Random restarts, no BSOD


Recommended Posts

So I recently built a new system and it's worked fine for a few weeks, but now I'm getting some crazy instability and I don't know what the cause is. The computer just randomly restarts, no BSOD - although I once had a BSOD stating "video_tdr_failure".

 

I looked at the Windows event logs and nothing, just an error stating an unexpected restart occured. Every time I run Furmark, OCCT, MemTest, etc., temps are normal, no reboot. But sometimes it'll occur when I'm browsing the web, or writing in Visual Studio, or playing a game, there doesn't seem to be any connection between what I'm doing and those restarts. It even restarted once when it was doing nothing, I was away and my account locked.

 

When it restarts, the screen goes black, sound continues playing for a few seconds. Also often these happen several times in a row, i.e. when it has restarted there's a high chance it'll restart again right after I get into Windows again.

 

I've updated all drivers to the latest as well as the BIOS. I loaded the optimized defaults to make sure my overclock wasn't the cause. This did nothing, and I was kind of expecting this because the computer was running fine before with these drivers. 

 

The only things I've changed recently were the PSU - I've switched to an XFX Pro 750W Black Edition because my Cooler Master 620W Silent Pro M2 had its fan spinning 100% all the time, and replaced the power strip which had started buzzing. I'm not excluding the PSU but at the same time, I can load the system 100% (OCCT Linpack + Furmark) and it doesn't cause reboots.

 

Gonna try running MemTest all day tomorrow, for what it'll be worth.

 

Configuration:

i5 4670K

Gigabyte Z87X-UD3H

2x Kingston DDR3 1600mhz 8GB

Samsung 840 Pro 256GB SSD

Western Digital Red 2TB Hard drive

2x GTX 760s in SLI

XFX Pro 750W Black Edition

Fractal Design Define R4

Monitor: BenQ XL2720T 120hz

Link to comment
Share on other sites

Well at least I still have the old noisy PSU around so I'll be able to test with it. It's super tedious to diagnose this because sometimes it seems to be fine for a day or more before rebooting, it's just completly random.

 

One thing that might be my fault is I've installed the SSD behind the motherboard plate on the case, and this prevents me from moving it anywhere else without removing the motherboard. However the way my new PSUs SATA connectors were designed, it was hard to get it to connect to the SSD, there just wasn't enough space. I removed the plastic cap on the SATA connector to free some space... not sure if that could cause issues with the PSU.

Link to comment
Share on other sites

Switched to my old noisy PSU, stable so far. Registered at XFX for an RMA, I hope the service will exceed its apparently dreadful reputation. After that I'll have to also RMA the Cooler Master I'm currently using, since the noise is due to a faulty fan PWM control. Funny fun fun.

Link to comment
Share on other sites

It's possible it's the nvidia drivers, there's a bug currently that's been causing lots of headaches and freezing/rebooting people's computers after 36 hours. The bsod message you get seems to support this. Try reading through this https://forums.geforce.com/default/topic/549618/gtx-780-freezing-and-stuttering-after-about-2-days-on-/?offset=80

 

"...

Here is a list of some bugs which are either fixed or targeted for our WHQL driver release in early August:

-No display on some monitors with Korean panels (Fixed in WHQL driver)
-TDR after leaving GPU idle for 36 hours (Targeting fix in WHQL driver)
-Faceworks demo shows DIgital Ira character with yellow beard (Fixed in WHQL driver)
-If driver without GFE is installed -> over install the driver which has GFE bundled ,uncheck GFE -> driver fails to install (Targeting fix in WHQL driver)

..."

 

I always lock up after the 36 hours when I open Chrome, sometimes it recovers and says the video card stopped working other times not so much.

Link to comment
Share on other sites

It's possible it's the nvidia drivers, there's a bug currently that's been causing lots of headaches and freezing/rebooting people's computers after 36 hours. The bsod message you get seems to support this. Try reading through this https://forums.geforce.com/default/topic/549618/gtx-780-freezing-and-stuttering-after-about-2-days-on-/?offset=80

 

"...

Here is a list of some bugs which are either fixed or targeted for our WHQL driver release in early August:

-No display on some monitors with Korean panels (Fixed in WHQL driver)

-TDR after leaving GPU idle for 36 hours (Targeting fix in WHQL driver)

-Faceworks demo shows DIgital Ira character with yellow beard (Fixed in WHQL driver)

-If driver without GFE is installed -> over install the driver which has GFE bundled ,uncheck GFE -> driver fails to install (Targeting fix in WHQL driver)

..."

 

I always lock up after the 36 hours when I open Chrome, sometimes it recovers and says the video card stopped working other times not so much.

I haven't experienced any TDR and only one BSOD; the reboots happen sometimes right after booting into Windows and I've never left the computer idle for more than 24 hours. So, this description doesn't fit what I'm experiencing. 

 

Having swapped the PSU for the old one and not had any reboots since, I'm getting more confident that it was indeed the power supply.

 

It could also be an over-heating processor

Please read the OP. I stated that CPU and GPU temperatures were normal even under stress testing.

Link to comment
Share on other sites

Shipping the PSU back to XFX (from Montreal, Quebec to California, US) costed me 40$. The replacement had better work.  :/

Wow why so much? Canada Post rate increases are getting out of hand...

 

Also, be sure they don't charge you customs fees when your replacement arrives! UPS/FedEx loves to do this. If they do make sure they understand that this is illegal under NAFTA.

Link to comment
Share on other sites

I had a hard drive die on me a couple of months ago, for some reason it was turning my pc on (I don't know how, or why)

(Speaking of which I need to go back and request that thread closed)

After hours of troubleshooting, and advise from members here, I got no where, til one day, I started the pc and there was apparently no os on drive C:

Tested my motherboard, and it was doing what it was supposed to be. (I'll try to find the thread date) So I fell back to the hard drive.

Tried it on my son's machine and the drive spools up, but that was it, I'm not saying this is the same problem you're having, just wanted to mention the possibility of a dying mobo or hard drive

(https://www.neowin.net/forum/topic/1163950-weird-startup/)

This was back in July maybe it can help

Link to comment
Share on other sites

Hard drives are responsible for the remainder of these inexplicable problems, but as they've become very reliable, just as RAM has, I do believe power supplies are the leading cause. By nature they are just more volatile, and proofing them is much costlier than with hard drives, so manufacturers cut corners. Also consumers are less likely to pay a premium for power supplies, but for storage everyone wants a brand name.

Don't worry man, the replacement will be fine, your PC will be back and running in no time

Link to comment
Share on other sites

How did the RAM test go? I've learned from my years of system building to go to the motherboard's website and by RAM specifically rated for the mobo.

Link to comment
Share on other sites

I didn't even attempt doing a long memtest; after switching to the other PSU I haven't had any reboot or BSOD. It's rock solid. I'm really sure the PSU was the problem.

Link to comment
Share on other sites

  • 1 month later...
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.