• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Page fault in nonpaged area and system_pte_misuse BSOD

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
I watched realbench, I saw that luxmarkx64 crashed. Well, I'm guessing that means my oc isn't stable? This actually my first time to use realbench. I never really used realbench before to determine if it was stable or not. I'll just keep on using this overclock and see if it gives me problems on regular use. So far the page fault BSOD hasn't replicated today yet.

Edit1:
I've decided I'll try 1.4v DRAM Voltage. To my surprise, realbench is running longer now. I set it on an 8hr stress test. Will update results when I wake up. Now, I have a question, if this ram cannot work on it's stock voltage can I rma it?

Edit2:
update, a few minutes after I typed edit1 I got the page fault bsod. Well, realbench lasted longer after I adjust the dram voltage. Now I swapped slots from a1b1 to a2b2. I saw in another thread someone's ram was unstable because of that. I double check the Maximus IX Hero's manual it also recommended A2B2 if only using dual channel kit. Well I re-run realbench. See if it does better.

Edit3:
Okay, after swapping DIMM slots it seems to be stable now. Almost finished in realbench 8hr stress. I'm trying at 1.4v Dram voltage. Will stress at stock voltage later tonight.

Edit4: I tried putting the DRAM Voltage back to stock, I immediately got welcomed by that page fault bsod as soon as I logged into windows. I guess the ram isn't really stable at stock 3200 MHz @ 1.35v. You guys think I should RMA this for not being stable at stock settings?
 
Last edited:
Okay, I set the DRAM Voltage to 1.4v seems to be stable. However, I found out that the BSOD (Page fault and System PTE Misuse) only happens if I start my computer from a cold boot. Also, it has come to my attention that it maybe because of some software or driver from Asus since the Page fault BSOD also mention IOMAP64.sys. But I really don't know, I guess I am to avoid cold booting to avoid BSOD on windows log on for now.
 
If your memory runs fine stock, then when you use XMP it is not stable then I would RMA.
 
Last edited:
I'll try to RMA it. I also tried talking to G.Skill whether this issue is RMA'able. Unfortunately, I think they still have a CNY hangover. lol

Edit: gotta add it isn't stable in stock either. The only way to make it more stable is 1.4v DRAM Voltage and it still gets BSODs just on cold boot though.
 
Last edited:
Why don't you run the memory stock speed for a while just to make sure there is not any problems.
 
Last edited:
Why don't you run the memory stock speed for a while just to make sure there is not any problems.

Stock meaning at 3200MHz? (it is a 3200MHz kit) or 2133MHz? If so I already tried it and can't get to pass realbench stress. Also get immediate BSOD of page fault after I enter my password to log in windows. If ever it gets past windows log on it will BSOD(page fault) on load. This is how I found out 1.4v makes its stable except on cold booting. (I tried .1v increments until 1.37v and it still bsods so I decided to try 1.4v and found out it was stable no matter what stress test I throw at it.)
 
I notice from your sig that your CPU is delidded. Could there be damage from that process to the chip?
 
I notice from your sig that your CPU is delidded. Could there be damage from that process to the chip?

Upon checking my delid it doesn't look like anything is wrong, no burns, chipped contacts or cracks on the wafer(used vice only method). I have been in contact with G.skill though, they requested I do two more final test before ruling out a faulty kit. Since, I haven't tried reseating the cpu yet I'll try it and they've asked me to try 2800MHz and 3000MHz if it would still BSOD(Although I have already tried running it without XMP[2133MHz] and it still gave a BSOD). If it still BSODs they said it is probably a faulty kit.

The CPU is actually stable at 5GHz on whatever stress test I throw at it.(Prime95(more than 8hrs), Realbench 8hr stress, Intel XTU(8hr stress), Cinebench, 3d Mark) this is with 1.4v Dram voltage. If I put it to stock dram voltage which is 1.35v it would crash realbench and do random page fault BSODs.
 
Last edited:
Sounds like defective memory if the only thing you can do to make your system stable is increase the Ram voltage. How stable is the PC with setting Bios optimized default?
 
With IOMap64.sys in the frame, I'd suggest you rule that driver/software (Asus Smart Doctor/iTracker 2, Asus GPU Tweak or GPU Tweak Streaming?) out by uninstalling it. Then setup your RAM using XMP and tweak the XMP settings to match those given by the RAM maker. On restart, check for any 3rd party/utility drivers loaded using DriverView.

STOP 0x00000050: PAGE_FAULT_IN_NONPAGED_AREA
Usual causes: Defective hardware (particularly memory - but not just RAM), Faulty system service, Antivirus, Device driver, NTFS corruption, BIOS
...
The PAGE_FAULT_IN_NONPAGED_AREA bug check has a value of 0x00000050. This indicates that invalid system memory has been referenced.
...
If the driver responsible for the error can be identified, its name is printed on the blue screen...


Common factor: Device driver.

Memory doesn't always refer to physical memory (RAM), it's often the data that's loaded into working memory: when a 'bad' driver loads into a location that includes data already loaded for *something else*, the next time *something else*'s data is called for by the CPU it crashes because the data that should be held there is corrupt, replaced by data from a 'bad' driver (invalid system memory has been referenced).
 
I think satrow's words need some consideration. How about running the system in safe mode to bypass the device drivers and seeing if it fixes the issue.
 
Back