PC freezes and restarts while gaming - help

[Werner Smal]

Hi everyone

Ok so for the past few weeks I’ve been having a problems that’s driving me crazy. Whenever I play Overwatch 2, every now and then my PC would freeze, sound will still be there for a few seconds but then that freezes as well, and a few second later the computer will restart. It is completely random, sometimes it takes 10 minutes sometimes I can game for hours. No blue screen or anything but I do hear an error message but I can’t minimize to see what it is. It doesn’t happen on heavier games, Cyberpunk runs smoothly.

While troubleshooting the problem, I had a few other issues. While installing Windows the PC would restart while installing corrupting the files, and I had to take my PC to a friend just to try and install Windows properly. Whenever I tried installing GPU drivers, PC would restart while installation is running, also corrupting the files.

The problem was gone for a week or so, but I bought a new case Yesterday and the problem is back.

What I’ve done to try and fix the problem:
- Sent my PC to Rectron, they replaced the motherboard and the CPU
- Reinstalled windows like 15 times and on a new NVME
- Bought a new power supply, I think that is what solved the problem at first, but now it’s back
- Reinstalled every driver I could possible think of And updated bios
- Tested it with another GPU
- Tested it with other RAM

My PC:
- x570 Auros pro
- Ryzen 9 5900x
- Powercolor 6900xt Red devil edition
- Trident Z Neo 3600mhz cl16
- 1 Sabrent NVME and 1 Transcend NVME
- Super Flower SF-10000 Plat 1000w power supply
- New case: Cooler Master Masterframe 700

Any help would be much appreciated!!

 

[Bloodlines]

Power supply is faulty ?

 

[Werner Smal]

I doubt the new power supply is faulty as it kinda did fix the problem, but I mean at this point anything is possible

 

[dash09]

Replicate the issue, then check event viewer logs. Thats the starting point

 

[CarboVan]

Which cooler are you using? Have you monitored temps to rule that out?

 

[Werner Smal]

Temps are as cold as can be, been checking them constantly and nothing out of the ordinary.

Event ciewer doesn’t help cause it doesn’t show why it crashed. Weird thing is everytime I go and check event viewer, there is a error message saying the computer restarted unexpectedly but at that time I didn’t game and there was no unexpected restart

 

[Xtreme]

Corrupting files while installing window means faulty ram or nvme. Check speed and timings or even new sticks.
Set all to default after replacing the ram then report.

 

[Werner Smal]

Made sure the other RAM I tested was accepted on Aorus website with my motherboard and XMP hasn’t been on in weeks

 

[CarboVan]

Your PBO settings?
Perhaps disable all those all well. Or can even try adding tiny positive offset volts? or lower the power limits 🤔
I'm just throwing ideas in a hat here

 

[Werner Smal]

Your PBO settings?
Perhaps disable all those all well. Or can even try adding tiny positive offset volts? or lower the power limits 🤔
I'm just throwing ideas in a hat here

haha thanks for the ideas, I need them and I haven’t tried this, will definitely give it a go

 

[animehero]

I would have said it seems like a grounding issue but you've changed the chassis.

So to put into perspective

New CPU
New MB
New PSU
New Case
Same GPU
Same RAM
Same Storage

And you are sure its the exact same symptoms from before the changes were made?

Sounds like something just doesn't like playing nice in this rig even if individual components test fine.
Its actually a bloody nightmare to figure out, I had a similar issue a while back and could never get to the bottom of it.
In the end I ended up selling off individual components and those all work 100% with their other owners while I got different parts.

Start at the basics though, Remove the one NVMe drive, use 1 stick of RAM in the first RAM slot of the MB and reinstall windows + GPU drivers. See what happens when trying to to play Overwatch 2. Rinse and repeat with other RAM and Slots on MB. If everything checks out, add another stick and see what happens

Essentially using the least amount of components possible to see if you can replicate the issue helps pin point the issue
Perhaps check out OCCT to stress test the system as well

 

[WurminatorZA]

Check your BUS speed is at default of 100Mhz, Revert any CPU, GPU, RAM overclocks start at default motherboard settings and work from there. Install a previous BIOS also to test

 

[Werner Smal]

I would have said it seems like a grounding issue but you've changed the chassis.

So to put into perspective

New CPU
New MB
New PSU
New Case
Same GPU
Same RAM
Same Storage

And you are sure its the exact same symptoms from before the changes were made?

Sounds like something just doesn't like playing nice in this rig even if individual components test fine.
Its actually a bloody nightmare to figure out, I had a similar issue a while back and could never get to the bottom of it.
In the end I ended up selling off individual components and those all work 100% with their other owners while I got different parts.

Start at the basics though, Remove the one NVMe drive, use 1 stick of RAM in the first RAM slot of the MB and reinstall windows + GPU drivers. See what happens when trying to to play Overwatch 2. Rinse and repeat with other RAM and Slots on MB. If everything checks out, add another stick and see what happens

Essentially using the least amount of components possible to see if you can replicate the issue helps pin point the issue

My one friend also things it might be a grounding issue, I could of made a small mistake putting everything in the new chassis?? I don’t know… how do I even start checking that?

The thing is I have done all that, even borrowed a friends nvme, remvoed the ram checked each one and the problem was still there.

At this point I feel like selling the whole damn thing and buying everything new

Jip, symptoms are exactly the same

 

[PerformancePCs]

Had a similair issue before, turned out the voltage on the cpu was below a certain threshold (under load) causing the memory controller to behave weirdly and cause restarts.

Try a different (lower spec) cpu on that board with the latest bios and see if it persists. I solved the issue by changing out cpu (in your case, you may have to change board if it solves the problem).

