Frequent Crash / No errors

[Wylder]

**UPDATE**

After trying for months to figure this problem out and nearly throwing the machine out the window I've found the issue for certain. It was the processor. I received the replacement a week ago and haven't had a problem since. Everything runs fast, cool, and without error. Thanks to all who helped out.

**UPDATE**

I’ve been having a problem where my computer will “freeze” and just sit there. No errors, no BSOD, just frozen until I do a hard reset. Upon reboot there are no indications of any errors, there aren’t any errors in any of the system logs, and there aren’t any logs in the minidump folder for windows. I’ve left the machine on for more than an hour and it just sits there frozen.

System specs:
Windows XP SP2
Asus P5N-D
Nvidia 9800GT
2x 2gb Corsair XMS DDR2 Memory
Corsair HX620 Power Supply

I believe it to be one of the components causing the issue and would love some help on how to isolate the troublesome component. Here’s where I’m at.

I thought it was the video card so I swapped it out with an ATI 1950X. I got another lockup. I removed one of the sticks of RAM and placed it into another machine and that machine runs great. No errors or lock ups. Just to verify that the 9800gt was good I placed it in the same machine as the RAM and it also runs like a champ.

So now I’m stuck. I’ve been getting lockups in the machine for several months and they were getting progressively worse. Something isn't working right. Any suggestions?

 

[Appzalien]

Put your memory back in and run Memtest86 from a boot floppy or cd if you don't have a floppy. Let it run for several hours on both sticks unless its obvious theres something wrong. If you get errors you might be able to RMA the ram, Corsair is really good about it, just copy down the error codes and go to there web site and make an RMA request. Sometimes bad ram gets out into the world and they know its coming back to them, so there really good about it. I once had two bad 512mb sticks or DDR they didn't even make anymore, so they sent me two 1gig sticks to replace them. Cool huh?

Oh, I forgot, you might want to go into the bios and set safe defaults and see what happens.

 

[Wylder]

Ok, thanks for the tip. I'll run Memtest tonight and see what the results are. I did notice something peculiar when I was checking out my memory timings. They are supposed to be at 5-5-5-18, but CPU-Z was reporting them at 5-5-5-33. Could that have caused any issues?

Everything is stock with no OC'ing.

 

[snowchick7669]

If you manage to get into your machine, then it pays to find the minidumps that your computer stores when it crashes. These can be read and debugged, which generally points to a faulting service/driver.

These can be found in

C:\Windows\Minidumps

That directory is for XP however. You can then upload these minidump files to be evaluated.

Let me know if you have trouble finding these

 

[Wylder]

This is what my minidump contains. It looks like nothing has been listed since 4-06-09. So my recent lock -up are not here. : /

Debugging Details:
------------------


FAULTING_THREAD: 8ad45ba0

DEFAULT_BUCKET_ID: GRAPHICS_DRIVER_FAULT

CUSTOMER_CRASH_COUNT: 1

BUGCHECK_STR: 0xEA

PROCESS_NAME: Crysis.exe

LAST_CONTROL_TRANSFER: from 00000000 to bd0ff6b0

STACK_TEXT:
00000000 00000000 00000000 00000000 00000000 nv4_disp+0xed6b0


STACK_COMMAND: .thread 0xffffffff8ad45ba0 ; kb

FOLLOWUP_IP:
nv4_disp+ed6b0
bd0ff6b0 49 dec ecx

SYMBOL_STACK_INDEX: 0

SYMBOL_NAME: nv4_disp+ed6b0

FOLLOWUP_NAME: MachineOwner

MODULE_NAME: nv4_disp

IMAGE_NAME: nv4_disp.dll

DEBUG_FLR_IMAGE_TIMESTAMP: 49dabcf7

FAILURE_BUCKET_ID: 0xEA_IMAGE_nv4_disp.dll_DATE_2009_04_06

BUCKET_ID: 0xEA_IMAGE_nv4_disp.dll_DATE_2009_04_06

Followup: MachineOwner

 

[snowchick7669]

Hi Wylder,

The minidumps don't always record every crash, so dont panic. Have you updated your graphics driver recently at all? It may pay just to download the latest update, have had issues with out of date drivers before.

