Jump to content


 


Register a free account to unlock additional features at BleepingComputer.com
Welcome to BleepingComputer, a free community where people like yourself come together to discuss and learn how to use their computers. Using the site is easy and fun. As a guest, you can browse and view the various discussions in the forums, but can not create a new topic or reply to an existing one unless you are logged in. Other benefits of registering an account are subscribing to topics and forums, creating a blog, and having no ads shown anywhere on the site.


Click here to Register a free account now! or read our Welcome Guide to learn how to use this site.

Photo

Computer Crashing - Video Card?


  • Please log in to reply
16 replies to this topic

#1 RichieP

RichieP

  • Members
  • 86 posts
  • OFFLINE
  •  
  • Local time:04:06 PM

Posted 20 April 2006 - 11:31 AM

I'm helping a friend sort out his computer. It keeps crashing for no reason. I turned off the reboot after serious error option to see if it got an error message but it still shuts down randomly.

I changed the power supply for one of mine to see if that helped. It was OK for a while but still keeps cutting out. there's nothing in the Event Viewer to give any clues.

I thought it may be the graphic card, as I've read quite a few problems relating to display drivers. It's got an Nvidia GeForce 6600GT card installed in a PCI slot. System Information for Windows shows the following, amongst other stuff:

Property Value
InfPath oem14.inf
InfSection nv4_NV3x
ProviderName NVIDIA
DriverDateData 00 40 1F 20 F5 96 C5 01
DriverDate 8-2-2005
DriverVersion 7.8.0.1
MatchingDeviceId pci\ven_10dedev_0140
DriverDesc NVIDIA GeForce 6600 GT
DeskCheckForDuplicates 1
CoInstallers32 nvcod.dll,NVCoInstaller

The motherboard is an MSI:

Property Value
Manufacturer MICRO-STAR INTERNATIONAL CO., LTD
Model Alderwood

Chipset Vendor Intel Corporation
Chipset Model 925X/XE Host Bridge / DRAM Controller (B1 Step)
South Bridge 82801FB/FR ICH6/ICH6R LPC Interface Bridge
SMBus Intel Corporation 82801FB ICH6 SMBus Controller @0500

CPU Intel Pentium 4 HT
Cpu Socket LGA775 [Socket 478]
Processor Upgrade ZIF Socket

System Slots 4 PCI

Memory Summary
Capacity 2048 MBytes
Location System board or motherboard
Maximum Capacity 4096 MBytes
Memory Slots 4
Error Correction 8-bit Parity
Error Correction Capabilities None
Name Physical Memory Array
Use System memory
Maximum Memory Module Size 1024 MBytes

Warning! Accuracy of DMI data cannot be guaranteed


I have found a page for downloading drivers, hopefully updating the current one, but not sure which I need, if any. They're on this page

Thanks.

BC AdBot (Login to Remove)

 


#2 stevealmighty

stevealmighty

    Bleepin' WormBreath


  • Members
  • 2,629 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Upstate NY
  • Local time:11:06 AM

Posted 20 April 2006 - 03:09 PM

Are all of the fans working for the entire machine? Check the case fans, the fan on the 6600gt, the fan on the power supply (it might even have 2 on it), and specially the fan on the processor. If one of the fans isn't working, then the system might be overheating and shutting it down at the power supply which cuts power to everything else (obviously!). If the cpu is over heating, it'll restart or shutdown to keep from damaging the hardware.

If your power supply has a switch on the back for selecting fan speeds (L, M, H), try setting it to H (High) to see if it's a cooling issue with the psu itself.

Hope this helps!
War produces veterans, wounded both physically and mentally. They have sacrificed for us.....and it is now our job to help these veterans, as they have already helped us in ways we will never know, in ways that we cannot fathom, and in ways that we take granted every day.
Posted Image

#3 RichieP

RichieP
  • Topic Starter

  • Members
  • 86 posts
  • OFFLINE
  •  
  • Local time:04:06 PM

Posted 20 April 2006 - 03:20 PM

Thanks for the reply Steve.

I've downloaded Speedfan and it tells me Temp1 is only 32C. Problem is, it also tells me Temp2 is 99C, and Temp3 is 128C. I don't know which is which.

Anyway, it's been running a virus scan for the past hour and a half, the temperatures have stayed constant around these marks so I don't think the processor is overheating, as it's not crashed yet.

Ooops, forgot to mention, I've taken a stick of RAM out to see if it's that.

I know faulty RAM can cause these symptoms, but what also happens when it shuts down, is that it won't power back up until I've taken the power lead out and then put it back in again.

Aren't computers weird?

#4 stevealmighty

stevealmighty

    Bleepin' WormBreath


  • Members
  • 2,629 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Upstate NY
  • Local time:11:06 AM

Posted 20 April 2006 - 05:20 PM

Hmmm....well, faulty RAM will do that. Ram can also overheat too.

If you've taken out a stick of ram, and the problem is solved, then try putting the ram back in and unplugging something else (like an extra cd drive if you have 2 of them, or an extra hard drive if you have 2, last case scenario is to unplug a cooling fan). If the problem goes away again, then it's a power supply problem. Either it's to small (not enough power to supply power to entire computer) or it's overheating.

