Question Random screen flickering and usbsequent full shutdown while playing certain games

Oct 15, 2024
3
0
10
To start off I'll list my specs:
Ryzen 9 7950x3d
MSI MAG Tomahawk x670e
G.skill flare ddr5-6000 ram, 32gb
EVGA FTW3 RTX 3070 ti
Crucial p3 plus
Deepcool 850 watt PSU

So I've been having this issue for a few days now. Basically, in some really specific games, most notably assetto corsa, kerbal space program, beamng, and rounds (weirdly enough). I'll be playing for an amount of time, sometimes a few minutes, one time even up to 3 or 4 hours, where the display will suddenly flicker all sorts of weird colors and then suddenly completely shut down. No bsod or anything, just sudden power off and reboot. And sometimes (like right now) I can't get it to boot at all. This seems to happen with varying severity, sometimes it just simply restarts, sometimes it requires me to switch to IGPU, boot, then switch to gpu, sometimes it won't even restart and I need to flick the switch on the PSU off and on, and sometimes it just won't boot, as said earlier.

I can't access the PC currently to get the exact error codes, but here's roughly what I remember:

-event ID 14, nvddlmkm
-Kernel, all sorts of different shutdown codes
-some COM thing

And here's what I've tried so far and the results:
[When I say "fixed it" I mean that I didn't experience the error for some time afterwards]

-Clearing CMOS: fixed it for a few hours, then started right back up
-System restore to a few days before the issue started: fixed it for about a day and a half
-Updating GPU drivers: Absolutely nothing
-Disabling/removing all my RGB software: Absolutely nothing

If I can get this piece of <Mod Edit> to boot again I'm going to try adjusting TDR, downgrading to an old driver, reseating gpu and testing my ram sticks.

Edit: While reseating mu GPU I noticed my sag bracket was set a little high, causing it to press against the top of the PCIe slot a little bit. Not sure if this is what was causing the issue but I guess we'll find out because I WAS able to get it to boot again.
 
Last edited by a moderator:
Welcome to the forums, newcomer!

Deepcool 850 watt PSU
Deepcool is the brand of the unit while 850W is the advertised wattage of the unit, what is the model of the unit? If the PSU's been in service for a while, include it's age.

Ryzen 9 7950x3d
How are you cooling the processor?

MSI MAG Tomahawk x670e
BIOS version for your motherboard?

Updating GPU drivers: Absolutely nothing
Using DDU?
 
Welcome to the forums, newcomer!

Deepcool 850 watt PSU
Deepcool is the brand of the unit while 850W is the advertised wattage of the unit, what is the model of the unit? If the PSU's been in service for a while, include it's age.

Ryzen 9 7950x3d
How are you cooling the processor?

MSI MAG Tomahawk x670e
BIOS version for your motherboard?

Updating GPU drivers: Absolutely nothing
Using DDU?
The PSU is good, I have a tester and also tried it in my old AM4 pc.

The processor has a Lian-Li ALC360 aio, and all of my benchmarks and stress tests never showed any overheating issues.

The motherboard is on the latest bios version from MSI, it's been updated for a few weeks now but the problem only started 4 or 5 days ago.

I tried basically every option, using outdated drivers, updating them, using ddu, downloading old drivers again, and then trying newer ones again, still the issue persists.

I'm going to see I'd I can replicate the issue through a benchmark or something that way I can isolate the problem if it's either a ram stick or the gpu itself.
 
The PSU is good, I have a tester and also tried it in my old AM4 pc.
just because a power supply "works" in a low load scenario doesn't mean that it can handle high load usage like gaming.

this is why knowing the model of this particular unit is necessary.
DeepCool is known for producing garbage products and you may be surprised to find that the majority of issues like yours are caused by inadequately manufactured power supplies.
 
To start off I'll list my specs:
Ryzen 9 7950x3d
MSI MAG Tomahawk x670e
G.skill flare ddr5-6000 ram, 32gb
EVGA FTW3 RTX 3070 ti
Crucial p3 plus
Deepcool 850 watt PSU

So I've been having this issue for a few days now. Basically, in some really specific games, most notably assetto corsa, kerbal space program, beamng, and rounds (weirdly enough). I'll be playing for an amount of time, sometimes a few minutes, one time even up to 3 or 4 hours, where the display will suddenly flicker all sorts of weird colors and then suddenly completely shut down. No bsod or anything, just sudden power off and reboot. And sometimes (like right now) I can't get it to boot at all. This seems to happen with varying severity, sometimes it just simply restarts, sometimes it requires me to switch to IGPU, boot, then switch to gpu, sometimes it won't even restart and I need to flick the switch on the PSU off and on, and sometimes it just won't boot, as said earlier.

I can't access the PC currently to get the exact error codes, but here's roughly what I remember:

-event ID 14, nvddlmkm
-Kernel, all sorts of different shutdown codes
-some COM thing

And here's what I've tried so far and the results:
[When I say "fixed it" I mean that I didn't experience the error for some time afterwards]

-Clearing CMOS: fixed it for a few hours, then started right back up
-System restore to a few days before the issue started: fixed it for about a day and a half
-Updating GPU drivers: Absolutely nothing
-Disabling/removing all my RGB software: Absolutely nothing

If I can get this piece of <Mod Edit> to boot again I'm going to try adjusting TDR, downgrading to an old driver, reseating gpu and testing my ram sticks.

Edit: While reseating mu GPU I noticed my sag bracket was set a little high, causing it to press against the top of the PCIe slot a little bit. Not sure if this is what was causing the issue but I guess we'll find out because I WAS able to get it to boot again.
Did the problem only show up after installing some software?
 
Last edited by a moderator:
Did the problem only show up after installing some software?
No, not that I can recall. Which is leading me to assume it's probably a dead/faulty part.

Which had better not be the GPU because I just got an rma for a month ago when it <Mod Edit> itself after only 3 weeks of having it installed.
 
Last edited by a moderator: