Solved Random freezes during Windows 7

[PenPen]

I used to have my computer running pretty fine until earlier this week, upon browsing on the interwebs, my computer suddenly restarted by itself without caution.

When it rebooted, I got a CMOS checksum error. To go further I continued with the rebooting, and Windows loaded without errors.

The computer seemed to run fine by then, but after some time it just froze (I wasn't playing any games with it, I'm just browsing with Firefox and maybe had Adobe Flash running in one of the tabs for YouTube). Tried to do Ctrl+Alt+Del but no response. I then tried to press the NumLock or Caps Lock key but the keyboard light didn't turn on or off.

I pressed the power switch to turn off the computer, and turned it on again. CMOS checksum error appeared again so I looked into BIOS. It actually froze in BIOS setup as well (while looking around), so I had to reboot. Entered BIOS again this time successfully, and changed my settings to failsafe instead.

The computer ran fine afterwards, so I did a chkdsk (no errors), the Windows memory diagnostic (no errors) and did the sfc /scannow thing...it had a log, which from what I read should involve my much earlier attempt of getting Windows Mail to work in Win7 (I can't fully tell if there are any other things wrong in there), but that was like 1 year ago and I have switched to using Thunderbird some time after that.

I also changed my CMOS battery just in case the battery was indeed dying, just to be safe. Did some cleaning of my computer's innards while I'm at it.

I also rolled back my NVIDIA drivers since about 1-2 weeks ago I installed the latest NVIDIA drivers, but the new driver crashes at times even when I'm not doing anything. I did a clean install of a previous version of the NVIDIA driver, thinking that the driver update was part of the reason as well.

After having this worked out, and the computer not crashing, I switched back my BIOS to the optimal setting and then played Borderlands, without issues, the computer didn't freeze or BSOD, and I can go back to Windows fine.

The computer ran fine until today where it just froze again. I rebooted, and to be safe I changed the BIOS setting to the failsafe one.

I haven't changed any hardware since I built the computer. The Win7 was upgraded from Vista and I migrated the HD to this new computer below. This build has been running for around a year or so without issues until now. My specs are below.

I really hope it's not a hardware or especially a PSU issue, though. Thank you!

OS: Windows 7 Home Premium 32bit SP1
CPU: AMD Phenom II X4 955
Mobo: MSI 870A-G54
RAM: 4GB DDR3
GPU: Galaxy NVIDIA GTX 560Ti
HD: Western Digital WD6400AAKS
PSU: Antec Neo ECO 620C

 

[usasma]

Why are you worried about your PSU? Most times we find that the user is the one who solves these problems - so it pays to listen to the user's concerns.

We'll need this info (even though you're not reporting BSOD's): http://www.sevenforums.com/crashes-debugging/96879-blue-screen-death-bsod-posting-instructions.html

Have you checked the date that the computer is set to in the BIOS - is it the correct date and time?

You're describing video driver errors - which usually may degrade into a BSOD if not fixed.
1 - update the drivers to the latest available from the nVidia website.
2 - follow the troubleshooting steps here (even though you're not reporting BSOD's): http://www.sevenforums.com/crash-lo...op-0x116-video_tdr_error-troubleshooting.html

 

[PenPen]

Thanks for the reply. I'm just hoping that I don't get hardware issues...hopefully it's just me being paranoid.

As for the dates and time, they're all current, even after I get CMOS checksum errors. I didn't need to change them.

I've attached the file with this post.

Must I update to the latest NVIDIA driver? I don't mind doing that, but the latest version crashed a few times for no reason which is why I wanted to rollback. To be safe, should I remove the current driver, and then install the new one, or just straight-up installing is good?

I'll also take a look at that link you posted as well.

Thanks again.

 

[seavixen32]

I accepted NVIDIA's latest driver update and discovered Windows Media Center no longer talked to my HDTV PCI card so I just rolled back to the previous driver and all was good again.

You don't have to update to the latest driver and I firmly believe they haven't been fully tested before being released. If you don't want to install the driver just close the prompt. If it's offered through Windows Update and you don't want it, just hide it.

If you custom install your older NVIDIA driver, it will automatically uninstall the one you're replacing, and then carry out a clean install.

 

[PenPen]

Well to be honest, I disabled the NVIDIA driver update notification a while ago. I just instead go to NVIDIA's page once in a while to get updates.

Still I'd be curious about it as my friends updated the drivers as well and they're fine. I wonder if it's just part of the puzzle to my Windows freezing for no reason, so I added that bit as well.

 

[seavixen32]

Depending whether you self-built your computer or bought it from a computer manufacturer you are probably better off using the computer manufacturer's drivers as OEM drivers can be configured differently to NVIDIA's drivers, hence the occasional freeze.

 

[PenPen]

Well my computer is self-built, yeah.

It froze again today. The computer doesn't respond to my key presses (Ctrl-Alt-Del or any lock keys I pressed has no response) or anything and I had to turn off the computer thru the power switch.

I just updated the driver to the latest version. Going to run Furmark first...and GIMPS later.

Was chatting to a friend about another thing I noticed with my computer as well. I have SpeedFan installed and it says my temperature for Temp1 and Core are both around 45-46 degrees normally. If I do anything more intensive with my computer, it shoots to 50 degrees or higher. My friend says his is usually around 38 degrees when it's not doing anything.

I haven't ran SpeedFan with a game, but I'd presume that it'll go up to the 60s. Does the high temperature play a factor in my random freezes as well?

 

[PenPen]

This is bad.

Okay, I ran Furmark, there's no stability test so I ran the 1080 and 720 benchmarks. They ran fine without problems, the GPU logged the highest temperature at 73 degrees.

I then did a 15 minutes 1080 test, it was running until 49% where my GPU went to 83 degrees and my computer froze. No response to whatever I did. So I had to reboot thru the power button.

The computer now freezes earlier when it goes to the "run safe mode" prompt when loading. After booting for a few more times, my monitor can't even detect any incoming graphic signals.

I'm kinda stuck here...

Edit:

Some further adventures, I unplugged everything in my computer, left the computer to cool down by itself and re-inserted the graphics card. I also reset my CMOS by taking out the battery and reinserting it after a few minutes (had to reset the clock as a result). I also left my computer door open to see if it's really a ventilation issue. So far so good for now.

 

[PenPen]

Well it appears that either my graphics card is dead since it's running, but the monitor detected no input, even if I used other DVI or the hdmi port. Or maybe my motherboard screwing up something.

Unfortunately I don't have a spare card or an onboard video output, so I'll see if my friends have one and go from there.

 

[PenPen]

Eventually I took it to the local mobo distributor and had my computer checked by them around 1 week ago. Turns out one of my RAM was bad, so I got them replaced.

Computer is running fine since I got the RAM replaced, but I surmised the ventilation might be a problem, so I added a case fan to help bring in cool air.

Computer has been working nicely since then. I consider this as solved (by myself?). Thanks.