Oh, and I don't believe that speed fan monitors the fan on your gfx card, or the fan on your power supply. I'd check them visually to ensure that they're working correctly.

Without looking at your machine, I can't say what your temperatures are for (cpu, case, gfx card etx.). I can tell you that that seems a bit high though. When I run a graphics intense game like battlefield 2, it only gets up to about 56 c in the case.

I'm going to stick with the overheating issue as the source of the problem. That and the under power issue. That and.....lol, kidding.

Maybe we can get someone else to jump in and talk about the temps of your computer, or offer other advice....

Edited by stevealmighty, 20 April 2006 - 05:22 PM.

War produces veterans, wounded both physically and mentally. They have sacrificed for us.....and it is now our job to help these veterans, as they have already helped us in ways we will never know, in ways that we cannot fathom, and in ways that we take granted every day.
Posted Image

#5 JPHarvey

JPHarvey

  • Members
  • 508 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Sydney, Australia
  • Local time:03:06 AM

Posted 21 April 2006 - 05:24 AM

Additionally, your Video Card may be overheating. I had purchased a brand new system with fans galore. Yet whenever I crank up the settings for gaming (I have an ATI Radeon) it would go into VPU recovery (so that the system didn't crash). I don't know if nVidia have the same thing so I guess the question is that is it crashing when the video is challenged?
Anyway, although my video card fan was working fine, I swapped it out for an ArcticCool aftermarket fan, and now NO DRAMAS!! If that helps you, I am unaware!! :thumbsup:
[CPU]Intel E6600 Core 2 Duo @ 3.19GHz
[MoBo]ASUS P5N32-SLI Premium (nForce590)
[RAM]4GB Corsair XMS2 DDR2-800 CL4 @ 710MHz
[GPU]XFX 8800 GTX 768MB [SLI] @ Stock
[PSU]CoolerMaster 1kW
[Audio]ASUS Xonar D2
[Case]Antec Nine Hundred
[OS]Windows Vista Ultimate 64
[LCD]SAMSUNG 226BW
[Other]WC'd CPU & SLI

#6 RichieP

RichieP
  • Topic Starter

  • Members
  • 86 posts
  • OFFLINE
  •  
  • Local time:04:06 PM

Posted 21 April 2006 - 07:05 AM

This machine has 5 hard drives (why??) but I've disconnected them all except for the one with the OS.

It's just crashed again, whilst running Spybot. Swapped the RAM na dit happened even quicker.

Next call I think is to remove the graphics card and see how it goes.

#7 RichieP

RichieP
  • Topic Starter

  • Members
  • 86 posts
  • OFFLINE
  •  
  • Local time:04:06 PM

Posted 21 April 2006 - 01:37 PM

Hmm. Took out the graphics card, then realised the on-board has a male termonail, same as my monitor. Can't try that yet.

Cleaned the card and took out the heat sink and fan. Cleaned a thick layer of dust from between the fan and sink and thought this might be it. Problem. The PSU won't stay on longer than 5 secs. This is the one from my own machine, and it still works OK in my own machine.

This is weird. I've ordered a new Antec 500W PSU, but any ideas why neither PSU will stay on longer than 5 secs? I can't have put the heatsink back in wrong becasue I know it will only fit one way, and it will work without one, before overheating.

#8 pascor22234

pascor22234

  • Members
  • 403 posts
  • OFFLINE
  •  
  • Local time:12:06 PM

Posted 21 April 2006 - 02:56 PM

No, there probably is no onboard video - thats a serial port with (9) male connector pins showing. Video ports have 15 pins in 3 rows..

Is the video card being overclocked ? If so, reduce or eliminate all overclocking.

Is the CPU or RAM being overclocked ? If so, reduce or eliminate all overclocking.

I just reread your last post ! Power supplies that turn themselves off in 5 seconds is almost always due to the heatsink not properly contacting the CPU die or heatspreader. You may have dislodged it when cleaning the dust out.

First, get some heatsink compound, preferably Arctic Siver 5. It's expensive, but is only a small fraction of the cost of a new power supply. Remove the heatsink & fan (HSF). Clean the old heatsink compound off both the underside of the heatsink and the CPU top, preferably with rubbing alcohol and a soft cloth or tissue. Place a SMALL dab of heatsink compound on top the CPU die or heat spreader. Replace the HSF. This may be easier (or not) by removing the fan first and then replacing it after the heatsink block is reattached to the motherboard. BE CAREFUL if you need a screwdriver to reattach the heatsink. A slip could cause the tool to dig into the motherboard and turn it into an expensive doorstop.

Reattach any cables you may have had to remove to get to the HSF. Turn on the power. It should not shutdown, but, if it does then take off the HSH and try to be more careful putting it back on.

When to computer stays powered up proceed to the next section to check out the machine.

-------------------------------------------------------
Try to find another video card to use long enough so can check everything else out:

To conclusively test the RAM download Memtest86. It comes in an ISO file format that can be burned to a CD using Nero, Roxio Easy CD Creator or the free CDBurnerXP Pro. This will make a bootable CD, so make sure the BIOS is set up to try to boot from the optical drive before it tries to boot from the hard drive.

