• Happy holidays, folks! Thanks to each and every one of you for being part of the Tom's Hardware community!

Question PC shuts off during gaming but RAM lights stay on, only reboots after flipping PSU switch off/on ?

Status
Not open for further replies.

relentless238

Reputable
Feb 21, 2018
8
1
4,515
Hey guys, so I have an interesting problem that just started happening the past few weeks. It has happened three times as of today.

My PC will shut off during gaming (playing escape from tarkov), but the RAM and IO lights will stay on. Almost like its in sleep mode. Both monitors will lose signal. All the fans will stop spinning, but the RAM stays lit up (similar to how it stays lit up in sleep mode). The kicker is that in this coma like state, i cannot turn the PC back on without flipping the PSU power switch on and off (the front panel power button is not responsive unless i flip the psu switch). I only have to flip it for like 1 second and then I can use the power button and boot the PC normally. No bluescreens or issues rebooting. I checked the minidump, but nothing is generated since it does not blue screen. I also checked event viewer and did not see anything out of the ordinary besides it saying that there was a critical error due to unexpected shutdown. There is also no error lights on the motherboard. Only lights that are on is the RAM and the ones near the IO interface.

It happens super randomly too, all three times it has happened during Escape from Tarkov, in game, not in the menus. It happened around 2 weeks ago, exactly as stated above, and after rebooting the PC the game ran fine up until now. I probably have put another 20 hours of gametime in since it happened two weeks ago. Today, it happened twice, both within 2 hours of each other. Once again, the computer backscreens, RAM and IO lights stay on, but cannot reset via power switch. I must flip the PSU on and off to reboot. It reboots clean and then I get back into the game normally.

The PC is connected into the wall outlet via surge protector. I have other electronics plugged in as well, most notably an Atom amp for my DT770 Pros 250 ohm. I just plugged the computer directly into the socket above the surge protector. Haven't played since but not sure if it will help. It is weird to me that it happened two weeks ago and I've played the game a ton since then and now it happened twice today, nearly two weeks later.

Some other important notes is that I have played MW2019 on this computer and have not had black screens. I also just moved this whole computer into a new case (O11 Dynamic EVO), while also installing my 5800X3D. No issues in previous case or previous CPU (3800X).

I am using a Lian Li GPU anti sag bracket. It is mounted via (2) standoff screws on the motherboard. I installed as the manual said. Not sure if this is causing a ground issue?

Here are my specs:

Ryzen 5800X3D (bios was updated for this)
EVGA FTW3 3080
ASUS TUF X570 WIFI +
Corsair RM 750W Gold
32GB Trident Z Neo 3600 mHZ ( DOCP turned on in BIOS)
9 upHere case fans connected to a hub thats powered via SATA
Samsung EVO 970 m.2 1TB (game is on this)
intel m.2 1TB SSD (OS on this, don't know the exact name of the part)
Noctua NHD15S cooler w/ two fans
 
Update -
Ran MemTest86 for 6 hours. No errors after 4 passes.
Also checked all three Kernel 41 logs, all have no data / no bug check codes (0x0).
Only difference is SleepTransitionToOn, which is 3 and then 1 and then 0. I’m assuming that’s from me trying to wake the computer. The last time it happened I didn’t even bother to try.
 
To test ram, download memtest86. Remove all but 1 stick and run it for a few minutes. If no errors remove that one and go to the next stick until you’ve tried all of them. If one is bad it should start throwing errors fairly quickly.
Hey Ohio. I can’t test with all RAM in place? I ran mem test last night and had no issues after 4 passes (please see my first reply). I ask because the ram is under my large air cooler and the air cooler would have to be removed to unseat the RAM.

thanks.
 
if it just shutdown and pc cant be powered on without disconnecting powersupply from wall (on/off switch on psu), then psu protection is kickin in
rtx 3080 does spike alot, since youre sayin it started after moving PC to another case, then id recheck cables
your PSU does have five modular connectors for EPS (cpu) and pci-e (gpu)
try using three separate pcie cables if you havent already
 
  • Like
Reactions: Zerk2012
You could leave ram in, but in my experience it's easier to do one stick at a time. However, the poster above has a good point to double check the power supply as well. Since you stated you moved everything, is it possible there is a grounding issue in your case like maybe the board is touching the case where it should not?
 
if it just shutdown and pc cant be powered on without disconnecting powersupply from wall (on/off switch on psu), then psu protection is kickin in
rtx 3080 does spike alot, since youre sayin it started after moving PC to another case, then id recheck cables
your PSU does have five modular connectors for EPS (cpu) and pci-e (gpu)
try using three separate pcie cables if you havent already

i will check all my cables. Bought another 1000 w PSU to test with. And I am already using 3 separate PCIE cables. Thank you for the reply. I’ve been reading about 3080 power spikes and my card apparently has a 420W max TDP but can spike higher. So I’m wondering if this 750W isn’t enough.
 
Last edited:
You could leave ram in, but in my experience it's easier to do one stick at a time. However, the poster above has a good point to double check the power supply as well. Since you stated you moved everything, is it possible there is a grounding issue in your case like maybe the board is touching the case where it should not?
I will check my cables and case. Yeah so I ran the mem test with both sticks and had no issues. I’ll try other solutions before removing my cooler to test each ram one at a time. I would have to remove the cooler multiple times to do that.
 
Try that other PSU (what is it by the way?). I'd be surprised if that's not the issue..I believe the RM series are not very well reviewed.
Hey Dudio. Thanks for replying. I bought a Corsair RMx 1000 W Gold. I read the RMx have better (Japanese) capacitors than the Chinese RM ones. However I seen both PSUs are “A-tier”. I’m going to be RMAing my 750W Corsair PSU if it’s the defective part. They come with a 10 year warranty. Then I’ll just have a backup! I just installed it. Will boot up later and game for a while and see what happens. It’s weird though because it happened on 6/29 and then didn’t happen again until yesterday. And it happened twice in one hour yesterday. So who knows if it’ll happen today when gaming or in two weeks.
 
Last edited:
  • Like
Reactions: dudio
Hey Dudio. Thanks for replying. I bought a Corsair RMx 1000 W Gold. I read the RMx have better (Japanese) capacitors than the Chinese RM ones. However I seen both PSUs are “A-tier”. I’m going to be RMAing my 750W Corsair PSU if it’s the defective part. They come with a 10 year warranty. Then I’ll just have a backup! I just installed it. Will boot up later and game for a while and see what happens. It’s weird though because it happened on 6/29 and then didn’t happen again until yesterday. And it happened twice in one hour yesterday. So who knows if it’ll happen today when gaming or in two weeks.

Hey did you ever fix this issue? Did the power supply change do the trick? Had this same issue about a year and a half ago; fast-forward to today and it started happening again with no change of parts. I currently use a Corsair RM650 PSU so I thought there could be a correlation.
 
Hey did you ever fix this issue? Did the power supply change do the trick? Had this same issue about a year and a half ago; fast-forward to today and it started happening again with no change of parts. I currently use a Corsair RM650 PSU so I thought there could be a correlation.
Yes, it was fixed by replacing the Corsair RM750W with a Corsair RMx 1000W. No issues since the swap!
 
Status
Not open for further replies.