BSOD - possible graphics, PSU, or motherboard failure?

laceyGB

Honorable
Apr 15, 2013
9
0
10,510
About two weeks ago, my XP machine got cursed by blue screens every time I ran it for more than 20 minutes. Mostly 0XA IRQL_NOT_LESS_OR_EQUAL and 0X1A MEMORY_MANAGEMENT errors. According to BlueScreenView, ntkrnlpa.exe was the main culprit. I tried a whole bunch of things, but they didn't help:

Ran memtest86 for 9 hours, it reported no problems with RAM
Updated graphics card drivers
Removed all graphics card drivers
System restored
Repaired Windows with the installation CD
tried RAM in different slots

Finally I checked my hard drive with Seatools, and it reported 205 bad sectors over 3 long tests. I figured the hard drive was responsible, so I bought a new one and got myself Windows 7 for the fresh install.

The blue screens persist. According to BlueScreenView, 85% of the minidumps are 0X1A MEMORY_MANAGEMENT errors, with the main culprit being ntoskrnl.exe. I ran Memtest86+ again, this time for 24 hours, but again it reported no errors. A bit disappointing, since my RAM has a lifetime warranty.

My system specs are:

Intel Q6600
Asus P5n32e-SLI
Seasonic 900w
Nvidia 8800 GTX
Corsair CM2X2048 - 6400CD
Western Digital 1tb blue

Any idea what the issue could be? Could it be a failing graphics card, motherboard, or PSU?

Thanks :)




 

johnnyq1233

Distinguished
Aug 15, 2007
1,233
0
19,460
I would look for an updated chipset driver since the memory controller and I/O functions are on the motherboard.
Other then that, I would consider borrowing a graphics card from a friend just to test if that's the culprit.
 

laceyGB

Honorable
Apr 15, 2013
9
0
10,510
Updated nForce chipset drivers didn't solve the problem, and I currently don't have any access to a spare graphics card. Is there a way of narrowing down the problem without replacing components?
 

laceyGB

Honorable
Apr 15, 2013
9
0
10,510


Hmm...

MEMORY MANAGEMENT

0x41284 - 14 instances in the past week
A PTE or the working set list is corrupted.

0x888A - 4 instances
Internal memory management structures (likely the PTE or PFN) are corrupted.

0x8884 - 1 instance
(Windows 7only). Two pages on the standby list that were supposed to have identical page priority values do not, in fact, have identical page priority values. The differing values are captured in parameter 4.

0x411 - 2 instances
A page table entry (PTE) has been corrupted. Parameter 2 is the address of the PTE.

0x403 - 2 instances
The page table and PFNs are out of sync . This is probably a hardware error, especially if parameters 3 & 4 differ by only a single bit.

PFN_LIST_CORRUPT

0x02 - 1 instance
A list entry was corrupted.

0x99 - 6 instances
A page table entry (PTE) or PFN is corrupted.


SYTSTEM_SERVICE_EXCEPTION - 4 instances

This error has been linked to excessive paged pool usage and may occur due to user-mode graphics drivers crossing over and passing bad data to the kernel code.

PAGE_FAULT_IN_NONPAGED_AREA - 1 instance
Bug check 0x50 usually occurs after the installation of faulty hardware or in the event of failure of installed hardware (usually related to defective RAM, be it main memory, L2 RAM cache, or video RAM). Another common cause is the installation of a faulty system service. Antivirus software can also trigger this error, as can a corrupted NTFS volume.


Put in a new graphics card yesterday, but that didn't solve the problem. So it's not the graphics card, it isn't the memory, it isn't the hard drive, and it isn't the software or drivers. That leaves the CPU, motherboard, and power supply.
 

TenPc

Honorable
Jul 11, 2012
2,471
1
11,960
Is it Windows 7 x32 or x64?
Did you upgrade or new clean install?
Did you delete all and any partitions prior to formatiing and install?

as per link -
http://www.asus.com/Motherboards/P5N32E_SLI_Plus/
The 12v connector is actually an 8-pin connection, there is a plastic lid on the other 4 pins so have you connected the 8 pin 12v connector or are just using the 4 pin connection?

What is your PSU make model and the new graphcs card?
 

laceyGB

Honorable
Apr 15, 2013
9
0
10,510


Mobo's revision is 1801, CPU revision is G0.





Windows 7 x64, completely new install on a new hard drive.

I've connected it via the 8 pin 12v vonnection.

PSU is a Seasonic X900, and the new graphics card is a GeForce 210.
 

TenPc

Honorable
Jul 11, 2012
2,471
1
11,960
I think it's the PSU, it's too high powered for all your hardware. I often overstate the PSU requirement but 900 Watts is way over the top especially for a motherboard that is more suited to XP. The 12 v rail would be about 850 watts plus, considering it's a Seasonic, cheaper brands would only have about 780 watts.

