• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

RAM or Hard Drive issue?

Pringle

2[H]4U
Joined
Sep 30, 2012
Messages
3,921
Hey guys

Recently put together a new build, formatted and did a clean install of Windows 10, and everything was working smoothly last night.

This morning, Windows is throwing up some BSOD's at me and refuses to repair itself or start in safe mode. The errors I've been getting are;

Page Fault Non Paged Area
PFN List Corrupt
Memory Management
Kernel Security Check Failure

I'm currently doing another wipe and clean install (this time with Windows 7) but what can I do to rule things out here?
 
start basic hardware testing

ram:memtest86
CPU:prime95
HDD: crystal disk info and the manufacors tool

Then basic software
sfc /scannow
chkdsk
reinstall drivers
reinstall windows
 
It screams of memory problems.
Too little info to work with tho.
Need exact memory, exact mobo and speed they are running at.
 
Hey guys

Recently put together a new build, formatted and did a clean install of Windows 10, and everything was working smoothly last night.

This morning, Windows is throwing up some BSOD's at me and refuses to repair itself or start in safe mode. The errors I've been getting are;

Page Fault Non Paged Area
PFN List Corrupt
Memory Management
Kernel Security Check Failure

I'm currently doing another wipe and clean install (this time with Windows 7) but what can I do to rule things out here?

Run stock settings bios settings, swap ram slots, swap dimms, etc.

Like Nenu said, sounds like a memory problem. I wouldn't even bother diagnosing it much. Just get it exchanged.
 
It screams of memory problems.
Too little info to work with tho.
Need exact memory, exact mobo and speed they are running at.

MSi Z270 Gaming M5
2 sticks, 8GB per of Corsair Vengeance LPX

Motherboard says the ram is running at 2133MHZ at 1.200v.

Ran Memtest, no errors on first pass. Going to run a couple more just in case.

I'm starting to think I just fucked something up with the OS install. Going to run Prime and Crystal Disk in a bit
 
MSi Z270 Gaming M5
2 sticks, 8GB per of Corsair Vengeance LPX

Motherboard says the ram is running at 2133MHZ at 1.200v.

Ran Memtest, no errors on first pass. Going to run a couple more just in case.

I'm starting to think I just fucked something up with the OS install. Going to run Prime and Crystal Disk in a bit

My take on it is that it installed fine and a hardware issue popped its head up as it started to bed in.
Could be wrong but thats what it smells of.
It will be interesting to see if further memory testing throws an issue because I expected it to fail on the first pass with the problems you are having, if the ram is the problem.

Check your motherboard is running the latest BIOS. This matters a lot as the updates often cover ram stability changes.
There may be other stability issues that have been sorted.
Hopefully this is all you need.
Failing that refit the ram.
Refit all the power and SATA leads and unplug USB devices you dont need, see if that changes anything.

Failing that this thread may help
https://forum-en.msi.com/index.php?topic=281412.0
These CPU sockets are very easy to damage and are not officially covered under warranty.
I had an Asus Sandybridge mobo that suffered bent pins and was refused warranty.
Take extreme care.
If there is a problem with the socket it might be wise to contact MSI first.

Footnote:
I have Corsair LPX Vengeance 3000MHz 4x4GB which took 3 motherboards before I could get it to run at rated speed.
It ran ok at 2133MHz on the first 2 though.
I needed a higher spec board with more layers to get 3000MHz.
The boards that failed were MSI Z170a Gaming Pro, ASRock Z170 Extreme 4+
The board that works and I am still using is the Asus Maximus VIII Hero and can now be overclocked to 3733MHz, thats the difference with a good spec board!
 
Page faults are RAM errors, the other ones could be attributed to a defective power supply as readily as the hard drive itself being bad (one, or the other, or a combination of both), and all the issues would absolutely happen with a defective power supply.

Of course, since Windows 10 is involved who the hell knows. :D

And remember: "new" hardware doesn't mean it works. :)
 
I've had a couple of cases where Memtest passes fine, but that runs under it's own little environment. When you boot to Windows, and it loads all kinds of other drivers, then the memory just doesn't seem to work anymore. Often it's something stupid like one DIMM slot on the motherboard doesn't work quite right, or the RAM won't run paired, only in single channel. The RAM and motherboard will otherwise work fine, so I've just attributed it to driver nonsense and lived with it once i finally get it in some stable configuration - usually by dropping to one DIMM and then working back up from there.

That isn't common though. Much more common to actually find a faulty DIMM, or PSU issues, or something like that.
 
Barf

Running Prime95 for the rest of the night. Good news is, after installing a fresh Windows 7, updating the BIOS and all drivers/updates, I haven't gotten any BSOD.

I'll check the pins in the morning, assuming I don't wake up the find it in flames
 
I've had a couple of cases where Memtest passes fine, but that runs under it's own little environment. When you boot to Windows, and it loads all kinds of other drivers, then the memory just doesn't seem to work anymore. Often it's something stupid like one DIMM slot on the motherboard doesn't work quite right, or the RAM won't run paired, only in single channel. The RAM and motherboard will otherwise work fine, so I've just attributed it to driver nonsense and lived with it once i finally get it in some stable configuration - usually by dropping to one DIMM and then working back up from there.

That isn't common though. Much more common to actually find a faulty DIMM, or PSU issues, or something like that.

Which memtest? there is a difference among them
 
Alright, ran Prime for a solid 13 hours with no issues. Going to run some benchmarks on the CPU and GPU today and see if everything's up to snuff
 
  • Like
Reactions: Nenu
like this
Alright then. Note to self; fuck Windows 10.

See you guys next week!

upload_2017-12-18_8-46-47.png
 
Passmark Memtest86

I ran some testing of those and a bunch of others
memtest86 v7.0x fairs pretty good. but if you are runinng the 86+ v5,x or the older v4 that is on the non uefi boot from passmark .iso you are losing a good amount of "detectabilty"
That might explain what you had experienced

might testing was based on timings and voltage issues though. no hard fault. and PRime95 blend trumphed everything else in for those errors.

Windows built in ramtest was absolutely sh#t and couldn't even detect memory errors in unbootable system or that had graphical glitches on the screen from ram errors.
its 100% useless in my book
 
Back
Top