RAM could cause your computer to intermitantly lock up.

Could be an unstable overclock on the CPU, aka not enough voltage, cooling, etc...

Could be a bad PSU, check the voltages to make sure thay are not dipping below the regular amount, a bad power supply can cause a computer to shut off if it cant give the right steady voltage.

Could be a software issue, or conflict. Might even be as silly as an IRQ conflict which rarely if ever happens.

Could be the motherboard is overheating, if the northbridge or southbridge get to hot it can lead to system instability.
Share
September 19, 2011 7:06:59 PM

Thanks Mighty. How would I check CPU temps? Would I find any evidence anywhere that would point to this? I do have an aftermarket cooler on my CPU, and I don't overclock so I didnt think of this initially, but it's possible I suppose.

Please advise.
m
0
l
Related resources
September 19, 2011 7:11:07 PM

Check your ram settings, looking at your description seems to be bad settings on the ram or cpu, check it is not overvolted and that timings are within manufacturers specs.
m
0
l
September 19, 2011 7:11:56 PM

You can check the CPU temperatures by using a program called speedfan, CPU-Z, speccy, or coretemp to name a few. If theres air bubbles when you put the thermal paste on, it can cause overheating because the transfer of heat wouldnt be as efficent.
m
0
l
September 19, 2011 7:14:40 PM

mightymaxio said:
You can check the CPU temperatures by using a program called speedfan, CPU-Z, speccy, or coretemp to name a few. If theres air bubbles when you put the thermal paste on, it can cause overheating because the transfer of heat wouldnt be as efficent.


Thanks again. I forgot CPUz did that, I have it installed already. I'll check it as soon as I get home and get back to you!

Edit: Forgot to add that I did not spread the paste myself (Though i do have some arctic silver if needed). It was pre-applied on the CPU fan i bought (Arctic Cooling Freezer 7 Pro REV.2 CPU Heatsink Cooler LGA13).
m
0
l
September 19, 2011 7:16:46 PM

_Pez_ said:
Check your ram settings, looking at your description seems to be bad settings on the ram or cpu, check it is not overvolted and that timings are within manufacturers specs.


These values are presumably at their defaults, but I should check.

Can I do all this in CPUz?

thank you very much.
m
0
l
September 19, 2011 7:34:25 PM

I think you can tell the voltages through CPU-Z if not i would check speccy, or speedfan.
m
0
l
September 19, 2011 7:50:45 PM

mightymaxio said:
It could be a heat issue with the CPU, you said you monitored GPU temps, what were your CPU temps? The max CPU temp is alot lower than the max GPU temp, so it could be overheating the CPU and shutting off the computer especially because videos and such use the CPU more so than the GPU unless you have GPU acceleration enabled.

Other things it may be:
Could be a RAM issue, check your ram with memtest86+ or another RAM tester to make sure one stick of RAM isnt bad. Anything below a perfect score on RAM could cause your computer to intermitantly lock up.

Could be an unstable overclock on the CPU, aka not enough voltage, cooling, etc...

Could be a bad PSU, check the voltages to make sure thay are not dipping below the regular amount, a bad power supply can cause a computer to shut off if it cant give the right steady voltage.

Could be a software issue, or conflict. Might even be as silly as an IRQ conflict which rarely if ever happens.

Could be the motherboard is overheating, if the northbridge or southbridge get to hot it can lead to system instability.


Quick question for you, what tools are used to check a psu? actual v/w meeters or a software? I have an issue on an existing system that will randomly "turn off" essentially. Like some one pulled the plug on it. Once this happens I cannon't restart the computer for several minutes at the minimum. I did a test of unpluging the power cable and waiting 10~ seconds before trying to start it up to see if it was the mobo protecting its self and it still wouldn't restart till i waited for around 5~ min. I would like to note that this issue didn't accure till after I travled from the US to Chile. Problem is present on both power settings, 220 is standard in Chile. 110 required a transformer; us standard.
m
0
l
September 19, 2011 8:03:16 PM

I would buy a PSU tester, they cost 20$ USD and are available at pretty much any computer shop even bestbuy carries them.
m
0
l
September 19, 2011 8:17:57 PM

mightymaxio said:
I would buy a PSU tester, they cost 20$ USD and are available at pretty much any computer shop even bestbuy carries them.


I don't know if that was addressed to me or not, but I think you missed the part where I said I am in Chile now. No best buys, no Fry's electronics. A link to one would be nice though, as I might get lucky and somewhere in the city a Store might have one.

On a side note... we do have MANY apple stores and some one said that their might be a radio shack or 2.
m
0
l
September 19, 2011 8:23:12 PM

mightymaxio said:
You can check the CPU temperatures by using a program called speedfan, CPU-Z, speccy, or coretemp to name a few. If theres air bubbles when you put the thermal paste on, it can cause overheating because the transfer of heat wouldnt be as efficent.


Mighty,

I have just booted up and started streaming video for about 5 mins. Here is what I see in SpeedFan for temps:

GPU: 31C
Temp1: 29C
Temp2: 25C
Temp3: 25C
HD0: 26C
Core0: 35C
Core1: 30C
Core2: 37C
m
0
l
September 19, 2011 8:26:57 PM

Ichy said:
I don't know if that was addressed to me or not, but I think you missed the part where I said I am in Chile now. No best buys, no Fry's electronics. A link to one would be nice though, as I might get lucky and somewhere in the city a Store might have one.

On a side note... we do have MANY apple stores and some one said that their might be a radio shack or 2.


