Jump to content

PC Random Complete Power Off Then Restart

 Share

I am quite lost with this issue and am hoping for some help/suggestions.

Long story short: I replaced my GPU not too long ago. I started getting blue screens constantly. I used DDU to uninstall all display drivers, went through every fix I could find.
Luckily, I found it had nothing to do with my GPU. While updating my Display Drivers after the DDU uninstall, I guess I updated my Intel ME drivers. I found this was causing the blue screening and a simple bios flash fixed this issue.

Thought that was the end of my troubles, but ever since, I RANDOMLY get complete PC loss of power and then it restarts. My motherboard lights even shut off. I kept thinking my old power strip I've been using was maybe the cause, it might still be, but everything else plugged into the strip doesn't turn off.

I dug really deep, I started using windows verifier to check for corrupted drivers. I was able to find one corrupted driver and uninstall it. (The driver found was from iObit Smart Defrag: "SmartDefrag.sys"). But after that first driver uninstall, the following restarts, I kept getting a driver error blue screen, but no provided driver.

So.. with the blue screens no longer providing me with the possible driver causing the problem, I disabled Automatically Restart from sysdm.cpl, as well, configured Small Memory dump (256 kb). After a few days, I got a random loss of power and restart again, and bluescreenviewer was able to pinpoint it to an RGB lighting software. I went down a whole rabbit hole to find a lot of RGB control software will conflict and cause these crashes. So I found a provided file from G.Skill to completely wipe ALL RGB software and their registry entries.(The only driver I found was ene.sys, which is what I found was a common driver to cause crashes from multiple RGB controler software)

This seemed to fix it. However, yet again, I started getting the power loss and restarts, but no MiniDump files have been created to check bluescreenviewer for the cause since,

This is an issue that is purely random. I can be sitting with nothing but discord open, and my PC loses power and restarts. Or it has happened while streaming. I have consistently checked my temps. Temps are completely fine. There is no consistent cause I at least can find.. I can game for 6 hours with no problem. I can also game for 30 minutes and get this issue. I can be sitting on windows desktop, and randomly get this issue. It's nothing related to CPU/GPU usage or temps.

With this all said, I can only assume it's a something like a corrupt file, or corrupt driver. I can't pinpoint anything.

Lastly, I never had this problem before the new GPU, Bios Flash or other driver updates. I was running 2x 1080ti's before, and now a single 3080ti. I have more than enough PSU power for this PC. I don't think it's PSU related. I've had no power surges, even if I did, I have a great surge protector I'm plugged into. I think it's a purely corrupt file(s) or driver(s) that are accessed randomly, causing this. But if that's the case. I can't explain why I don't get a blue screen when this occurs..

PC Specs:
Motherboard: Asus Z390-A
CPU: i9-9900K
GPU: EVGA 3080ti FTW3
PSU: Corsair RM750 750 W 80+ Gold Certified Fully Modular ATX
RAM: 32 GB, 4x8, Tident Z RGB
Storage: 980 Pro EVO 2 TB, SanDisk 1TB drive

Link to comment
Share on other sites

Link to post
Share on other sites

The 3080ti is a 3090 basically and it consumes more power than 2 1080tis. Couple that with the spiking behavior of the 3000 series and a rm750 (especially pre 2018) is not going to handle that.

 

I know this because there are a lot of people with even current rm750s that have issues with the 3080 due to them having a bit more powerhungry cpu.

 

Im very very certain this is just the regular 3080ti has a useage spike, spikes to 600w real quick, rm750 thinks a short happens and it shuts off.

Link to comment
Share on other sites

Link to post
Share on other sites

6 minutes ago, jaslion said:

The 3080ti is a 3090 basically and it consumes more power than 2 1080tis. Couple that with the spiking behavior of the 3000 series and a rm750 (especially pre 2018) is not going to handle that.

 

I know this because there are a lot of people with even current rm750s that have issues with the 3080 due to them having a bit more powerhungry cpu.

 

Im very very certain this is just the regular 3080ti has a useage spike, spikes to 600w real quick, rm750 thinks a short happens and it shuts off.

I would agree, but if I can run games for 6 hours with no issues and massive GPU power draw, I don't think the GPU/Power usage is the issue at all. Especially when considering majority of the crashes have occurred while I'm simply sitting with discord and/or youtube open. 

Link to comment
Share on other sites

Link to post
Share on other sites

15 minutes ago, ComicallyBad said:

I would agree, but if I can run games for 6 hours with no issues and massive GPU power draw, I don't think the GPU/Power usage is the issue at all. Especially when considering majority of the crashes have occurred while I'm simply sitting with discord and/or youtube open. 

See thats the weird part too.

 

Other users have reported this exact issue. When idle the pc shuts off and this does have to do with a sudden loadspike.

 

Do you still have a 1080 to test the systen with for stability?

Link to comment
Share on other sites

Link to post
Share on other sites

4 hours ago, ComicallyBad said:

I am quite lost with this issue and am hoping for some help/suggestions.

Long story short: I replaced my GPU not too long ago. I started getting blue screens constantly. I used DDU to uninstall all display drivers, went through every fix I could find.
Luckily, I found it had nothing to do with my GPU. While updating my Display Drivers after the DDU uninstall, I guess I updated my Intel ME drivers. I found this was causing the blue screening and a simple bios flash fixed this issue.

Thought that was the end of my troubles, but ever since, I RANDOMLY get complete PC loss of power and then it restarts. My motherboard lights even shut off. I kept thinking my old power strip I've been using was maybe the cause, it might still be, but everything else plugged into the strip doesn't turn off.

I dug really deep, I started using windows verifier to check for corrupted drivers. I was able to find one corrupted driver and uninstall it. (The driver found was from iObit Smart Defrag: "SmartDefrag.sys"). But after that first driver uninstall, the following restarts, I kept getting a driver error blue screen, but no provided driver.

