PC Randomly Rebooting During Gaming

[victor6946]

GPU: GIGABYTE RX7800XT - Driver 23.10.01.10

CPU: Ryzen 7 3700x - Chipset 5.08.02.027

Motherboard: ASRock Fatal1ty X470 Gaming K4 - BIOS P4.90

RAM: Corsair Vengeance LPX 16 GB (2 x 8 GB) DDR4-3200

PSU: Corsair TX750M Gold 750 W 80+ Gold

OS: Windows 11 Pro - 10.0.22631 Build 22631
 

Problem: So far it's only been with MWIII, but after about an hour of gaming my PC will just restart. My screens will go black and it takes about 5 seconds before the boot screens come up. In Windows Event viewer I get the following WHEA-Logger messages...
 

A fatal hardware error has occurred.

 

Reported by component: Processor Core

Error Source: Machine Check Exception

Error Type: Cache Hierarchy Error

Processor APIC ID: 4



So far this has happened twice, one today and once yesterday. The errors are logged in pairs. In the screenshot above it APIC ID 4 and 10. The first time it was APIC ID 2 and 13. The only thing that is new to my system is the GPU, which I got about a month ago.
 

Anyone have any ideas as to what's going on?
 

What I've tried: I let MemTest do 4 runs, but at the end I didn't get any errors. Same with Prime95. My CPU has never been overclocked. And XMP is not set for the RAM. The GPU has not been overclocked either.

[Neutraliz]

I don't know if thoses WHEA errors are the same as WHEA_UNCORRECTABLE_ERROR bsods. 
But WHEA_UNCORRECTABLE_ERROR is definitly not a good sign for your CPU, or ram. 
I really can't say for sure, but it wouldn't be the first time I've seen Zen 2 CPUs dying randomly on the internet. 

[Whatisthis]

Most likely ram timings. Your new gpu is pushing your cpu/ram harder. That Corsair lpx is the most issue prone ram with Ryzen.

 

bump up the voltage / loosen the timings / drop the speed and see if the issue goes away.

[Mark Kaine]

On 11/19/2023 at 7:07 AM, victor6946 said:

What I've tried: I let MemTest do 4 runs, but at the end I didn't get any errors. Same with Prime95. My CPU has never been overclocked. And XMP is not set for the RAM. The GPU has not been overclocked either.

99.99% its your ram, just try some none corsair that are on the qvl.

 

 

Vengeance + Ryzen is never a good fit and a known issue, tbh

 

[Bjoolz]

19 hours ago, Neutraliz said:

I don't know if thoses WHEA errors are the same as WHEA_UNCORRECTABLE_ERROR bsods. 
But WHEA_UNCORRECTABLE_ERROR is definitly not a good sign for your CPU, or ram. 
I really can't say for sure, but it wouldn't be the first time I've seen Zen 2 CPUs dying randomly on the internet. 

It's basically the same. And WHEA doesn't monitor RAM, only the CPU and PCIe devices. It uses the CPU's error log to see what is wrong. If you use RAM that is clocked higher than the recommended from AMD/Intel you could see this crash, but that's because it overclocks the memory controller. The recommended speed is often lower than what people use for XMP. 

 

@victor6946 The WHEA event is pointing to the CPU. Remove any and all overclocking/undervolting, that includes XMP/DOCP speeds over 3200MT/s. 

[victor6946]

On 11/21/2023 at 4:12 PM, Bjoolz said:

It's basically the same. And WHEA doesn't monitor RAM, only the CPU and PCIe devices. It uses the CPU's error log to see what is wrong. If you use RAM that is clocked higher than the recommended from AMD/Intel you could see this crash, but that's because it overclocks the memory controller. The recommended speed is often lower than what people use for XMP. 

 

@victor6946 The WHEA event is pointing to the CPU. Remove any and all overclocking/undervolting, that includes XMP/DOCP speeds over 3200MT/s. 

CPU has been running stock since I bought it, no overclocks/undervolting done to it. XMP is also not set at the moment, but the restarts are still happening. 

