Windows 10 NTFS.SYS BSOD Loop - Unique Situation.

MNDMorris

New Member
Joined
Sep 24, 2018
Messages
15
Hi

This is my first post on the forums as i'm unable to find the solution online. I've searched all over Google for a similar case to mine and i'm yet to find any.

I recently ran in to a BSOD during gaming on Windows 10 64-bit. It occurred when I was playing Black Ops 4. I managed to get two quick games in and a BSOD occurred on the third session.

I then managed to reboot in to Windows and my nVidia drivers were out of date for Black Ops 4, I downloaded them and noticed Windows had a security update also. I installed the Windows update AFTER the nVidia drivers. I've now learnt that those Window drivers were Spectre/Meltdown patches, which for some reason, Windows hadn't installed when they first came out.

I was unable to rectify the BSOD occurrance and it continued to happen when trying to run the Black Ops 4 client and the launcher itself. This got worse and eventually lead to a BSOD loop. The BSOD each time was different, never came over a pattern or anything to point me in the right direction.

I was initially able to boot in to Safe Mode in Windows and remove all recent drivers to see if it was an issue from that perspective, this did not resolve the BSOD issues I was having. I eventually decided on a fresh install for Windows 10 via USB. This installed fine and I figured I had a corrupt driver. After the fresh install, I attempted to apply my overclock settings and XMP profile to the CPU and RAM respectively. For information, I was previously sat at 5ghz on the i5 7600k CPU @1.29vcore, with the XMP Profile set to 3000Mhz which was the rated speed for the RAM I bought. This has been running stable since October 2017, not one single BSOD from this set up. On attempting to apply the same settings, I was only able to clock the CPU at 4.5ghz @1.29vcore. A whole 500mhz out! I assumed this was because of the Spectre/Meltdown patches. The RAM continued to run at the rated 3000Mhz via the XMP Profile.

Two days went by and I was streaming a football match online. 80 minutes in and the system BSOD again! I wasn't doing anything unusual and the system was running as normal, no lagging or display issues to pre-warn of a BSOD.

This time around there was a common theme around NTFS.SYS BSOD, it occurred more often than not and sometimes pointed towards a Kernel issue. I didn't make note of the BSOD codes at the time but it was primarily an NTFS.SYS code. I'm now stuck in a permanent BSOD loop. I can't boot to safe mode, advanced repair or install Windows 10 from a USB without a BSOD occurring. There are times when Windows will boot in to the desktop, but on trying to do ANYTHING, it'll BSOD.

I've since flashed my BIOS to a previous version, prior to the Spectre/Meltdown patches. This hasn't helped and the system is still stuck in a BSOD loop.

I've removed my GPU, disconnected my HDD, disconnected the PC tower USB connection, removed one RAM stick and changed them through to check, removed any additional peripherals from USB etc. I still have the BSOD loop issue. I've even disconnected my M.2 SSD on top of the above listed and tried to run a Windows 10 USB, the USB attempts to run up the installation media, but BSODs shortly after initialising, that's with NO HDD/SSD connected either! I've also applied a CMOS reset several times. The motherboard's EZ Debug tool is highlighting VGA and DRAM as being the issue, even though i've removed my GPU and i've also tested the RAM as far as possible.

I'm able to run memtest86 and this showed that there were NO errors on the memory.

I now feel i've exhausted all possible options and i'm at a complete loss as to what to do next. If no one can help, i'll have to start returning components individually under their 12 month warranty. Which might not guarantee a fix...

I'm intending on running a third party application tonight to wipe and format the SSD and leave the HDD disconnected for now. My concern is that M.2 SSD architecture is different to a HDD, can I still format and wipe it with DBAN for example? I've been unable to install DBAN on to a USB as of yet. I have access to a laptop.

For information my system specification is:

i5 7600k, delidded and previously clocked at 5.3Ghz stable, downclocked to 5Ghz for the last 10 months. Peak temperatures on stress testing are no more than 55 degrees on full load.
NZXT Kraken X62 AIO Liquid Cooler.
Corsair Vengeance LPX @3000Mhz on XMP.
Gigabyte GTX 1080 G1 Gaming.
Samsung 970 Pro 1TB M.2 SSD (used for Windows OS and games library)
Western Digital HDD 1TB (used for storage)
MSI Z270 Gaming Pro motherboard
Corsair RM550x PSU with Corsair premium sleeved cabling
NZXT H500i tower

Any advice would be greatly received, I honestly don't know what to do next.

Thanks for reading guys, please leave comments below.

Mike
 
Sort by date Sort by votes
Hi,
for some reason i didn't see your post till now.. Anyhoo, so you can now, at least, run the cpu at stock?

If this is the case why not try running the Intel diagnostic again?

Something else you could try is to keep everything at stock values and then stress test first the cpu and then the gpu.

Prime95 will help with the cpu.

Furmark you can use to test the gpu.

See if either crashes/blue screens when tested. (An hour should do)

Also if you could post a dump file or two or even run the app in the sticky that would be awesome.
 
Upvote 0 Downvote

Hey Kemical

I ran a CPU stress test overnight at a manual clock of 3.8ghz. Which is stock, yes. I did however, disable any boost functions in the BIOS. The CPU stress test ran for around 4 hours. Finishing around 3:30am. There was a BSOD shortly after the test, followed by two more BSODs and a corrupt BSOD minidump. I was asleep at the time, so I came to a frozen BSOD screen in the morning.

