Unstable system - CPU, RAM or MOBO?

[wizeton]

I have very similar use to this one created by @ReD-EyeD

Idle BSODs on 3700x

I'm in dire need of a solution or direction to what to do with my problem. Sorry if it's a long post, but I'm trying to do my best providing as much info as possible. I'm getting random BSODs during idle or low load stuff like watching youtube or fast scrolling pages. It can happen randomly...

www.techpowerup.com

My specs:

• Ryzen 7 3700x
• 4x16GB Crucial Ballistix LT Gray 3200Mhz
• Asus TUF Gaming X570

Background:
The main difference comparing to thread that I mention is that I bought this PC in November 2019 and I used it to this day without any issues in heavy games such as Metro Exodus, or Half Life Alyx playing on max settings.
Problems start to happen 2-3 weeks ago. I was getting a lot of BSODs just like that:

• PAGE FAULT IN NONPAGED AREA
• SYSTEM SERVICE EXCEPTION
• IRQL NOT LESS OR EQUAL
• CRITICAL SERVICE FAILED

Solving issues:
First thing I did was just Windows diagnostic. sfc scannow showed me some issues. Long story short, I decided to reinstall system on different SSD drive.
Issue is still there but at least I can run PC, BSODs not happens that often.
I used memtest86 on each stick RAM separately, which gave me a lot of errors at Test 6 Block move - tested in 2 different slots.
I used Prime95 which shows error message on 7 and 8 thread.
I used OCCT for ~10-15 minutes without any issue.
I updated BIOS to newest one: 1407

Whats happening now:
I can launch PC, but sometimes I can't and I getting boot loop.
I can't launch PC in DOCP profile but I set 1.375 on CPU and 1.35 on RAM manually, and that I worked at some point. Now Im back on default factory settings, and that somehow works again for me.
Im getting BSODs but not that often as before.
My CPU is constantly acting very strange in terms of overall usage.
Im getting freezes when doing basic stuff such as watching YT or Facebook,
I can't launch some apps, or apps are closing itself without my interaction,
A lot of issues with internet connection, corrupted files, packet loss.


What concerns me the most is that what causes that issue? I used it very often without any issue but when I noticed a similar issue on forums, people complain that it starts to happen soon after the assembling system.
I used stock settings in BIOS from start.

 

[Chomiq]

Why are you running with fixed cpu voltage? Memtest86 shows errors so it's most likely timing issue.

You're running quad channel with dual rank memory kits. Get Dram calculator and Thaiphoon burner and get stable timings for Micron E-die in dual rank setup.

All these errors have also resulted in corrupted system files so that's why you're seeing such behavior.

 

[Regeneration]

Your system is unstable due to improper memory configuration.

Remove any CPU overclocking until you find stable memory settings.

Use MemTest64 OR HCI MemTest (one instance by thread) OR Linpack Xtreme with maximum memory.

You should be getting ZERO errors.

 

[xtreemchaos]

what psu and gpu do you have ?. when you get time could you fill out your system spec in your sig it will save folk from asking

 

[laszlo]

in case you are on default bios settings and still get memtest errors most likely ram fault; try other ram and test them

 

[lsevald]

Lately I've been using GSAT (Google Stressful Application test) on a USB stick and booted from that to test RAM stability before risking the contents of my drives. I downloaded the one I've been using from here:

GSAT linux live cd (how to easily and safely stress test memory)

It's set up to run indefinitely, run it for an hour or two and press ctrl+c to end the test and get the report.

 

[wizeton]

what psu and gpu do you have ?. when you get time could you fill out your system spec in your sig it will save folk from asking

be quiet! Straight Power 11 650W 80 Plus Gold
RTX 2070

in case you are on default bios settings and still get memtest errors most likely ram fault; try other ram and test them

yes, I have plan to test other RAM

in case you are on default bios settings and still get memtest errors most likely ram fault; try other ram and test them

but 4 sticks got faulty at the same time?

 

[Chomiq]

be quiet! Straight Power 11 650W 80 Plus Gold
RTX 2070


yes, I have plan to test other RAM



but 4 sticks got faulty at the same time?

It's not that sticks are faulty. Your motherboard fails to set correct timings for them. You can get them to work by manually setting timings in BIOS.
Continued usage despite errors have lead to OS corruption so if you want to do anything from windows you have to reinstall OS.
So it's a mix of tricky setup (4x16), incompatible hardware running on auto and user errors.

Boot with single stick. Load optimized defaults.
Reinstall windows.
Install Thaiphoon burner and Dram calculator.
Follow some guide for the calculator to get correct timings for Micron E die, dual rank, 3200 safe preset. Print these settings.
Get one stick running correctly with no errors.
Install all four sticks. Then:

And get safe timings for 4 DIMM modules.
Verify if it runs without errors.

Yeah, only 8 GB Crucial sticks are listed on QVL for the board.

Getting a different set of 4 x 16 GB, no matter what brand, will probably lead to same sort of troubleshooting scenario.

 

[freeagent]

Those are all memory errors. Your mobo doesn’t seem to like how you have your ram setup, or it just doesn’t like that ram.

 

[wizeton]

Those are all memory errors. Your mobo doesn’t seem to like how you have your ram setup, or it just doesn’t like that ram.

I just came back from a RAM test which I did using friend G. Skill Trident Z stick 1x16GB, default BIOS - i also got errors on that on Test 6 Block Move. Then I used my Crucial RAM inside friend PC, default BIOS, no issue during 30 minutes of test on Intel machine, when I was getting soon after 5-6 minute on my PC with the same stick.

It's not that sticks are faulty. Your motherboard fails to set correct timings for them. You can get them to work by manually setting timings in BIOS.
Continued usage despite errors have lead to OS corruption so if you want to do anything from windows you have to reinstall OS.
So it's a mix of tricky setup (4x16), incompatible hardware running on auto and user errors.

Boot with single stick. Load optimized defaults.
Reinstall windows.
Install Thaiphoon burner and Dram calculator.
Follow some guide for the calculator to get correct timings for Micron E die, dual rank, 3200 safe preset. Print these settings.
Get one stick running correctly with no errors.
Install all four sticks. Then:
View attachment 156760
And get safe timings for 4 DIMM modules.
Verify if it runs without errors.

Yeah, only 8 GB Crucial sticks are listed on QVL for the board.

Getting a different set of 4 x 16 GB, no matter what brand, will probably lead to same sort of troubleshooting scenario.

I tried boot my pendrive with my Windows, but turns out Im getting BSODS also there.

 

[wizeton]

Update: I made some hardware tests

• I tested friend RAM, I got errors in memtest,
• I tested my RAM in friend PC, he didn't got any error in memtest
• I borrow Ryzen 7 2700, swapped CPU and during 3h of test, I didn't got any error

I'm working already on my PC with 2700 on board at least 4-5h without any BSODs. Everything is very smooth, stable.
Conclusion: CPU was the issue. I working right now on my warranty request, that's the only way.