Computer regularly crashing/freezing

[Squid]

OK, here's the issue I'm facing. My computer regularly crashes/freezes. Usually it happens within 5 or 10 minutes of booting up. A freeze is as it sounds and everything becomes unusable. Crashes involve BSDs of various forms, the error message might change and sometimes the BSD freezes and doesn't complete the memory dump. My mobo also occasionally gives me the error beep code that says I have a problem with my PSU.

Today I brought my PSU to work and plugged it into a PSU tester and it looks like it checked out OK and its currently hooked up to a computer at work and has been running fine for a while, but admittedly its not under as high a load as the computer does not have a graphics card and I've got an nVidia 9600GT card.

I'm looking for other ideas of what I can check before I give up on the computer.

[Totalheadache]

post us full specs but yeah testing your psu with another pc arent the same variables.

e.g psu brand. overclocking? gpu brand etc...

come to think of it if rest of pc is as old as the gpu...new build..most likely

[Darkpriest667]

OK, here's the issue I'm facing. My computer regularly crashes/freezes. Usually it happens within 5 or 10 minutes of booting up. A freeze is as it sounds and everything becomes unusable. Crashes involve BSDs of various forms, the error message might change and sometimes the BSD freezes and doesn't complete the memory dump. My mobo also occasionally gives me the error beep code that says I have a problem with my PSU.

Today I brought my PSU to work and plugged it into a PSU tester and it looks like it checked out OK and its currently hooked up to a computer at work and has been running fine for a while, but admittedly its not under as high a load as the computer does not have a graphics card and I've got an nVidia 9600GT card.

I'm looking for other ideas of what I can check before I give up on the computer.

PSU was a good idea. Is it still freezing now with the new PSU?


First suggestion take out all ram sticks but 1. And run memtest with each stick (only 1 stick in at a time) make sure your memory is stable. I suggest memtest86. It is free and you can make your own bootable CD. http://www.memtest86.com/

If all sticks test ok the second suggestion would be to flash your bios to the most updated version. If you're still getting freezes (which may or may not be related to bios voltages) third suggestion

Back up data and reinstall your OS.

[Totalheadache]

PSU was a good idea. Is it still freezing now with the new PSU?


First suggestion take out all ram sticks but 1. And run memtest with each stick (only 1 stick in at a time) make sure your memory is stable. I suggest memtest86. It is free and you can make your own bootable CD. http://www.memtest86.com/

If all sticks test ok the second suggestion would be to flash your bios to the most updated version. If you're still getting freezes (which may or may not be related to bios voltages) third suggestion

Back up data and reinstall your OS.

uhm psu isnt a great idea as it's hooked up to a different rig but if it's still working then yeah you can try what you said...(i.e a work pc most likely uses less power)

but yeah psu's tend to go when they go rather than be a bit iffy. still a better test would be a different psu...(not readily obtainable i admit).

i think you need to get your hands on another gpu somehow and test with that. you can of course just plug your psu up to a voltmeter to check it's doing what it should..

[Darkpriest667]

uhm psu isnt a great idea as it's hooked up to a different rig but if it's still working then yeah you can try what you said...(i.e a work pc most likely uses less power)

but yeah psu's tend to go when they go rather than be a bit iffy. still a better test would be a different psu...(not readily obtainable i admit).

i think you need to get your hands on another gpu somehow and test with that. you can of course just plug your psu up to a voltmeter to check it's doing what it should..

I agree usually when a PSU goes it goes. My suspicion is it is memory related or OS related. He's indicated he has the latest bios.


If you get a chance Squid the next time it blue screens write down as many of the #s of the blue screen code as you can.

[Squid]

uhm psu isnt a great idea as it's hooked up to a different rig but if it's still working then yeah you can try what you said...(i.e a work pc most likely uses less power)

but yeah psu's tend to go when they go rather than be a bit iffy. still a better test would be a different psu...(not readily obtainable i admit).

I can take home the psu from the work computer but its only 230W and not likely to be able to power my computer.

i think you need to get your hands on another gpu somehow and test with that.

I wish I had a spare GPU to do that with, but I don't.

you can of course just plug your psu up to a voltmeter to check it's doing what it should..

I already did that, I used a psu tester and it came back with nothing wrong according to the hardware guys at my office who I consulted with.

[Darkpriest667]

I wouldnt even use a 230W to power my phone much less a computer these days.


Yeah best to not test with that 230W

[Squid]

PSU was a good idea. Is it still freezing now with the new PSU?

My computer doesn't have a new psu, the existing psu is running in a different computer.