[Bjoolz]

4 minutes ago, victor6946 said:

CPU has been running stock since I bought it, no overclocks/undervolting done to it. XMP is also not set at the moment, but the restarts are still happening. 

Then the CPU is the main suspect. Have you also disabled Precision Boost Overdrive in the BIOS? 

[victor6946]

25 minutes ago, Bjoolz said:

Then the CPU is the main suspect. Have you also disabled Precision Boost Overdrive in the BIOS? 

Just checked my BIOS, Precision Boost Overdrive was set to AUTO so I set it do DISABLED. Although isn’t AUTO the same as DISABLED? Running a stress test in 3DMark seems to trigger the reboots, so I will try that now.

[victor6946]

On 11/21/2023 at 5:41 AM, Whatisthis said:

Most likely ram timings. Your new gpu is pushing your cpu/ram harder. That Corsair lpx is the most issue prone ram with Ryzen.

 

bump up the voltage / loosen the timings / drop the speed and see if the issue goes away.

I have basically left everything as Auto / Default in the Bios. 
 

Memory speeds are already 2667MHz.
 

If it’s an issue that is happening because of the new GPU, would it be better to just upgrade to the newest CPU my motherboard (still AM4) can handle? And swap out the memory with something else?  

[victor6946]

On 11/21/2023 at 6:55 AM, Mark Kaine said:

99.99% its your ram, just try some none corsair that are on the qvl.

 

 

Vengeance + Ryzen is never a good fit and a known issue, tbh

 

Is there anyway to know for sure if it's just the memory and not the CPU? 

I only ask since @Bjoolz seems to think the CPU is the main culprit. 

[saminator0107]

1 hour ago, victor6946 said:

Is there anyway to know for sure if it's just the memory and not the CPU? 

I only ask since @Bjoolz seems to think the CPU is the main culprit. 

I agree. Lets try to pinpoint the issue. I dont blame OP for not wanting to throw the "parts cannon" at the computer....lol.

[Mark Kaine]

1 hour ago, victor6946 said:

Is there anyway to know for sure if it's just the memory and not the CPU? 

I only ask since @Bjoolz seems to think the CPU is the main culprit. 

i mean there are many ways...

 

update your bios

 

update chipset drivers

 

ddu/update gpu drivers

 

run av scan

 

buy new ram (finally) 

 

buy a new cpu...

 

 

26 minutes ago, saminator0107 said:

I agree. Lets try to pinpoint the issue. I dont blame OP for not wanting to throw the "parts cannon" at the computer....lol.

yeah, i understand why they don't want to lol... but that's what it ends up being usually anyways,  there's just some steps before doing that.

 

 

fact remains, Vengeance ram =

[Mark Kaine]

1 hour ago, victor6946 said:

If it’s an issue that is happening because of the new GPU

I mean that's entirely possible,  think about it, so far pc has been working fine i guess??

(bit a lack of info in your op tbh)

 

so you get a new gpu and it doesn't anymore??

 

why should it be the cpu's fault (or ram) *in this case*, doesn't make any sense?

 

 

[victor6946]

1 minute ago, Mark Kaine said:

i mean there are many ways...

 

update your bios

 

update chipset drivers

 

update gpu drivers

 

run av scan

 

buy new ram (finally) 

 

buy a new cpu...

 

 

yeah, i understand why they don't want to lol... but that's what it ends up being usually anyways,  there's just some steps before doing that.

 

 

fact remains, Vengeance ram =

BIOS is on the latest version that isn't a BETA release.
Chipset and GPU drivers should be on newest, unless something new released that I completely missed. Have tried reinstalling with using DDU as well
Windows Defender shows nothing. 

Good thing I live near a MicroCenter... How important is it that the memory Mfr Part# shows up exactly the same in the Compatibility List? For example this Trident Z RGB 32GB (2 x 16GB) DDR4-3600 PC4-28800 CL18, I don't find it in the list.

