PC Crashes on High-Demand Games but Runs Smoothly on League of Legends

[RDaniel]

Hi everyone,

I'm facing some unusual issues with my PC setup and could use your insights. Here's the configuration:

Motherboard: MSI MEG X570 Godlike

CPU: AMD Ryzen 9 5950X

RAM: G.Skill Trident Z 16GB x2

GPU: RTX 3090 FE

Storage: Samsung NVMe 970 Evo Plus

PSU: EVGA T2 1600W

Initially, the system wouldn't boot due to a CPU error indicated on the motherboard display. After swapping in new RAM, the system started working fine for everyday tasks and games like League of Legends. However, it crashes when launching demanding games such as Cyberpunk, Helldiver 2, Elden Ring, or running the 3D Mark Time Spy stress test.

I've tried several troubleshooting steps including reinstalling Windows, updating the BIOS, switching back to the original RAM (which reintroduced the CPU error), and updating all drivers. I also tested the GPU, NVMe, and PSU in another system, where they performed without any issues.

As this is an AM4 system, I'm hesitant to invest further in a new motherboard and CPU without understanding the root cause of the problem. Could anyone offer some advice or potential solutions?

 

[Hellowpplz]

2 minutes ago, RDaniel said:

Hi everyone,

I'm facing some unusual issues with my PC setup and could use your insights. Here's the configuration:

Motherboard: MSI MEG X570 Godlike

CPU: AMD Ryzen 9 5950X

RAM: G.Skill Trident Z 16GB x2

GPU: RTX 3090 FE

Storage: Samsung NVMe 970 Evo Plus

PSU: EVGA T2 1600W

Initially, the system wouldn't boot due to a CPU error indicated on the motherboard display. After swapping in new RAM, the system started working fine for everyday tasks and games like League of Legends. However, it crashes when launching demanding games such as Cyberpunk, Helldiver 2, Elden Ring, or running the 3D Mark Time Spy stress test.

I've tried several troubleshooting steps including reinstalling Windows, updating the BIOS, switching back to the original RAM (which reintroduced the CPU error), and updating all drivers. I also tested the GPU, NVMe, and PSU in another system, where they performed without any issues.

As this is an AM4 system, I'm hesitant to invest further in a new motherboard and CPU without understanding the root cause of the problem. Could anyone offer some advice or potential solutions?

 

try running memtest64

[Blasty Blosty]

2 minutes ago, RDaniel said:

I also tested the GPU, NVMe, and PSU in another system, where they performed without any issues.

Have you tested other parts in YOUR system?

 

For example, do you have a friend with a PC where you could swap GPU or PSU to see if you still face the issues? 

[RDaniel]

2 minutes ago, Blasty Blosty said:

Have you tested other parts in YOUR system?

 

For example, do you have a friend with a PC where you could swap GPU or PSU to see if you still face the issues? 

I replaced the MB and CPU with X670e and Ryzen 7 7800X 3D, it worked perfectly. Only the MB and CPU that I got no other way to test it.

[Blasty Blosty]

1 minute ago, RDaniel said:

I replaced the MB and CPU with X670e and Ryzen 7 7800X 3D, it worked perfectly. Only the MB and CPU that I got no other way to test it.

One of them are faulty, or both then.

 

I don't think there really is a way to test them, aside from using known working components with them. You could take them to a computer shop for a diagnosis?

[RDaniel]

9 minutes ago, Hellowpplz said:

try running memtest64

Thanks, will try that later

[RDaniel]

4 minutes ago, Blasty Blosty said:

One of them are faulty, or both then.

 

I don't think there really is a way to test them, aside from using known working components with them. You could take them to a computer shop for a diagnosis?

I did bring it to local PC repair store, and they just ran a furmark test and CPU stress test then told me they fixed the issue it(tried to charged me $80)

[TylerD321]

Sounds like a cooling issue to me. Download hwinfo64 and look at cpu temps using “sensors only” page.

[Agall]

1 hour ago, RDaniel said:

I did bring it to local PC repair store, and they just ran a furmark test and CPU stress test then told me they fixed the issue it(tried to charged me $80)

I'd suspect some sort of C-state/P-state issue. I could see something like League never triggering the highest P-state, so operating at a lower P-state or C-state the whole time.

 

The ways of testing that would be changing Windows Power Plan for the CPU and/or disabling C-states in the UEFI, and/or TDP limiting to 105W with PBO's ECO mode. For the GPU, there's PCIe C-state controls that change PCIe version, where it'll negotiate as low as PCIe 1.1 to save power. Those are UEFI features that can be disabled then validated with GPU-Z by comparing it while idle versus with a game in the foreground.

[RDaniel]

17 hours ago, TylerD321 said:

Sounds like a cooling issue to me. Download hwinfo64 and look at cpu temps using “sensors only” page.

Just Checked, the temperature of the CPU did not go above 65 c during furmark 4K + CPU stress test at the same time. Issue only appeared when I ran 3D mark and graphics intensive game.

[RDaniel]

Just spotted a new issue, when I try to backup my personal files with external hard disk. The mouse movement became extremely laggy, does that relate to my current issue?