So.. with the blue screens no longer providing me with the possible driver causing the problem, I disabled Automatically Restart from sysdm.cpl, as well, configured Small Memory dump (256 kb). After a few days, I got a random loss of power and restart again, and bluescreenviewer was able to pinpoint it to an RGB lighting software. I went down a whole rabbit hole to find a lot of RGB control software will conflict and cause these crashes. So I found a provided file from G.Skill to completely wipe ALL RGB software and their registry entries.(The only driver I found was ene.sys, which is what I found was a common driver to cause crashes from multiple RGB controler software)

This seemed to fix it. However, yet again, I started getting the power loss and restarts, but no MiniDump files have been created to check bluescreenviewer for the cause since,

This is an issue that is purely random. I can be sitting with nothing but discord open, and my PC loses power and restarts. Or it has happened while streaming. I have consistently checked my temps. Temps are completely fine. There is no consistent cause I at least can find.. I can game for 6 hours with no problem. I can also game for 30 minutes and get this issue. I can be sitting on windows desktop, and randomly get this issue. It's nothing related to CPU/GPU usage or temps.

With this all said, I can only assume it's a something like a corrupt file, or corrupt driver. I can't pinpoint anything.

Lastly, I never had this problem before the new GPU, Bios Flash or other driver updates. I was running 2x 1080ti's before, and now a single 3080ti. I have more than enough PSU power for this PC. I don't think it's PSU related. I've had no power surges, even if I did, I have a great surge protector I'm plugged into. I think it's a purely corrupt file(s) or driver(s) that are accessed randomly, causing this. But if that's the case. I can't explain why I don't get a blue screen when this occurs..

PC Specs:
Motherboard: Asus Z390-A
CPU: i9-9900K
GPU: EVGA 3080ti FTW3
PSU: Corsair RM750 750 W 80+ Gold Certified Fully Modular ATX
RAM: 32 GB, 4x8, Tident Z RGB
Storage: 980 Pro EVO 2 TB, SanDisk 1TB drive

Check the motherboard. 

Link to comment
Share on other sites

Link to post
Share on other sites

14 hours ago, jaslion said:

See thats the weird part too.

 

Other users have reported this exact issue. When idle the pc shuts off and this does have to do with a sudden loadspike.

 

Do you still have a 1080 to test the systen with for stability?

 

10 hours ago, STELDR said:

Check the motherboard. 

I have my 1080ti I can test still, however, I have no other spare parts unfortunately to test. (I can't really afford to buy more spare parts to test either).

I will try to swap back to my 1080ti, (though it's an annoying process... using DDU, wiping drivers, re-installing lol). I can give it a try, however, with that; considering how random this issue occurs, I would have to swap GPU's and test for weeks if not a full month. If the issue persists with both cards, potentially I could assume that my PSU is no good maybe from a previous power surge (even with my good surge protector), if not, hopefully, I find I Just need a higher wattage PSU, and this one isn't "damaged."

Furthermore; I would try simply running heavy benchmarks with this 3080ti, and then my old 1080ti. But again, I haven't been able to reproduce the PC power loss by any method. I've ran many benchmarks on this 3080ti and didn't get this issue. I ran the same benchmarks with my 1080ti, before upgrading, with no issues either. It is only these very bizarre timings I get this issue. Meaning; I can't simply use a benchmark on both to get a conclusion...
Of course this adds a level of frustration, as if I'm trying to narrow down to a lack of power to the 3080ti, you'd think a benchmark would be the best go-to for checking this.

I myself have put a lot of emphasis on it maybe being hardware. But what are the chances it's not hardware and maybe a software (driver) issue? Due to the pure randomness, not being able to reproduce consistently, and verifier causing blue screens showing there is a corrupt driver (although, not providing which one), I want to also consider the chance it's a software issue... BUT, if it is software, why no blue screen? That's what's difficult for me the most right now. Whether a corrupt driver, or lack of power, windows should blue screen rather than complete power cut off, correct?

As I don't blue screen, I want to blame hardware.. due to the randomness, and verifier proving existence of corrupt drivers, I want to blame software.

Final notes and main points:

- I can't reproduce this error by any method. It's purely random.

- I get power cut off most times with simply discord and/or chrome open, it's happened a few times during recording gameplay.
- This hadn't happened before upgrading. After upgrading, I also can't say I encountered this issue right away.

- After updating multiple things such as my GPU drivers, my Samsung 980 Pro EVO firmware, Intel ME drivers. This is when I started having issues. First, I had the bluescreening issue, due to the Intel ME driver requiring a BIOS flash. After flashing my BIOS, fixing the bluescreening, this is around the time I started getting or at least noticing the power cut offs.

Link to comment
Share on other sites

Link to post
Share on other sites

I flashed my BIOS for a second time, just in case. As well, instead of using the provided driver downloads from my motherboards website (Asus), I went to Intel's site for the latest Intel Management Engine driver. The Intel ME driver from Intel's website was very different from the one Asus provides.. The ZIP had completely different files and folders. When I installed it, it seemed to be a newer version, as it completely wiped the old ME driver and replaced it.

I can't say for sure this has fixed it, but I have not had a cut off of power in a whole week. I had gone 3-5 days before without a cut off, but never a whole week.

As I said, the Intel ME driver caused the blue screening I spent 2 days straight debugging.
I admit, it would be strange/confusing, if flashing the BIOS again, or installing Intel's provided ME driver fixed this issue. However, if the Intel ME driver, alone, could cause non-stop blue screening for 2 days straight, there's a chance this power cut off was related as well.

I will update if I ever have a power cut off again. For now, it looks like I've resolved this issue.

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share


×