[victor6946]

4 minutes ago, Mark Kaine said:

I mean that's entirely possible,  think about it, so far pc has been working fine i guess??

(bit a lack of info in your op tbh)

 

so you get a new gpu and it doesn't anymore??

 

why should it be the cpu's fault (or ram) *in this case*, doesn't make any sense?

 

 

I was running a 1070ti before this with no issues. And had been running that same setup since the release of the 3700X. (2019)

Only thing that changed was I swapped the 1070ti with the 7800x on 10/16. And it was running fine with no problems until 11/17, which when I first got these random reboots / event viewer errors.

[Mark Kaine]

On 11/21/2023 at 11:12 PM, Bjoolz said:

It's basically the same.

the thing is, yes... but it doesn't tell us *anything* about what is actually causing it... 

[Mark Kaine]

Just now, victor6946 said:

I was running a 1070ti before this with no issues. And had been running that same setup since the release of the 3700X.

Only thing that changed was I swapped the 1070ti with the 7800x on 10/16. And it was running fine with no problems until 11/18.

well ok, so a month.... sure is weird, but entirely possible its a gpu driver issue or something.

 

did you use an older driver before? 

maybe reverting it could fix it. 

[victor6946]

15 minutes ago, Mark Kaine said:

well ok, so a month.... sure is weird, but entirely possible its a gpu driver issue or something.

 

did you use an older driver before? 

maybe reverting it could fix it. 

I was using a different GPU driver before, but I updated the driver right away when CoD MWIII came out. So roughly on 11/10 is when I updated the drivers.

Reverting to an old one is something I haven't tried... 

[victor6946]

I will also add since I got the new GPU, Overwatch has become unplayable. 

I cannot seem to get more than ~65fps in it, it just stutters so bad. And yes, I have triple checked my settings, resolution.. Not to mention when I exit the game, it still remains in memory at least according to Task Manager until I force close it.

 

[Mark Kaine]

36 minutes ago, victor6946 said:

I will also add since I got the new GPU, Overwatch has become unplayable. 

I cannot seem to get more than ~65fps in it, it just stutters so bad. And yes, I have triple checked my settings, resolution.. Not to mention when I exit the game, it still remains in memory at least according to Task Manager until I force close it.

 

easy fix: buy a nvidia gpu.

 

i mean its pretty obvious a gpu/driver issue by now, and anything other than trying to find a driver that works or rma the gpu isn't my expertise,  i use nvidia gpus only (i wonder why lol)

 

[Mark Kaine]

39 minutes ago, victor6946 said:

cannot seem to get more than ~65fps i

yeah i get 330fps almost max settings with ease in ow... something clearly isn't right. 

[victor6946]

11 minutes ago, Mark Kaine said:

easy fix: buy a nvidia gpu.

 

i mean its pretty obvious a gpu/driver issue by now, and anything other than trying to find a driver that works or rma the gpu isn't my expertise,  i use nvidia gpus only (i wonder why lol)

 

Lesson learned… I have a 4070 on hold at micro center. Hopefully it goes well with the swap!

 

The Overwatch issues should’ve been my first red flag It stutters so bad and is unplayable.  

[Bjoolz]

3 hours ago, Mark Kaine said:

the thing is, yes... but it doesn't tell us *anything* about what is actually causing it... 

It does. The CPU cache. I can do a full debug just in case, but Windows is pretty good at it.

 

@victor6946 In the WHEA event, copy the content of the Details, post it in a pastebin (Like pastebin.com) and link to it here. Do a few of them in case there are differences. 

[Mark Kaine]

7 minutes ago, Bjoolz said:

It does

no it doesn't,  it tells you where it happened not what caused it.

 

i know what i said...........

..... and why i said it...

 

 

 

[victor6946]

Well... an update, after I reinstalled Windows (2nd time) I was getting the same random reboots with the 7800. So traded it in for a 4070 and used DDU to uninstall AMD drivers. Haven't ran into the issue at all anymore.

@Bjoolz @Mark Kaine