The maximum 12v watts for all your current hardware would not exceed, even at peak, more than 350 watts, even the total power would not exceed more than 500 watts. You are using a high powered PSU that is 400 watts more than is nnecessary, it can impact on the motherboard so check for bulging capacitors. I also think your Power Mangement system cannot manage the PSU properly and is having a bad day.

Try to acquire a PSU with less total power.

Check in Device Manager for any inconsistencies.
 

TenPc

Honorable
Jul 11, 2012
2,471
1
11,960
Also, in the memory QVL list, there are no modules for 800mhz supporting 2gb, only one in the 667mhz range -
2048MB NANYA NT5TU128M8BJ-3C DS NT2GT64U8HB0JY-3C

According to NANYA, their DDR2 is 1.8v
http://www.nanya.com/PageEdition3.aspx?Menu_ID=61&def=210&lan=en-us
Your ram is probably only 1.5v.

Ram can cause those BSoDs if they are not compatible or have the wrong timings or latency's or speeds, or voltage.

You may neeed to acquire DDR2 800mhz 1024mb (1gb) sticks x4 or 667mhz 2048mb sticks x 2 NANYA brand, @ 1.8v

 

laceyGB

Honorable
Apr 15, 2013
9
0
10,510


Hmm...my system originally had two Nvidia 8800 GTXs in SLI, and my problems started long before I changed the graphics cards. 900w doesn't seem too high for two power hungry cards, and it ran fine for 5 years before all the BSODs. The capacitors on the motherboard look fine.
 

laceyGB

Honorable
Apr 15, 2013
9
0
10,510


Yep, I used the RAM for 5 years without any problems.

 

TenPc

Honorable
Jul 11, 2012
2,471
1
11,960
So, what were you doing when the first issue occured? And subsequent issues?

Maybe it's not the ram or PSU or graphics card or even the hdd, or cpu, what about DUST. So much dust in this world that nobody seems to notice.

Edit - Asus did not mention voltages for the ram, in the spec sheet, so it's possible it accepts most types.
 

laceyGB

Honorable
Apr 15, 2013
9
0
10,510


When it first happened I was running a whole bunch of programs - a BitCoin client that was processing an old wallet, a torrent client downloading a bunch of different of stuff, and Firefox with about 30 tabs open. Since then, it's been somewhat random. During normal browsing it can happen every half hour to every few hours, but if I try to play game or have, say, 20 youtube tabs loading at once, it happens really quickly.

Dust certainly could be a problem, but what about a failing CPU? I know that the CPU is usually the last thing to go, but can high heat contribute to a premature demise? Before I knew any better, the cpu in this PC (Q6600, default clock) was running at about 65C idle for about four years, before I learned about the importance of cooling and changed the CPU fan.
 

scout_03

Titan
Ambassador
your bios os ok and on the link of your motherboard http://www.asus.com/Motherboards/P5N32E_SLI/#support_Download_8 they have support up to vista also the corsair memory stick you use work at 2.1 volts according to corsair and list of the kit compatible with you board work at 1.8 volts see the list for specs and timing http://www.corsair.com/us/learn_n_explore/ it could be a matter of settings recheck all of them in the bios .
 

TenPc

Honorable
Jul 11, 2012
2,471
1
11,960
In another thread, the OP resolved his own issue -
"Litecoin mining - But I found the solution, It was a simple fact of the program I was using was just buggy and unsupportive of my current set up.
Changing the software from GUIminer to CGminer and now everything has been stable for a day and a bit."

http://www.tomshardware.co.uk/forum/id-1662517/system-hangs-10mins-mining.html#10730431

If you can get to Safe Mode, you might want to try that option.

"20 youtube tabs loading at once" - that's rather a lot, and seems rather innocuous.
"Firefox with about 30 tabs once" - that seems rather a lot and also innocuous.
Torrents etc seems to infer something but am not to sure.

There is no current 64 bit Firefox, only BETA versions used for testing purposes only. Mozilla had announce last year or so that they wouldn't be considering opting for the 64 bit version but it seems that users have forced their hand.

For all the things you are doing, you should consider a faster CPU/motherboard with 16gb/32gb of ram DDR3.
You can't play a game while you are doing other things, or have just done a heap of other things, better to restart the PC then go play a game.
You only got 4gb of ram and although the hardware is over the alps, you need a fast cpu with heaps better ram.

edit - "ntkrnlpa.exe " is a backward compatibility system file for Windows 2000 programs and services. It is required by all Windows OSes past 2000 for compatibility reasons as there are still some hardware and programs that utulize the Windows 2000 platform.
 

johnnyq1233

Distinguished
Aug 15, 2007
1,233
0
19,460
I'ld agree. That's a lot of information processing that you are throwing at an old 4 core processor with only 4GB of ram.
Do you ever get the low memory warning or monitor you memory usage?
Does the machine crash if you boot it up and just let it sit idle?
How does it work if you open say only 2-4 pages at a time?