I'll have to start putting the minidumps in to a folder and upload them to here. All the minidumps seem to be pointing towards drivers, specifically GPU drivers. A friend of mine has an Intel Z270 motherboard for me to use next weekend, so I'll carry out further testing then. I should be able to swap the motherboards through and determine whether it's the CPU or the motherboard. Fingers crossed it's the motherboard because it's still under warranty and my CPU isn't, due to the delid and overclock in the past.

I'll get some info over tonight
 
Upvote 0 Downvote
Morning MNDMorris,

so it sounds like from what you say above that the cpu didn't bsod during testing but did so (3 times) after?

So perhaps the cpu is ok after all?

Further testing will show either way and you getting the spare moBo will be a great help.
MNDMorris said:
All the minidumps seem to be pointing towards drivers, specifically GPU drivers.
Click to expand...

You could always try older gpu drivers. The following link has all recent Nvidia gpu drivers available for download:
Link Removed

I always use the DDU when changing drivers just so nothing is left of the previous driver:
Display Driver Uninstaller (DDU) V18.0.0.1 Released. - Wagnardsoft Forum

Good luck with the testing.
 
Upvote 0 Downvote

I'm honestly not sure.

I'll run a stress test on the GPU and CPU again tonight, with the software you've recommended, to see if that shows any problems.

To reiterate, any issues i've had all seem to be relating to how the CPU is being ran. For instance, underclocking the CPU with everything else on Auto allows it to boot in to Windows fine. As soon as I attempt to run stock on Auto, or overclock manually, BSOD loops and issues occur immediately.

I have no confidence in what the BSOD minidumps are reporting, with regards to GPU/Driver issues. It's been SO varied in terms of which BSOD comes after the other. The motherboard is highlighting VGA and DRAM as being issues also, which doesn't help at all. It's just a general direction.

As is stands, the CPU won't run at stock settings. That indicates either the CPU or MoBo are at fault. The difficulty I have at the moment is confirming which one of them is at fault. I'm unable to send off the MoBo on a RMA because I have zero confidence in diagnosing the fault from my side. I don't want to waste potentially a month, for MSI to bounce the MoBo back to me saying it's fine.

Extremely frustrating scenario, to say the least.

Thanks for the support the last couple of days, i'll post back some results once i've tested this evening.
 
Upvote 0 Downvote
Agreed, it does sound more likely that the cpu is in error. I just hope testing will bring a definite answer, good luck!
 
Upvote 0 Downvote
kemical said:
Agreed, it does sound more likely that the cpu is in error. I just hope testing will bring a definite answer, good luck!
Click to expand...

Put my CPU in to another motherboard last night. BSOD's again on stock speeds. Only seems to run stable below 3.8ghz. Not sure how the CPU is damaged and i'm unable to confirm through means of testing, which part of the CPU is actually broken. I find it crazy how it's been sat at 5ghz overclocked at 1.29vcore for 10 months and it suddenly goes. I've only been using it for light gaming sessions and general use.

It also seems like a HUGE conincidence that this occurred on installing new drivers, Windows updates and installing Black Ops 4 on the same day. Bearing in mind the BSOD started occurring whilst playing Black Ops 4. God only knows how this has happened lol
 
Upvote 0 Downvote
Hi,

it might be the case that this particular cpu was gonna crap out on you whether you overclcocked it or not and it's simply a case of coincidence? (and the silicon lottery of course).
Is it too late to RMA? (I know overclocking is supposed to invalidate warranties but could be worth a try? Just don't tell em lol )
 
Upvote 0 Downvote

I've delidded the CPU to replace the TIM with liquid metal! Not a chance of an RMA unfortunately...I didn't even reseal the gasket and used the MoBo CPU clamp to hold it in place. Wanted to be able to replace the liquid metal in the future.

It STILL runs at 3ghz stable, it'll do the job until 9th gen Intel comes out or Zen 2. Honestly thinking about Zen 2 though

Thanks for the last week or so. Appreciate the help.
 
Upvote 0 Downvote
MNDMorris said:
I've delidded the CPU to replace the TIM with liquid metal! Not a chance of an RMA unfortunately...I didn't even reseal the gasket and used the MoBo CPU clamp to hold it in place. Wanted to be able to replace the liquid metal in the future.
Click to expand...
You fiend...

Best of luck with the new build and would love to see it once it's done. We never close threads so please always feel free to drop by if only to share some pics..
 
Upvote 0 Downvote
kemical said:
You fiend...

Best of luck with the new build and would love to see it once it's done. We never close threads so please always feel free to drop by if only to share some pics..
Click to expand...

My system is pretty nice at the moment, only difference with next gen intel or AMD will be the aethestics on the motherboard. I'd like a TUFF Asus board to be honest, they look amazing! Even a Maximus or Crosshair would be sick.
 
Upvote 0 Downvote
I have the Crosshair Hero which is my first ASUS moBo and I really like it. I always used Gigabyte in the past as their top boards are feature rich and usually hard to beat. Now would i go back to gigabyte now I've tried the Crosshair? No chance!
 
Upvote 0 Downvote
You must log in or register to reply here.

Similar threads

P
  • Solved
Windows 10 Win10: BSOD loop due to driver verifier, cannot disable verifier, cannot boot to safe mode
Replies
2
Views
4K
ubuysa
U
A
  • Solved
Windows 7 ntfs.sys BSODing Randomly
Replies
15
Views
5K
AGuyInNeed
A
J
  • Solved
Windows 7 Dubugging shows BSOD "probably caused by Ntfs.sys, ntkrnlmp.exe, win32k.sys."
Replies
6
Views
4K
usasma