Series of BSOD's and issues I can't wrap my head around

[bo.engelen]

My apologies for the long topic beforehand but I'm desperate.

So a a good month ago I got myself a new PC with the following specs:

CPU: AMD Ryzen 7 1700X
Mobo: MSI X370 GAMING PRO CARBON
GPU: MSI GeForce GTX 1060 ARMOR 6G
RAM: Corsair 16 GB DDR4-3200 (LED one)
SSD: Samsung EVO 850 500GB SSD
PSU: Corsair RM650X 650W ATX24

And I have been getting one issue after the other and I can't find what is causing this...

I have been getting a couple of different BSOD's with the following stopcodes:

- PAGE FAULT IN NONPAGED AREA (this one most of the times)

- KERNEL SECURITY CHECK FAILURE

- DRIVER OVERRAN STACK BUFFER

- MEMORY MANAGEMENT

- FAST ERESOURCE PRECONDITION VIOLATION

These are extremely random, most of the time when I"m just browsing or even when I try to shut down my pc...
Funny thing is that it has not happened while gaming for hours or editing footage in Adobe After effects and or Premiere, so that that tells me it isn't "stress" related?.
It does however happens a lot when I alt tab a game...


On top of that I also have encountered the following issues:

- Screen going black with only the mouse visible, when I restart I can see a "Display driver has failed to start" error in windows notifications when I force quit and restart.

- Noticing the windows loading circle comes up A LOT, like every time I switch a tab in chrome or when I hit the start menu or even just random when doing nothing...

- Overall sometimes sluggish performance, especially with everyday normal tasks (browsing chrome, navigating windows) not while gaming or video editing...


What I have tried so far:

- Complete clean Install of windows
- Set windows power options to high performance
- Updated the BIOS of my motherboard
- Manually downloading the latest stable drivers for my GPU, CPU, etc
- Performed a check disk and a memory test in windows which both gave no errors

My best guess:

- When researching all my BSOD's there is one thing that keeps popping up, while not always on top of the list of possible issues, bad ram keeps popping up in all of them.
Tough Windows memory check didn't report any problems

- I also found out that the RAM modules I bought (CMU16GX4M2C3200C16, Vengeance LED, XMP) are not on the list of compatible RAM list of both my motherboard and AMD's Ryzen list...
These ones where in serious discount when I bought them so I didn't bother to double check...
That being said, my CPU Ryzen 7 1700x) is also not on the list of compatible cpu list of my mobo..... which is very weird since all the ryzen 5 series and the 1800 are...



I'm at a point where I feel my PC is so unstable it becomes ridiculous 🙁
If new RAM resolved my problems than I would't mind to pay the price.
But RAM isn't cheap these days so I would like to have some opinions on the matter.

Any help is deeply appreciated!

 

[jdanniece]

As you said probably bad ram.

 

[bo.engelen]

Ok so now add to that a "rpc server unavailable" error when trying to opening files from the desktop and unable to go to settings from the start menu....
Temporary resolved when restarting

 

[johnbl]

normally, you would test your memory timings by downloading and running memtest86 on its own boot image.
if it works as expected (no error) you would then start to look at the windows memory dump files to see why windows is shutting down the system with a bugcheck. bad memory timings will cause corruption of the data in memory, this can cause any windows device driver to call a bugcheck. Some windows components assume a virus infection and call a bugcheck if its internal data held in memory gets modified.

sometimes it is something simple like several copies of over clocking drivers being loaded at the same time. These are easy to see with a windows debugger in the memory dump files. c:\windows\minidump directory

 

[bo.engelen]

normally, you would test your memory timings by downloading and running memtest86 on its own boot image.
if it works as expected (no error) you would then start to look at the windows memory dump files to see why windows is shutting down the system with a bugcheck. bad memory timings will cause corruption of the data in memory, this can cause any windows device driver to call a bugcheck. Some windows components assume a virus infection and call a bugcheck if its internal data held in memory gets modified.

sometimes it is something simple like several copies of over clocking drivers being loaded at the same time. These are easy to see with a windows debugger in the memory dump files. c:\windows\minidump directory

Ok so I ran memtest89, it took over three and a half hours and found 1 error in test 5.
While I was hoping for more drastic results, I will change the RAM and see if that solves my issues..

 

[johnbl]

BB = 1011 1011
3B = 0011 1011

dropped 1 bit during the hammer test.
generally you want to update the BIOS if it was not already updated, check the BIOS timings with the vendors chip timings. check the bios voltage is correct and rerun the test.

if it still fails check to see if it is at the same memory address.
maybe tweak the bios memory timings (slower) or voltage.

as a guess I would put a 2T command rate in BIOS for the secondary memory timings and rerun the test. cheaper RAM modules often require 2 clock ticks to setup the address bits. While most BIOS versions set the default timing to 1 T.

the memtest only shows the first 4 memory timings 16-18-19-36
Generally the 5th timing is the command rate. it should be indicated by a N or T

 

[bo.engelen]

BB = 1011 1011
3B = 0011 1011

dropped 1 bit during the hammer test.
generally you want to update the BIOS if it was not already updated, check the BIOS timings with the vendors chip timings. check the bios voltage is correct and rerun the test.

if it still fails check to see if it is at the same memory address.
maybe tweak the bios memory timings (slower) or voltage.

as a guess I would put a 2T command rate in BIOS for the secondary memory timings and rerun the test. cheaper RAM modules often require 2 clock ticks to setup the address bits. While most BIOS versions set the default timing to 1 T.

the memtest only shows the first 4 memory timings 16-18-19-36
Generally the 5th timing is the command rate. it should be indicated by a N or T

Updating the bios is the first thing I did. I also checked voltages and tried lowering speed to even 2133.
But didn't help anything. Adjusting the command rate I didn't try yet...

Now a buddy of mine has 2x8gb ddr4 ram laying around that IS on the compatible list of both AMD ryzen and my motherboard.

I'm going to try that in my pc and see what it does over the weekend.