First suggestion take out all ram sticks but 1. And run memtest with each stick (only 1 stick in at a time) make sure your memory is stable. I suggest memtest86. It is free and you can make your own bootable CD. http://www.memtest86.com/

I forgot about memtest, I'll take my copy home from work and run that tonight.

If all sticks test ok the second suggestion would be to flash your bios to the most updated version.

I checked that a week or so ago and I'm already running the most recent version and asus's bios update tool won't let me flash it again, and their individual file to flash the bios manually won't work on a 64-bit OS, which I'm running.

If you're still getting freezes (which may or may not be related to bios voltages) third suggestion

Back up data and reinstall your OS.

I've been hoping to avoid this option, that was my last one, but backing anything up is a pain since my computer usually only stays on for about 5-10 minutes at a time.

[Squid]

No overclocking, the PSU is an OCZ 600W ModXStream Pro. The system has 8G of RAM and an Asus mobo and no other peripherals. The specifics of the mobo product num and video card manufacturer I'll post when I get home as I can't remember them right now.

[Totalheadache]

def try running some progs like coretemp/realtemp for cpu. msi afterburner for gpu.

old gpus can have dust and overheat.

run a program like occt or prime95 but then if your pc is crashing after ten mins these progs will likely shut off the pc within minutes.

if it's ten mins it could be an overheating issue...i.e cpu/gpu are reaching a certain temp and the pc will failsafe off.

[Crazyeyesreaper]

sounds like either PSU or MOBO / RAM

easy way to test if you dont feel like waiting on memtest,

load up your BIOs set your ram to the slowest setting your BIOS has so if your running DDR2 800 drop it to DDR2 667 or 533 etc if that fixes it well your memory isnt stable anymore and means you should run a memtest on each stick at its actual speed and see how they do, when done test 1 stick of ram in each DIMM see if its a particular DIMM thats gone faulty

also agree with total headache check temps just to err on the side of caution

could be age aka Motherboards are now manufacturered to fail after 3 years if your system is older then that well your system has lived past its expected life span its done this way to keep people purchasing products, but i doubt this to be the case for you.

everyones already stated the obvious

memtest , check dimms, lower Ram speed test for stability,
check temps
test with a different PSU,

also if running Vista /Win 7

check event viewer for logs on your BSODs see what it says there as well it may help us narrow down the issue

[Squid]

I agree usually when a PSU goes it goes. My suspicion is it is memory related or OS related. He's indicated he has the latest bios.


If you get a chance Squid the next time it blue screens write down as many of the #s of the blue screen code as you can.

I'll do what I can, but as I noted earlier the BSD is not always the same.

def try running some progs like coretemp/realtemp for cpu. msi afterburner for gpu.

old gpus can have dust and overheat.

run a program like occt or prime95 but then if your pc is crashing after ten mins these progs will likely shut off the pc within minutes.

if it's ten mins it could be an overheating issue...i.e cpu/gpu are reaching a certain temp and the pc will failsafe off.

I've checked for dust, and if I turn the computer on again immediately and check the temps in the BIOS they are normal. I'll provide exact temps the next time it crashes.

sounds like either PSU or MOBO / RAM

easy way to test if you dont feel like waiting on memtest,

load up your BIOs set your ram to the slowest setting your BIOS has so if your running DDR2 800 drop it to DDR2 667 or 533 etc if that fixes it well your memory isnt stable anymore and means you should run a memtest on each stick at its actual speed and see how they do, when done test 1 stick of ram in each DIMM see if its a particular DIMM thats gone faulty

also agree with total headache check temps just to err on the side of caution

could be age aka Motherboards are now manufacturered to fail after 3 years if your system is older then that well your system has lived past its expected life span its done this way to keep people purchasing products, but i doubt this

everyones already stated the obvious

memtest , check dimms, lower Ram speed test for stability,
check temps
test with a different PSU,

Good ideas, I'll do that when I get home.

[Totalheadache]

sorry i thought you took your psu into work and hooked it up to a work pc?

what am trying to say if your psu was iffy it might appear to be ok as it is connected to stuff with less power demands. this is very unlikely as psu's dont tend to work "sort of". and if you have used a voltmeter you have to look at another part i/e bad ram/overheating/mobofailure.

cpu's tend to be rare to fail but other parts like ram/mobo more common.

i'd def check your gpu. whip it out and check it's fan is ok...dust etc..

p.s your pc sounds about the same as my bros...it's almost a miracle his is still working.

sure nobody at work has another similar gpu?

[Totalheadache]

aye could that your tim on either cpu/gpu has gone...esp if you havent replaced.

that would explain a quick rise in temps and then shut off. you need to sit and get some monitoring apps installed and then sit and watch for crash.

but yeah am thinking its something more sinister than gpu/cpu overheating. you never know some paste can be pretty cained like.