Going in a completely different direction, why not running speedfan (can be downloaded for free via google). This program sits and monitors the heat of your machine, just in case it is actually overheating. I had a recent problem where my computer would freeze when I was using alot of graphics and I ended up working out that the GPU was overheating and so I took the sides and front off my machine and it solved the problem.

It pays to leave speedfan going for a few hours, and trying various applications so you can put some good stress on the components and see if they heat up above the normal range.

Other then that, the previous memtest suggestion would be mine as well, run that scan and see how you go. A general technician rule of 20 passes for good RAM, but I have found that sometimes I will find 1 error after 72 hours of running the scan. Overnight should be sufficient for most cases

 

[Route44]

To add to what snowchick7669 has said your error is as follows per auhma.org; note the description and cause.

0x000000EA: THREAD_STUCK_IN_DEVICE_DRIVER
A device driver problem has caused the system to pause indefinitely (hang). Typically, this is caused by a display driver waiting for the video hardware to enter an idle state. This might indicate a hardware problem with the video adapter, or a faulty video driver.

Now if we look at your minidump that you posted we get the following:

1. GRAPHICS_DRIVER_FAULT

2. nv4_disp.dll which is an Nvidia display driver.

As snowchick7669 rightly directs updating your video card drivers is a wise course of action. Do the following...

1. Download Driver Cleaner Pro or Driver Sweeper (free versions) to your desktop screen and install.

2. Download the latest diver(s) for your video card but don't install them.

3. Uninstall your video card drivers and reboot your PC into Safe Mode. Run Driver Cleaner Pro. If it doesn't find any video card drivers that is quite okay.

4. Reboot and install new video card drivers.

 

[Wylder]

Update -

- Ran a Memtest overnight and had zero errors.
- Lockups without errors continued
- Uninstalled video drivers
- Ran Driver Cleaner
- Installed 185 Nvidia drivers
- Lockups without errors continued
- Uninstalled and reinstalled the latest chipset drivers
- Same issue
- Updated BIOS and reloaded all default settings
- Same issue
- Formatted HD and installed Win XP SP 2
- Same issue
- Formatted HD again and installed Win 7
- Same issue
- Took all components out of the PC except for video card
- Same issue
- Installed another Video Card
- Same issue
- Replaced RAM
- Same issue

I am going nuts here. What more can I do?!?

 

[snowchick7669]

Now this may be a long shot, but have you tried a different PSU?

Also, when you replaced the RAM, did you replace both the sticks in both slots? Or just use one fresh stick and try it in both RAM slots (just to eliminate a faulty slot itself)

 

[Wylder]

I have tried the old PSU from my previous system and the results are the same.

I tested both sticks of RAM in the following configurations. A1/A2, A1/B1, A1/B2, A2/B1, A2/B2, B1/B2 and I've also tried single sticks in each slot by themselves. After that I replaced the RAM with RAM from another machine. The lockups have not only continued, but are getting worse. I used to be able to stay on the machine for a few hours, but now I'm only able to make it a few minutes before it locks up. Again, with no errors.

I just RMA'd the motherboard. At this point I've exhausted all options thoroughly. If the the replacement motherboard isn't the issue then I'm completely lost as to the cause of this problem. I may go officially insane of it locks up again after I install the new board.

 

[snowchick7669]

Let me know how it goes!

Am rather eager to hear if anyone else gets satisfaction out of taking an axe to a machine

 

[Wylder]

Motherboard is being shipped back. They found no errors with the motherboard. So far I've replaced everything in the computer and all the components on their own work just fine. All together I get persistent system hangs.

Thanks all for your help.

Any other clues before I just dismantle the thing?

 

[Wylder]

Updated the original post.