Two computer problems

March 12, 2012 2:50 PM

So I've had this problem since I built my computer a few months ago, it is completely sporadic and seemingly random. Sometimes my computer won't freeze for days, other times shortly after start up. When it freezes everything stops, the keyboard and mouse are unresponsive (no Ctrl-alt-del etc.) and if sound is playing there is a sort of "glitch" then silence. I don't know yet whether this is hardware related or software related. The other problem (separate then the freezing) I've been having on startup: I can't use it normally at all, because when I start a program like firefox or anything it else it freezes and I have to hard restart every time, or I have to wait 30 minutes and then firefox or other programs will start. Or the other issue arises where no programs will start and I'm forced to hard restart again. I can only run it working in safemode with networking, which prevents me from doing what I built it for, gaming and videos etc. Running chkdsk /r fixes this problem, but it sometimes returns. This is a secondary problem unrelated to the first.

Before you respond, see what I've tried for the freezing problem:
- I've checked the temperature for cpu etc., it is normal (voltage range of the ram as well)
-I've run Memtest x86, it detected no problems
-I've had a suggestion of getting a higher watt PSU, but others have said 750 should work fine (I may do this as a last resort)
-I've posted this about it on numerous other tech forums with no results

My specs are as follows:
ASUS Black Blu-ray Drive SATA Model BC-12B1ST/BLK/B/AS - OEM

XFX HD-697A-CNFC Radeon HD 6970 2GB 256-bit GDDR5 PCI Express 2.1 x16 HDCP Ready CrossFireX Support Video Card with Eyefinity

COOLER MASTER HAF 932 Advanced Blue Edition RC-932-KKN3-GP ATX Full Tower Computer Case with USB 3.0, Black Interior and ...

Intel Core i7-2600K Sandy Bridge 3.4GHz (3.8GHz Turbo Boost) LGA 1155 95W Quad-Core Desktop Processor BX80623I72600K

CORSAIR Enthusiast Series TX750 V2 750W ATX12V v2.31/ EPS12V v2.92 80 PLUS BRONZE Certified Active PFC High Performance Power ...

Intel 320 Series SSDSA2CW080G310 2.5" 80GB SATA II MLC Internal Solid State Drive (SSD) - OEM

Western Digital Caviar Black WD1002FAEX 1TB 7200 RPM SATA 6.0Gb/s 3.5" Internal Hard Drive -Bare Drive

ASUS P8Z68-V PRO LGA 1155 Intel Z68 HDMI SATA 6Gb/s USB 3.0 ATX Intel Motherboard with UEFI BIOS

Patriot Viper Xtreme 8GB (2 x 4GB) RAM

posted by johnx to Computers & Internet (14 answers total)

Those symptoms mean check the RAM first. Pull out the Blu-Ray and try. Pull out one stick and try. Then replace that stick and try. Try a 2gb stick alone if you have one.

If those all work, there are only 1,000,000 other things to try, but I would check that RAM first.
posted by caclwmr4 at 3:03 PM on March 12, 2012

It really stinks of the memory. I would want to run memtest for several hours, maybe even a 24 hour test if feasible.

I would run Prime95 on a minimum 4 hour test to check for CPU and overheating issues, although I think this is secondary to the memory.

Assuming no issues with the PSU, the 750 Watt PSU is more than enough for your system as the PSU calc indicates you need 550 or so given your setup.
posted by Mister Fabulous at 3:08 PM on March 12, 2012

Response by poster: Didn't anyone see that I ran memtest though?
posted by johnx at 3:15 PM on March 12, 2012

All memtest is good for is finding a problem. If memtest (or any other) says there's a problem, there is. If it shows everything is OK, that does not mean it is! It doesn't test like Windows or Linux actually works.
posted by caclwmr4 at 3:20 PM on March 12, 2012

Those are the symptoms I would have when my motherboard would blow a capacitor. It is really hard, however, to prove that a board has gone bad without going and buying another board.

Seconding that you need to run memtest for a LONG time to catch more random errors that only show up occasionally.

Run the Western Digital tool for testing the hard drive to determine if it has a problem.

