EventID: 1003 Error: 10000050

[MoeSzyslak]

I just built a new computer and recently have been getting BSODs at very random times. I have run a lot of stress tests on the computer (testing oc stability) and they have never failed. But at the most random times, like surfing the internet, the computer will get a BSOD.

I followed all the instructions above. My memory passed 14 passes of memtest86, I ran chkdsk, and reset my page file after defragmenting.

I have some of my minidumps attached to this thread.

System specs:
Intel Q6600 G0 Stepping @ 3.2Ghz (1.25V) - TRUE 120 Extreme lapped
Gigabyte DS3L
G.Skill 4GB PC-8000
Western Digital 320GB
eVGA 8800GT

 

[Route44]

4 errors were 0x8E: KERNEL_MODE_EXCEPTION_NOT_HANDLED

A kernel mode program generated an exception which the error handler didn’t catch. These are nearly always hardware compatibility issues which sometimes means a driver issue.

* The drivers that were cirted were core Windows drivers.

Your other dump was an 0x50: PAGE_FAULT_IN_NONPAGED_AREA

Requested data was not in memory. An invalid system memory address was referenced. Defective memory (including main memory, L2 RAM cache, video RAM) or incompatible software (including remote control and antivirus software) might cause this Stop message, as may other hardware problems (e.g., incorrect SCSI termination or a flawed PCI card).

1. What are your RAM voltage settings in your BIOS? Have you overclocked and if so have you tried easing back on the voltage? Sometimes the RAM may need more; I've seen this happen too. What does G.Skill recommend?

2. Bad RAM has been known to pass MemTest; try running one stick at a time.

3. Also, try running with just 2 sticks and see if stability returns.

* By the way, Moe, nice system.

 

[MoeSzyslak]

4 errors were 0x8E: KERNEL_MODE_EXCEPTION_NOT_HANDLED

A kernel mode program generated an exception which the error handler didn’t catch. These are nearly always hardware compatibility issues which sometimes means a driver issue.

* The drivers that were cirted were core Windows drivers.

Your other dump was an 0x50: PAGE_FAULT_IN_NONPAGED_AREA

Requested data was not in memory. An invalid system memory address was referenced. Defective memory (including main memory, L2 RAM cache, video RAM) or incompatible software (including remote control and antivirus software) might cause this Stop message, as may other hardware problems (e.g., incorrect SCSI termination or a flawed PCI card).

1. What are your RAM voltage settings in your BIOS? Have you overclocked and if so have you tried easing back on the voltage? Sometimes the RAM may need more; I've seen this happen too. What does G.Skill recommend?

2. Bad RAM has been known to pass MemTest; try running one stick at a time.

3. Also, try running with just 2 sticks and see if stability returns.

* By the way, Moe, nice system.

Wow, thanks so much for the help. This place is awesome!

The one 0x50 error happened yesterday. That one actually happened right after I opened up my computer and connected a sd card reader to my mobo. I turned on the computer and windows automatically installed the drivers. Then, maybe a minute later, I got a BSOD.

1. The Gigabyte DS3L has a default mem. voltage of 1.8 while the G.Skill dimms are rated at 2.0-2.1v. I have the ddr2 overvoltage for my mobo set to +.3 so it should be at 2.1v but on CPUID @ idle it is at 2.03 and I think it drops just like all my other voltages during stress tests. I had it set to +.2 when I initially started getting the BSODs and increasing it by the .1 hasn't helped.

I actually got a BSOD in the middle of typing this message....I had to write that paragraph above all over again. This is the first time I could read what the BSOD said because I just figured out how to stop windows from restarting from an error last night. I have a picture of the BSOD attached to this thread.

It is an 0x8E error and it lists the file "watchdog.sys" underneath. Just searched it on google and not much came up. Could it be a virus? I have had NOD32 installed on this computer since the first day I turned it on so I would be surprised if I got a virus.

2. I haven't tested one dimm at a time but I guess I should give it a shot. What tests do you recommend I do. Should I just stick with 7 passes of the entire test set? Some people recommend doing mutiple passes of test #10.

3. I only have two sticks. The G.Skill 4GB i have is a set of two 2gig dimms. Im guessing you mean that I should run windows with just one at a time and see if its stable.

 

[Route44]

Actually Watchdog.sys is part of Windows. Since all of your BSODs cite Window drivers it is a good bet you've got a hardware issue soemewhere.

Did you uninstall that SD card and its drivers?

I am still thinking it is your RAM voltage. MemTest on each stick for 7 passes and yes then try running one stick to see if it brings stablity.

 

[MoeSzyslak]

Haven't uninstalled the card reader yet but I was having this issue for a few days before I installed it. The most recent error I got was the same as the ones I was getting the day before I installed it.

Right now I have just one stick in and I'm running MemTest.

Do you think my voltage is too high or too low? The memory is rated at 1000Mhz and I have it running at 800Mhz so maybe even +.2 is too much?

