Possible RAM Issue, or not?

[ERBarratt]

Hi all,

 

So I've had the same mobo+mem+cpu for ages, and decided to upgrade my HDD to an onboard m.2 pcie ssd.

 

Motherboard:ga z170n wifi

CPU 600K

Mem: HyperX Fury 16gb

Vcard: Gigabyte vega 56

SSD: Intel M.2 Pcie ssd.

PSU: 500W SFX something...

 

Re-installed windows ten onto it, and changed my GFX card to a second hand Vega 56 (I have freesync monitor).

 

Also did some cable updates (added braided cable extensions). 

 

All well for a few weeks....

 

Then I started getting BSOD's with thread_stuck_device_driver errors, AND freezes, in games and browsing or general work (I'm a software engineer). When freezing, audio also sticks to the last tone it was on (a loud buzzing tone) as I run audio through the displayport cable. No ryme or reason, and an unpredictable frequency, enough to where I'm paranoid in games saving all the time.

 

I started by thinking maybe Chrome, so I uninstalled, CCleaner cleaned registry and reinstalled. Crashes still happen.

 

Updated all drivers, did clean installs with AMD drivers etc. Machine up to date as best as I can see.

 

Still occuring.

 

Then one time it happened and the system refused to boot. So I cycled the power, hoping to drain residual electricity. Eventually posts, but I run into various BSODS, the most common being 0cx0000098 failed to load winload.exe but occasionally getting achi bios error, or pte_system_misuse, or pagefile errors. No definable pattern, always different on every boot.

 

The 0cx error suggested I use installation media, so I got my laptop and created a usb. Tested it on laptop, boots fine into windows install.

 

I try the usb on my machine and get the same errors.

 

After a bit of searching, I found a pattern of "memory issues" emerging from all the results. So I removed both sticks (2X hyperx fury 8gb) swapping them from each of the two mem slots, one at a time to see if it'd change anything. Nothing, same errors (0cx error when trying to load normal windows, other errors when trying to boot from usb).

 

Thinking I was going mad, I marked one mem stick to make sure I was trying both out in both slots. Every time, unplug mains power, cycle PSU switch, cycle main PC switch. One miraculous occasion, with one stick in the rightmost slot, it boots!!

 

Try auto repair through USB windows 10 install media, but it finds nothing, so I boot into windows, and everything seems fine.

 

I run mdsched.exe and test both sticks independently in both slots, then both together. All fine, no errors occured.

 

So my best guess right now is something like excess electricity in something like mem controller, or memory itself, or something and I just wasted enough time to the point where it drained and reverted. I'm typing on the PC in question to write this post.

 

I've done the wiggle test on all the cables, checked they're all in, checked the bios to make sure CMOS battery ok (overclock settings I've had for years still present and ok - and I don't have any weird memory settings on).

 

So, I've literally registered to see what the communities opinion is? 

Thanks in advance.

[PineyCreek]

Random question...does it do it with the old GPU installed?

[ERBarratt]

Had a RX 580 in it before, and not that I can remember - or at least so infrequently it wasn't annoying.

 

Now I am thinking, after some of the crashes I lose the audio device through the GFX card and have to restart to see it again.

[HoneyBadger84]

Definitely something where you should try a different video card if you have one available to do so.  Obviously if the problem continues, it's not the GPU itself.  

[ERBarratt]

Is there some sort of log output that might help someone smarter than me about these things diagnose this?

 

I've now run memtest86 through all it's tests using it's bootable usb feature and have found no errors so far.

 

Still got occasional freezes going back into the BSOD loops.

 

Interestingly, I can boot into stock safe mode when this happens - then I run dskchk, memtest etc and eventually get back into windows normal.

[Bitter]

If booting to safe mode works fine every time then it's likely some kind of driver issue, finding which one is going to be the fun part.

[ERBarratt]

You're telling me!

 

I have just updated AMD drivers to optional 19.2.3 with a clean install to see if that helped. Still froze.

 

It's a tad confusing as I don't have all that many thing that require drivers on the system. Just the main components themselves plus a usb 3 hub...

[Bitter]

It's been so long since I've had to diagnose an issue like that, but I seem to recall something called (maybe it doesn't exist anymore) selective startup where you can disable things one at a time but I think that may be only for startup programs and nothing to do with drivers. Everything uses drivers, whether you install them or Windows 'just comes with them', they're being loaded at boot time and used. Windows (and I'm showing my age) used to have a startup mode where you would line by line tick Yes to each thing loading until you hit what broke it so you could figure it out, but I think that vanished with Windows 98. You may be able to get some more information from the dump files I'm assuming are being generated at the failed boot, try to locate those and see what exact failures are being referenced.

[HoneyBadger84]

Did you try a different video card or your other video card to see if that has the same issue yet?

[ERBarratt]

I only have one - I've tried using the onboard display out hdmi.

 

Still doing it :(