Finally, your power supply might be rated at the right wattage, but that doesn't mean it isn't somehow broken.
posted by gjc at 3:44 PM on March 12, 2012

Yeah, when I said "It really stinks of the memory" I meant it in that you need to run memtest for an extended period of time to hit the failure. I've had memtest run for 24 hours cleanly on memory that ended up being the culprit, but was only spotted on the second 24-hour run.
posted by Mister Fabulous at 3:59 PM on March 12, 2012

I've also had problems with outdated video card drivers--it might be worth checking ATI to see if there's a more recent version. (For example, StarCraft 2 used to cause an insane heat spike and would kill my computer within like 10 seconds until I updated the drivers.)
posted by fifteen schnitzengruben is my limit at 4:41 PM on March 12, 2012

And when I said "try" I meant more try using it normally, than memtest. Memtest was a good thing to do first, but it won't always pick up the problem fast. You could run it 7 days straight before it picks up a problem, a problem which you might find in normal use right away. Or not right away.

Aren't computers fun.
posted by caclwmr4 at 6:08 PM on March 12, 2012

Two things to try:

1) remove all unnecessary components and uninstall their drivers (blu-ray, video). Update all other drivers. Test the system for several days/weeks. If it fixes the problem then add components back in one at a time until the problem occurs again.

2) Run a liveCD (such as ubuntu) for a few days to see if that resolved the problem. If it does, it's a software issue, if it doesn't, it's probably a hardware issue.

It's also possible that your SSD is failing; try another drive if you have one.
posted by Sonic_Molson at 6:20 PM on March 12, 2012

As everyone agrees, it sure sounds like a hardware problem. Oh, you bought brand-new quality components and of course you expect them to work right out of the box, and most often they do. But I think you have one or more bad components. This happens.

Hopefully you have some (known-to-be-good) spare components lying around (or a working PC you can disassemble) so you can swap them into your new PC. Just keep replacing components until everything works stably.
posted by exphysicist345 at 7:33 PM on March 12, 2012

I'm going to chime in that this sounds to me like an SSD issue. These things are still very unstable IMHO. Is your OS installed on the SSD? If so, try installing the OS on the HDD and see if the problem persists.

Also... do you never get crashes in safe mode? This points at something else. But my money is still on the SSD.

You don't mention this in the question, but have you updated all your drivers? I would especially go to the ASUS page for your motherboard here and make sure all your drivers are updated (particularly the Intel RST).
posted by chicxulub at 6:32 AM on March 13, 2012

I had a similar problem about 3 years ago with a PC I built at home too. It would randomly freeze up, sometimes every day, sometimes not for weeks. I would get blue screens. I would need to run chkdsk. Memtest wouldn't find any problems after running for 4 days straight. Then a month later it would find a problem in the first 10 minutes of running. You say that the freezing is unrelated to the non-responding programs at startup, and unrelated to the need to run chkdsk /r, but that may not be true.

In my situation, I finally stripped my rig down to the individual components and started rebuilding piece by piece to see where things started malfunctioning. What I discovered was that I had accidentally screwed an extra riser into the motherboard tray where it didn't belong. It was making contact with the back of the motherboard and causing intermittent IO problems. Thus, issues with hard disk corruption, occasional memtest errors, video and audio glitches, and full-on BSODs.

Removing this riser, reinstalling all my components and then reformatting and reinstalling all the software solved the problem. I'm still running that rig with all the original gear, except the video card. It seems that was the only part possibly compromised by that mistake, it died a few months after I fixed the problem.

Even if your problem isn't exactly the same, try taking out each component and reseating one at a time. First just put your motherboard and RAM back in, see if it POSTs correctly. Then install your optical drive and run Memtest for at least a week straight. Then run a Live CD designed to tax your CPU setup to make sure the motherboard and processor are solid. Then plugin your hard drives, but run a Live CD of any flavor and stress test the hard drives. If everything comes out OK at that point, you should install your standalone video card and re-run all the above tests plus a stress-test of video capability. Only then should you feel comfortable installing your OS and software. Obviously don't restore any backups made with your current rig, since any underlying corruption in files not caught by chkdsk will carry over to your clean build.

Good luck!
posted by trivia genius at 8:12 AM on March 13, 2012

There is no good reason why chkdsk should return errors on a new disk. Disk and memory integrity problems only get worse. They never get better. NEVER!

Sure it can: if the operating system is always freezing up, chances are it is going to happen during a write some of the time, and when that happens the filesystem gets messed up. You can have a corrupted filesystem on a perfectly good disk.
posted by gjc at 7:15 PM on March 13, 2012

I had really similar problems with, guess what: Corsair RAM. They were defective and I've had no problems since RMA'ing and getting a different brand.
posted by tremspeed at 7:52 PM on March 13, 2012