[PoleCat]

You mention that you're occasionally getting a beep code from your MoBo? What's the code and what brand, make and model MoBo do you have? Are you sure it's indicating PSU instead of something else?

As everyone else has mentioned the obvious I'll ask the obscure question - what brand of keyboard are you using? Good luck!

[Squid]

You mention that you're occasionally getting a beep code from your MoBo? What's the code and what brand, make and model MoBo do you have? Are you sure it's indicating PSU instead of something else?

Well this goes with my previous promise to provide complete system specs:

PSU: OCZ 600W ModXStream-Pro
Mobo: Gigabyte GA-EP43-UD3L
GPU: EVGA nVidia 9800GT 512MB
RAM: 4G OCZ PC6400 RAM (2 x 2G) + 4G Kensington KVR800D2N6 (2 x 2G)
HD: Seagate Barracude 1T 7200RPM SATA

The code is a lot of (I think) short beeps, more than the 9 for any other code, it pauses and then repeats. If I'm right about short beeps it means a power issue, but given my testing of power today I might be wrong, if the beeps are long beeps it means a GPU not inserted correctly, which I doubt given I've reseeded it a couple of times.

As everyone else has mentioned the obvious I'll ask the obscure question - what brand of keyboard are you using? Good luck!

Benq wireless keyboard, it hasn't caused me problems and I've been using it longer than I've owned this computer.

[PoleCat]

Well this goes with my previous promise to provide complete system specs:

PSU: OCZ 600W ModXStream-Pro
Mobo: Gigabyte GA-EP43-UD3L
GPU: EVGA nVidia 9800GT 512MB
RAM: 4G OCZ PC6400 RAM (2 x 2G) + 4G Kensington KVR800D2N6 (2 x 2G)
HD: Seagate Barracude 1T 7200RPM SATA

The code is a lot of (I think) short beeps, more than the 9 for any other code, it pauses and then repeats. If I'm right about short beeps it means a power issue, but given my testing of power today I might be wrong, if the beeps are long beeps it means a GPU not inserted correctly, which I doubt given I've reseeded it a couple of times.

If you've reseated the GPU and you're still getting the problem then I'd rule that as off the list of things to check. However - maybe the slot on the mainboard that the GPU is plugged into is dying? Something to consider. Look below....

Benq wireless keyboard, it hasn't caused me problems and I've been using it longer than I've owned this computer.

The reason I asked about the beep code is that they can be very inconsistent indicators when it comes to diagnosing hardware faults. The older IBM codes list the PSU as a possible fault with three different codes. The problem is that the keyboard is also a possible fault on two of those codes as is the mainboard.

You say you've been using that keyboard since before you owned that computer? Have you ever looked into newer drivers? Maybe a new driver will cure the headache? Conversely, maybe Windows automatically updated the driver for that keyboard and the new driver is causing the problem? And who can say that it's the problem in the first place?...

The only cogent advice I can offer regarding keyboards - download and install the Intellitype driver. It's fixed some obscure problems with M2 in the past with wireless keyboards. Maybe it will do the same for you.

Squid, with most people I'd be suggesting the obvious solutions but I know you know your way around a computer. If you can't figure it out on first pass then I'd suggest looking at the more outre problems. Something just doesn't smell kosher here. Good luck!

[Taiji]

Would like to help but I need Stop error codes or I can't say anything

The code is a lot of (I think) short beeps, more than the 9 for any other code, it pauses and then repeats. If I'm right about short beeps it means a power issue, but given my testing of power today I might be wrong, if the beeps are long beeps it means a GPU not inserted correctly, which I doubt given I've reseeded it a couple of times.

Read the manual.

edit:

(Finds there is no manual for the board where it's meant to be on the website.) Phone them and ask.

[Squid]

I'm starting to lean towards a GPU issue of some sort. I did manage to get a second GPU, by taking apart my computer in my office, and I haven't had a crash yet. Its not conclusive as occasionally I do go a few days without a crash just not recently. I'm going to run with this other GPU for the weekend and see if that cures the problems, if so I'll need to get a new GPU or figure out how to fix this one..

[PoleCat]

I'm starting to lean towards a GPU issue of some sort. I did manage to get a second GPU, by taking apart my computer in my office, and I haven't had a crash yet. Its not conclusive as occasionally I do go a few days without a crash just not recently. I'm going to run with this other GPU for the weekend and see if that cures the problems, if so I'll need to get a new GPU or figure out how to fix this one..

If I may ask what is the new GPU?

Looking back a dying GPU could have caused all of your problems. Crashes to BSOD and other crap seems like memory related at first glance but it could easily be a byproduct of a bad filter though the graphics.

Keep in touch and good luck!