Wierd 12v?

[ViktorHark]

I got a crash with the following error "PAGE_FAULT_IN_NONPAGED_AREA." I then ran a memtest86, which passed, since I read somewhere that it was related to RAM errors. Afterwards I contacted BFG's tech support (since I figured it was related to video card RAM, and I had just got the card RMA'd), who prompted me to check out my 12v line on my PSU, which was at 12.37 in the Bios. I'm not the most literate as it comes to power, so does it sound likely that a 12v with that kind of reading would cause crashes in numerous programs, albeit graphically intense games most of the time (nearly constant lock-ups/crashes in Crysis, TF2, Sins, HL2:EP2, and occasionally in WoW in more intense spots)

Here are my system specs
AMD x2 6000+
BFG 8800 GTX
Asus nForce 570 SLi motherboard (Onboard Audio)
750W PC P&C PSU Silencer Quad
4 GB DDR2 RAM
Vista Home Premium

 

[BigBadBiologist]

BIOS readings aren't the most accurate. However, 12.37 should be fine.

How many times did you let memtest run? Are you running the latest chipset and video drivers? Are you overclocked at all?

 

[NoxTek]

You cannot cannot CANNOT TRUST the BIOS readings, they will fluctuate lots and in most cases be completely off from the real value.

Want to measure the +12v on your PSU? Turn on a digital multimeter and adjust it to read in the 20v DC range. Now connect the black lead from the DMM to the black pin on a molex connector from your PSU and connect the red lead to the yellow pin on a molex connector from your PSU. (The black lead can also be connected to any metal part of the case chassis, ie a common ground). Power up your PC and take a reading, presto! And I bet it's totally different from the +12v reading in the BIOS.


Since your PSU is a single-rail unit that's all you need to do. If you had a PSU with multiple +12v rails you'd need to measure each rail. (On a good quality unit the different rails are signified by a different colored 'stripe' on the yellow wires)


Whatever the case I highly doubt the PC Power & Cooling Silencer 750w unit is the source of your problems. It's one of the most rock solid Seasonic designs around... and if by chance it IS defective PC Power & Cooling / OCZ will take great care of you.

 

[ViktorHark]

How many times did you let memtest run? Are you running the latest chipset and video drivers? Are you overclocked at all?

I let the memtest run 3 times, all were pass. I have the latest chipset and video card drivers. I have nothing overclocked. For comparison sake, the RAM is Corsair XMS

Want to measure the +12v on your PSU? Turn on a digital multimeter and adjust it to read in the 20v DC range.

Whatever the case I highly doubt the PC Power & Cooling Silencer 750w unit is the source of your problems. It's one of the most rock solid Seasonic designs around... and if by chance it IS defective PC Power & Cooling / OCZ will take great care of you.

I don't have a digital multimeter, so how much would one cost me? (and where would I be able to pick one up for that matter? Radio Shack?) And regarding a defective PSU, I've had it for nearly a year without issue so it seems wierd that it'd just start breaking.

That being said, is there any other possible causes or things I could investigate to fix this? While I do occasionally get the page error BSoD, most of the time it just locks up or crashes.

Edit: In case I do get a multimeter, what is an acceptable 12v range for a stable system? I only ask because the person at BFG's tech support seemed pretty certain that anything above 12.25 would be bad.

 

[NoxTek]

I don't have a digital multimeter, so how much would one cost me? (and where would I be able to pick one up for that matter? Radio Shack?) And regarding a defective PSU, I've had it for nearly a year without issue so it seems wierd that it'd just start breaking.

You can grab one at Radio Shack for around $20 last time I checked.... no self respecting geek should be without one!

That being said, is there any other possible causes or things I could investigate to fix this? While I do occasionally get the page error BSoD, most of the time it just locks up or crashes.

The cause could be anything really, bad memory, overheating CPU, flaky motherboard, driver conflicts, BIOS issues. It's hard to say without more information really...

Edit: In case I do get a multimeter, what is an acceptable 12v range for a stable system? I only ask because the person at BFG's tech support seemed pretty certain that anything above 12.25 would be bad.

The ATX 2.2 specs say a deviation of plus or minus 5% is fine on the +12v line. So you've got a range of 11.40v - 12.60v where everything is within acceptable specs. Ideally you don't want to be at the extreme end of either side of that range, but if your +12v line really IS at 12.37v there isn't really anything wrong with that.

As I said, measure with a DMM at idle (BIOS or desktop) and at load (running 3DMark and Prime95 for example) and compare your DMM readings to readings seen in your BIOS or in HWMonitor, you'll probably be suprised at the difference between the two measuring methods.