Try googling PSU paperclip test, that might help you, and you can do it anywhere.
m
0
l
September 19, 2011 8:41:37 PM

No your all wrong, hes using tri channel kit with a dual channel board, that ram isn't supported. Please purchase the correct ram for your board.

the ram you're using is for a X58 tri channel board and your chipset is a P55

Move along message board kiddys! this geek just solved your highly over looked issue everyone makes so much.


$51.99 shipped

G.SKILL Ripjaws Series 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1600 (PC3 12800) Desktop Memory Model F3-12800CL9D-8GBRL
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
m
0
l
September 19, 2011 8:47:02 PM

Are you sure GSX? I'm using this:

http://ncix.com/products/index.php?sku=52648
Looks like dual channel to me (although someone on another board said that i shouldnt be using 3 sticks in a dual channel setup...)

Please advise GSX
m
0
l
September 19, 2011 9:05:14 PM

hey buddy.

Had quit a familiar problem very similar to yours. It wasn't actually video, surprisingly it was sound (specifically my Realtek HD audio)

Just to test it out, i disabled my audio via BIOS setup, and installed a x-fi sound card. My Freezing (with the crazy sound) stop. try it out.
m
0
l
September 19, 2011 9:09:11 PM

rozz said:
hey buddy.

Had quit a familiar problem very similar to yours. It wasn't actually video, surprisingly it was sound (specifically my Realtek HD audio)

Just to test it out, i disabled my audio via BIOS setup, and installed a x-fi sound card. My Freezing (with the crazy sound) stop. try it out.


Thanks Rozz, I might give that a try. Very good advice.

I want to verify this memory issue first and then i'll try your route.
m
0
l
September 19, 2011 9:30:35 PM

also - if you are having issues with your RAM not being supported by your mobo - (tripple channel ram as opposed to dual channel) just pull one stick out ;) 
m
0
l
September 20, 2011 12:00:09 AM

Thanks Rozz.

I have removed one stick, now down to 2x2GB sticks.

How much of an impact will I see if i ditch my current setup and get 2x4gb sticks (total 8gb)? RAM seems really inexpensive now, the prices are way lower than i remember.

What brand would people recommend?

G.Skill
Mushkin
Patriot

thanks.
m
0
l
September 20, 2011 12:40:09 AM

I recommend either G.Skill or Mushkin. Usually Mushkins have way higher overclocking potential imo.
m
0
l
September 20, 2011 12:57:46 AM

Theres really no easy way to test the PSU without having at least a multimeter although by far the easiest way to test is to get a power supply tester. If you have a multimeter then you can individually check the rails by hand. Heres a guide: http://forums.extremeoverclocking.com/showthread.php?t=...

I wonder if it could be the hard drive going bad too, just another possibility. I would test with Hdat2 or Vivard for any disk errors you can get both from either Hirens boot CD or UBCD(Ultimate boot cd) both are free from their websites.

I have a feeling its a power supply issue but its hard to determine the root cause without actually being there imo. Each computer is unique in some ways of troubleshooting.
m
0
l
September 20, 2011 1:23:52 AM

Yes, I wish you were here. I'll try calling Corsair tomorrow, if not Wednesday.
m
0
l
September 21, 2011 2:00:27 PM

mightymaxio said:
Theres really no easy way to test the PSU without having at least a multimeter although by far the easiest way to test is to get a power supply tester. If you have a multimeter then you can individually check the rails by hand. Heres a guide: http://forums.extremeoverclocking.com/showthread.php?t=...

I wonder if it could be the hard drive going bad too, just another possibility. I would test with Hdat2 or Vivard for any disk errors you can get both from either Hirens boot CD or UBCD(Ultimate boot cd) both are free from their websites.

I have a feeling its a power supply issue but its hard to determine the root cause without actually being there imo. Each computer is unique in some ways of troubleshooting.


Hey mighty,
So those diagnostics for the HDD run off a boot disk, am i understanding correctly? I'll look into it.
m
0
l
September 21, 2011 2:38:28 PM

Read the guide on testing voltages. I think i could pull it off if I needed to but should eliminate other things first I think.

After testing HDD and RAM, and eliminating them as culprit. I will test with onboard GPU to see if its the GFX card. After that... it will either be mobo or psu ...

Does that order make sense?

If ram is not culprit, should i put 3rd stick back in, or keep it out?
m
0
l
September 21, 2011 2:50:29 PM

I would put the 3rd stick back in, and then test again because it could be the third stick is bad.
m
0
l
September 22, 2011 2:11:27 PM

Thanks Mighty. You are my hero btw.

Let's assume (just for now) that the third stick of ram is causing the problem. Is this a normal by product of using a third stick, or is something actually defective in that stick?

Seems surprising to me that simply having a third stick would cause such major problems...
m
0
l
September 23, 2011 7:04:54 PM

A third stick could cause problems but its not likely, i have used 3 sticks of ram and have not had issues but it could still be a bad stick because that was excluded from the test.

Trying to narrow down possibilities for rebooting, you could try removing the 460 gtx and just booting from integrated for a while just to see if the problem occurs again because the only thing you really havent tested is the GPU.
m
0
l
September 28, 2011 12:53:54 PM

Hey Mighty,

So far I have been stable, not a single crash, for a whole week. This means RAM was the culprit. I am down to two sticks of 2gb and running smoothly. Do you suggest that I put the third one back in and fix the voltages etc, or simply stick with two with current settings.

Are you able to help me look at my voltages and whatnot, and tell me if they look right?
thanks
m
0
l
September 28, 2011 12:54:38 PM

Best answer selected by dio_free.
m
0
l
!