 

[JollyJamma]

Had a similair issue before, turned out the voltage on the cpu was below a certain threshold (under load) causing the memory controller to behave weirdly and cause restarts.

Try a different (lower spec) cpu on that board with the latest bios and see if it persists. I solved the issue by changing out cpu (in your case, you may have to change board if it solves the problem).

Rectron replaced the CPU and motherboard.

OP have you tried using manually set voltages for almost all settings?

 

[Werner Smal]

Rectron replaced the CPU and motherboard.

OP have you tried using manually set voltages for almost all settings?

No I haven’t and in all honesty I’m not clued up with changing voltages and all that

I’ll do some research and see what happens

 

[WurminatorZA]

Looking at reviews of your Motherboard on newegg, seems others have similar issues with this board. GIGABYTE X570 AORUS PRO AM4 ATX AMD Motherboard - Newegg.com

 

[JollyJamma]

No I haven’t and in all honesty I’m not clued up with changing voltages and all that

I’ll do some research and see what happens

I had an issue with my X570 Tomahawk where I was getting errors that I couldn't track down. I've got like 5 RAM kits and multiple SSDs.

I set my RAM (1.45v) and SOC voltage (1.1v for SOC was important) to manual and all was fine. Looked like the SOC was the issue.

I also had issues with Ryzen Master software being a complete dick. AMD does seem to have more bugs than Intel but they have been very good of late. The only issue I had recently was that voltage issue which can be a poor BIOS setting from the manufacturer.

 

[animehero]

My one friend also things it might be a grounding issue, I could of made a small mistake putting everything in the new chassis?? I don’t know… how do I even start checking that?

The thing is I have done all that, even borrowed a friends nvme, remvoed the ram checked each one and the problem was still there.

At this point I feel like selling the whole damn thing and buying everything new

Jip, symptoms are exactly the same

Well grounding isn't an exact thing, I guess making sure nothing but the motherboard stand offs are touching the back of the MB.
Making sure no loose hanging stuff touches other components etc. Basically anything that could short/ground components around it should be safely placed away to avoid accidental damage. Although to be clear I don't think this is the problem with your setup at the moment

 

[baasgene]

diagnosing an issue can lead to a fuck load of frustration lol. My laptop for example will freeze once in 3 months whilst I'm working, and then it takes me literally 10 minutes to get it powered on again. From there it's fucking rosy for another 3 months at least, I've redone windows and reseated NVME and Ram to no avail. Memtest passes with flying colors, nvme seems perfectly good according to hddsentinel.

Anyway that sounds like a slight instability with the ram perhaps? Have you run a proper memtest yet? Maybe do 30 loops in safe mode and see if it gets any errors?

Easiest way to check if it's the ram is to disable xmp probably and see if it still happens.

 

[animehero]

I had an issue with my X570 Tomahawk where I was getting errors that I couldn't track down. I've got like 5 RAM kits and multiple SSDs.

I set my RAM (1.45v) and SOC voltage (1.1v for SOC was important) to manual and all was fine. Looked like the SOC was the issue.

I also had issues with Ryzen Master software being a complete dick. AMD does seem to have more bugs than Intel but they have been very good of late. The only issue I had recently was that voltage issue which can be a poor BIOS setting from the manufacturer.

To add to what @JollyJamma was saying here

This is what I came across in my quest to find a solution to my issue at the time

https://www.reddit.com/r/Amd/comments/kw1o9f

Basically boils down to how MB/CPU and RAM co-exist in the system

 

[baasgene]

Ryzen used to be notorious for instability, and it does seem like some combinations can still cause such headaches. I've been one of the fortunate ones to report that not a single intel in my life has ever been this stable, and yet I would still upgrade to intel 13th gen next year just because my blood runs blue.

And of course 6ghz sounds like a fun number!

 

[JollyJamma]

Here Help with manual overclock of Ryzen 9 5900X

 

[Xander8807]

pm'ed our mate, willing to go through the whole system and see what the issue could be (15yrs tech exp behind me and helped out a few other members till date), and Yes Aorus BIOS sometimes does tend to quack out with its PBO and Voltages.
Have 2 amd systems thats compatible with his hardware so ill test each thing on my rigs too see if i can get this issue as im running Asus and been stable for more than year no issues.
Will update you guys on what the issue was if i manage to find it and help out our mate to get him back up.

 

[Areola Grande]

@niestillhere looks like some PC parts just became available.

 

[Xander8807]

@niestillhere looks like some PC parts just became available.

that oke seems like a scammer to me, how is he even on carb still? gonna make it a friday funny? @Toy Machine ?

 

[JollyJamma]

Oh and I don't want to be a dick but Gigabyte are known for their weird as hell general slap dash approach to BIOS quality.

I used to be their biggest fan but I moved on ages ago.

Z690 Gigabyte motherboards are hated on in many forum posts labelling the quality as 'BETA' as best.

 

[Tapsumbong]

My Gigabyte b550 board caught fire whilst flashing the bios

 

[AceSwiftShooter]

are you using an aftermarket nvme heatsink? I had an issue once where I installed a EKWB heatsink, but it was not compatible (NVME connector height too low).
This caused the pc to always restart cos of a short fromm the heatsink to mobo.

 

[Xander8807]

are you using an aftermarket nvme heatsink? I had an issue once where I installed a EKWB heatsink, but it was not compatible (NVME connector height too low).
This caused the pc to always restart cos of a short fromm the heatsink to mobo.

highly doubt he is, seems like a base build, so will be the factory product