 

[JonGerow]

Despite what MEMTEST says, I'm still going to say the problem is probably RAM.

You have 4GB... Is that 2x 2GB sticks or 4x 1GB sticks? If it's easy to replicate the error (running a certain benchmark or game), certainly you can take some memory out to see if the Page Fault error goes away.

 

[ViktorHark]

Despite what MEMTEST says, I'm still going to say the problem is probably RAM.

You have 4GB... Is that 2x 2GB sticks or 4x 1GB sticks? If it's easy to replicate the error (running a certain benchmark or game), certainly you can take some memory out to see if the Page Fault error goes away.

My RAM is 4x1GB. I'll look into that.

I got a digital MM which rated my 12v at 12.04 so yeah, don't think thats an issue.

Edit: the DMM was at idle, I'll do another test at load.

 

[NoxTek]

My RAM is 4x1GB. I'll look into that.

I got a digital MM which rated my 12v at 12.04 so yeah, don't think thats an issue.

Edit: the DMM was at idle, I'll do another test at load.

I bet you it will hardly fluctuate, probably not more than .1v. The Seasonic designed PSUs are awesome like that...

 

[ViktorHark]

I bet you it will hardly fluctuate, probably not more than .1v. The Seasonic designed PSUs are awesome like that...

Yeah, you're right, it pretty much stayed the same. I called BFG tech support again, and the person I talked to agreed that even at 12.37v I would have been fine. I'm also more inclined to agree with the video card problem, since the new card is about a week or two old (since I just got it back from RMA) and prior to this I never had this issue (the reason why I had mine RMA'd was for a totally different problem, which was fixed and then replaced with this new problem) (the RAM and the rest of my system is over a year old now and I've never seen this before)

She said it was probably a corrupted VRAM or the card was drawing too much power. Either way, thanks for all the help guys!

 

[ViktorHark]

I'd hate to make a new topic, but my problem seems surmised here already.

Either way I was fooling around, and noticed that my audio drivers were locked to 7.1 channel sound (where I only have 5.1 channel sound) so I reinstalled my audio drivers and reconfigured my audio, and lo and behold I was able to play a solid day of TF2 without any lag, lock ups, crashes, or anything of the sort.

Could this have been the source of my problem?

 

[NoxTek]

I'd hate to make a new topic, but my problem seems surmised here already.

Either way I was fooling around, and noticed that my audio drivers were locked to 7.1 channel sound (where I only have 5.1 channel sound) so I reinstalled my audio drivers and reconfigured my audio, and lo and behold I was able to play a solid day of TF2 without any lag, lock ups, crashes, or anything of the sort.

Could this have been the source of my problem?

Who knows *shrug*... that board has Realtek onboard audio right? Did you upgrade to the latest driver when you reinstalled them? Realtek are CONSTANTLY updating the drivers with fixes and improvements... (hell, I wish Creative spent half the time Realtek does on driver development). Maybe your system is all better?

 

[JonGerow]

My RAM is 4x1GB. I'll look into that.

I think it's your easy recourse here.

Take two sticks out.... run the machine.... swap out for the other two sticks.

You're making this a little harder than it has to be when you have four sticks of RAM at your disposal. You're also putting too much faith in MEMTEST. Yeah, when MEMTEST says you have bad RAM, you have BAD RAM, but I've also had MEMTEST pass KNOWN bad RAM as well. MEMTEST is a good tool, but far from perfect.

 

[ViktorHark]

Who knows *shrug*... that board has Realtek onboard audio right? Did you upgrade to the latest driver when you reinstalled them? Realtek are CONSTANTLY updating the drivers with fixes and improvements... (hell, I wish Creative spent half the time Realtek does on driver development). Maybe your system is all better?

Its actually a Soundmax audio chip, which the current drivers are the one I installed, but are still like nearly a year old.

And @RAM, I'm going to look into that if it crashes again, but like I said, so far, its been pretty good since I've updated the audio drivers. I think its also worth noting that the largest flag I had for the audio was how my system's audiodg.exe (vista's audio program) was glitching and taking up 60-70% CPU resources, now its down it 1-2.

 

[polemos4u]

I think it's your easy recourse here.

Take two sticks out.... run the machine.... swap out for the other two sticks.

You're making this a little harder than it has to be when you have four sticks of RAM at your disposal. You're also putting too much faith in MEMTEST. Yeah, when MEMTEST says you have bad RAM, you have BAD RAM, but I've also had MEMTEST pass KNOWN bad RAM as well. MEMTEST is a good tool, but far from perfect.

An answer without guile, memtest has done that too me.