Reading CPU and motherboard temperatures can usually be done with the free version of Everest Home. When it is running look in Computer > Sensor. The CPU temp should never exceed 55C. A really tough CPU test is Prime95. Monitor the Everest temperatures closely when running Prime95.

Also in the Sensor section take a look at the power supply outputs. At all times the readings should be withing +/-5% of the nominal voltages:
+3.3: 3.135V to 3.465v
+5: 4.75V to 5.25V
+12: 11.4V to 12.6V
You did say you swapped out the power supply so this really shouldn't be the problem.

Perhaps there are errors in the file system. From a command prompt type 'chkdsk c: /f /r'. It will ask you if you want to schedule a scan at the next restart - answer with a y and then restart the machine.

Your hard disk can be checked with Hitachi's Drive Fitness Test. It comes downloadable in an ISO image file, too. It will check the drive mechanics rather than the file system and won't cause any data loss.

Edited by pascor22234, 21 April 2006 - 03:18 PM.


#9 ColdinCbus

ColdinCbus

  • Members
  • 312 posts
  • OFFLINE
  •  
  • Local time:12:06 PM

Posted 21 April 2006 - 03:15 PM

Quick question, is it crashing with BSOD or hanging? If it is hanging, it may be related to the latest update from Microsoft. Some Nvida drivers are having an issue with the update.

edit - Nevermind, I can't read. You said it was randomly shutting down right there in the first post.

Edited by ColdinCbus, 21 April 2006 - 03:20 PM.


#10 RichieP

RichieP
  • Topic Starter

  • Members
  • 86 posts
  • OFFLINE
  •  
  • Local time:04:06 PM

Posted 22 April 2006 - 05:49 AM

Thanks for replies so far.

In reply to pascor:

Would the heatsink not making proper contact really make it turn off? I thought it was a bit odd that it wouldn't start after putting the heatsink back on. I need to get some compound when I get the chance this week.

It's quite a difficult one to get back in. It's got black plastic bits that twist quite stiff, if you get my explanation. I'll try taking it out and putting it in again.

As for diagnostics. You're right about the serial connector. No on-board video then.

I ran the Maxtor diagnostics. It reports no problems with the hard drive.

If I can get it going again, I'm really hoping that the thick layer of gunk I removed from the heatsink was the cause.

I'll keep you informed of progress. Thanks for replying.

#11 Herk

Herk

  • Members
  • 1,609 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:S.E. Idaho, USA
  • Local time:12:06 PM

Posted 22 April 2006 - 11:35 AM

Re-applying the heatsink compound may work - wouldn't that be great? When you apply it, use an old credit card or something and a very small amount. The compound works by filling microscopic pits and scratches on the surfaces and too much is as bad as too little. You don't want it squeezing out all over the motherboard. You just want enough to cover the surface in a very thin layer.

#12 RichieP

RichieP
  • Topic Starter

  • Members
  • 86 posts
  • OFFLINE
  •  
  • Local time:04:06 PM

Posted 22 April 2006 - 01:38 PM

I'll ask another question then.

Can it really make the PSU fail if the heatsink compound, or heatsink itself aren't connected/fitted correctly? I thought it still worked as normal, but the CPU got hotter much quicker. Surely it doesn't get that hot in 5 seconds it shuts itself down?

#13 Herk

Herk

  • Members
  • 1,609 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:S.E. Idaho, USA
  • Local time:12:06 PM

Posted 22 April 2006 - 02:46 PM

If the heatsink/processor aren't fitted correctly, the processor is most likely to fail. But the first thing a computer does when it boots up is a self-diagnostic. The power supply would shut down if something's wrong, maybe. (Sometimes, you wish it had.) You said it had five hard drives? I find that interesting, too. What kind of drives are we talking about? Are these SCSI drives? How were they hooked up? If SCSI, you need a termination on the end of the SCSI cable.

#14 RichieP

RichieP
  • Topic Starter

  • Members
  • 86 posts
  • OFFLINE
  •  
  • Local time:04:06 PM

Posted 22 April 2006 - 02:55 PM

Fair enough. I guess I'll find out when I get some compound and reseat the heatsink. Properly, hpefully.

As for your question, 4 are connected by SATA, including the one with the OS, and one by IDE.

#15 pascor22234

pascor22234

  • Members
  • 403 posts
  • OFFLINE
  •  
  • Local time:12:06 PM

Posted 22 April 2006 - 07:39 PM

Yes, a improperly fitting heatsink doesn't pull the heat away from the processor fast enough. The CPU will heat up extremely fast (2-5 seconds) which will cause the motherboard to go into an emergency shutdown mode which cuts off power to the CPU. The CPU chip can easily reach 100C or more before emergency shutdown goes into effect. This is generally very bad for the chip but probably won't toast the CPU if it happens once or twice (I have accidentally done this myself with no ill efect). Just don't let this happen over and over again or the CPU will fail at some point.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users