UPDATE: So I've tested one of the dimms. It passed memtest 7 times so I restarted and loaded up windows. I left it on the login screen and went to watch the game. I just got back and it was still at the login screen so I figured that it was running fine but when I logged in, I got the windows message saying I just recovered from a serious error. I checked my event log and it shows an 0x8e error at around 8:30pm. Wierd thing is that I turned off the automatic restart after an error so when I turned on my monitor, I should of have the BSOD in front of me but the computer restarted for some reason.

Then I decided to go into the BIOS and changed the ddr overvoltage to +.1 and then loaded up windows again. Right when my desktop loaded up I got a NVidia message saying that there wasn't enough power going to my video card and that the software was going to drop the clocks to a stable level.....but my video card wasn't overclocked in the first place and when i checked after i got that message, they were still at the default levels.

Could this somehow be a video card issue? Lol, I feel like my computer is slowly dying on me.

 

[MoeSzyslak]

Just ran two instances of HCI Memtest with the new overvoltage setting (+.1v) and it failed within a few minutes so the memory definately wasn't being overvolted. Changed it back and now I'm running the test again. Then I gotta do all this testing again with the other dimm

 

[Route44]

Thanks for the updates. Would love to hear how you solve your issue. What is the make, model, wattage of your PSU and what is you amperage on your 12v. rails?

 

[MoeSzyslak]

Thanks for the updates. Would love to hear how you solve your issue. What is the make, model, wattage of your PSU and what is you amperage on your 12v. rails?

Heh, I'm kinda hoping I hear it from you. I'm really stumped at this point. HCI Memtest runs completely stable now that my overvoltage is back to +.3v but I'm just gonna let the computer run for a while before I try the other dimm out. The computer never BSODed during stress tests but at random times with maybe one program open so I'm gonna give it a few hours to run with a movie playing plus firefox open.

My PSU is a Corsair HX520 and im getting 12.22v on the 12volt rail.

 

[Route44]

So if I am reading you correctly, with your voltage at +.3v with one dimm in slot it is stable, but with two dimms it has random BSODs, correct?

Sometimes motherboard's can be quirky in the memory configuration. Does Gigabyte say anything about 4 gigs in a 2x2 configuration as opposed to a 4x1 with your mobo? Have you tried different slots?

I also looked up your PSU and that is more than sufficient for what you are running. Are you able to use a multimeter on your molex connecters while system is under a load?

 

[MoeSzyslak]

So if I am reading you correctly, with your voltage at +.3v with one dimm in slot it is stable, but with two dimms it has random BSODs, correct?

Sometimes motherboard's can be quirky in the memory configuration. Does Gigabyte say anything about 4 gigs in a 2x2 configuration as opposed to a 4x1 with your mobo? Have you tried different slots?

I also looked up your PSU and that is more than sufficient for what you are running. Are you able to use a multimeter on your molex connecters while system is under a load?

Well, right now I'm running my computer in windows to make sure that it is actually stable. The computer never had problems getting through stress tests but had issues during basic tasks so I'm just letting windows xp run for the night to make sure that this first stick is stable in windows for an extended period. Then I'm gonna take this stick out, put in the other one and then run memtest for 7 passes and then windows xp for a few hours.

Maybe the other slot is faulty? After I test out this other stick, putting it into the same slot as the first stick to keep that variable constant, i'll try out the other slot that was being used originally when i had both sticks installed.

I don't see why this motherboard would have a problem with a 2x2 config. It is a pretty popular board and I've seen many people with similar configs using 2 2gig sticks.

Good news is that the computer has been running windows for about 6hrs now (with two movies playing, firefox & charting software open) and still hasn't gotten a BSOD. Hopefully when I wake up tomorrow itll still be running and then I can rule out the first stick as the problem.

 

[MoeSzyslak]

Just an update. I woke up this morning and windows was still running fine so I took out the first stick and put in the second one. Ran memtest for 7 passes with no errors so I loaded up windows. I had it running just some basic programs for a while and then decided to run HCI Memtest to test the memory stability in windows. I did get one error 30% through but nothing after that and now I'm running Prime95 blend and everything seems fine. More importantly, I've been running windows for about 3 hours on this second stick and no BSOD. After Prime95 is done I'm gonna run HCI Memtest one more time with my other programs closed to see if i get an error again.

Next thing to test out is if the other slot was faulty.

BTW, I did some research on the Nvidia error message I was getting and a lot of people have this issue. Something to do with certain brands of power supplies although I read instances where the person switched out to another power supply and still got the message so i have a feeling it is an issue with the video card. Apparently, eVGA released a set of drivers that addressed this issue so I removed my old drivers and installed those. I was only getting the insufficient power message every 2 or 3 restarts and i've only started up windows once since the new drivers were installed so i can't be sure yet that its fixed.