Question Desperate for help, failing PSU? NVME? Virus?

[kevmachine]

Hello, about 5 days ago I woke up and noticed my pc had restarted overnight. I thought that it was weird because I have it on 24/7 and never shut it down but I thought maybe it was an update or something. Later It crashed on me when playing a game, which never happened since I owned it. Since then it has crashed dozens of times and I started trying to trouble shoot it. What I have come to notice is, when launching a program off my 2nd nvme, it tends "not always" to crash, and when I say crash I mean no screen, no bsod just the pc shuts completely off and restarts itsself.

I do some transfering of large files, and I had a 40gb file on this nvme, and when ever I copied and pasted it, because I was trying to move stuff off the drive onto my other nvme, as soon as I clicked paste, the very moment, it would crash. I then thought it must be the hard drive... I experimented, by copying the file and pasting it on the same drive in a new folder. it worked no crash... I tried deleting that 2nd copy, no issues.

I even thought maybe its the other nvme failing and I tested that same thing, no issues, but as soon as I copied and pasted from my 2nd nvme J: to my C: it crashed. I thought maybe it was a corrupt file. i transfered something smaller, a game off my J to my C drive, that worked! now im not so sure what it is... I reset my mb to factory no overclocks, ran memory test came back fine. even reseated my ram and gpu seeing how games would crash sometimes and no luck still crashed. very strange why would my pc crash the moment I clicked paste to transfer files from one drive to another? does that pull power from the psu in a specific way that could indicate a bad psu?

I just launched a game on the J drive, the drive I suspected was bad and it crashed in about 5 seconds, well the pc crashed. I unistalled it and gonna install it on another drive to see if it works. I am at a loss. any help or thoughts would be greatly appreciated.

 

[kevmachine]

I even thought maybe it was the plug in the wall so moved it to another outlet and still would crash. the file also was a single file 40gb that i tried to transfer from J to C drive when it would crash as soon as I clicked paste, I deleted the file because I thought it was corrupted. Nothing is overheating, I looked at all temps and nothing seem's hot. drivers are all up to date as far as I know. windows up to date. This might be stupid but could the nvme have shifted a bit? could i remove it and re insert it lol.

 

[--SID--]

Define 'crash'. What's happening? Blue screen, black screen, restart, shutdown or something else?

List full system specs includes brand and model of the PSU.

 

[kevmachine]

Crashing as in the pc shuts off abruptly, as if I unplugged it from the wall. sometimes it restarts on its own 7/10, sometimes I have to press the power button. I did some testing today, took everything apart and put it all together this includes the cpu, gpu, psu ram ect. I thought maybe the nvme my windows was on was bad so I reinstalled fresh windows on another nvme, still same thing. Updated all drivers on my motherboards website. I just don't know what is going on.

cpu is ryzen 9 3900x
aorus x570 ultra
64gb ddr4 3600 ram xmp enabled i tried disabling xmp still crashed
power supply
Cooler Master MPY-6501-ACAAG-US MWE Gold 650 Watt 80 Plus Gold Certified Power Supply

 

[kevmachine]

I can play games on it sometimes for awhile 1h+ but then sometimes it will crash before the game even opens. I am just at such a loss. I am at the point where I want to try a new power cable from my psu and sata cables. Could it be a bad fan or something in the case that could do this? when I look at event viewer I get the same Kernal-Power code 41 (63) (70368744177664),(2) its the same critical log everytime it does this. If I had a spare psu id try that but don't have one at the moment. Maybe its my gpu going bad? gtx 960 2gb.

It loves to crash when I go to steam and try to install a game, I don't know if its during the allocation process most likely it is, but it tends to crash when ever I do that. Took psu cleaned it out a bit, removed my 4 internal sata hard drives, removed some fan y cables I had, still crashed, removed all my case fans unplugged them still crashed. now I am on experiment god... 17? lol Using a different cable that plugs into the wall to psu. so far so good, been around 15 minutes thou so not much time. will update as I go.

Crashed again.

 

[kevmachine]

So I was in the middle of a Youtube video and paused it to go run some earrands. came back 3 hours later, computer was still on perfectly as I left it. finished watching the video and as soon as I tried to open a program steam in this case it shut down.

 

[kevmachine]

I don't know if this helps but I ran cinibench R23 and it ran perfectly got a score of 18,238 with my 3900x temps got to 76. as soon as the test ended I closed it and double clicked on steam and boom shut down! Now I am gonna resinstall steam and see if that helps, the thing is it wasn't just steam that was causing it to happen thou it seems steam was causing this issue a lot for me. The tricky thing is I did a fresh install of windows on a different nvme and it still shut down on me, pretty sure i might of had steam installed on there also but can't say for sure as I was just downloading stuff to test.

 

[--SID--]

Motherboard BIOS up to date?
Test with another reliable PSU?

 

[Vic 40]

Can you maybe do this to see all hardware in the pc and also show what bios you run on?

download hwinfo,
install and open it=click run,
close the top window which is the system summary,
in the main window at the left top click "save report",
at the bottom of the next window check "Summary for Clipboard",
after that you'll see what's in the pc,
copy by clicking "copy to clipboard" and rightclick+paste in your next respons

 

[InvalidError]

Random shutdowns and restarts on a PC that used to work perfectly fine are most commonly caused by a flaky PSU.

The last time I had one of those on one of my PCs was simply due to a clogged PSU air filter.

(Well, I had a couple of unexpected shutdowns due to my stupid CyberPower UPS deep-frying its battery pack while still pretending that everything was normal.)

 

[kevmachine]

yes updated motherboard, its bios F35d. Don't have another PSU handy but that will be the next test I think ill have to do.

Computer: GIGABYTE X570 AORUS ULTRA
CPU: AMD Ryzen 9 3900X (Matisse, MTS-B0)
3800 MHz (38.00x100.0) @ 3767 MHz (37.75x99.8)
Motherboard: GIGABYTE X570 AORUS ULTRA
BIOS: F35d, 10/13/2021
Chipset: AMD X570 (Bixby)
Memory: 16384 MBytes @ 1330 MHz, 20-19-19-43
- 16384 MB PC28800 DDR4 SDRAM - G.Skill F4-3600C18-16GVK
Graphics: GIGABYTE GeForce GTX 960 WindForce 2X OC
NVIDIA GeForce GTX 960, 2048 MB GDDR5 SDRAM
Drive: Seagate FireCuda 520 SSD ZP2000GM30002, 1953.5 GB, NVMe
Drive: Samsung SSD 970 EVO Plus 500GB, 488.4 GB, NVMe
Drive: Seagate FireCuda 520 SSD ZP2000GM30002, 1953.5 GB, NVMe
Drive: MSFT XVDD, Disk drive
Drive: MSFT XVDD, Disk drive
Drive: MSFT XVDD, Disk drive
Drive: MSFT XVDD, Disk drive
Sound: NVIDIA GM206 - High Definition Audio Controller
Sound: AMD Family 17h/19h - HD Audio Controller
Network: Intel Wi-Fi 6 AX200 160MHz
Network: Intel I211AT Copper (Pearsonville) Network Adapter
OS: Microsoft Windows 10 Home (x64) Build 19044.1415 (21H2)

I think its a flaky psu but its just odd how I can run a cinibench r23 benchmark fine without it crashing and as soon as I try to open a program or 9/10 it crashes. but then again psu work in strange ways when they get bad I am guessing.

 

[kevmachine]

Bringing the gpu and psu to a friends house to test on his rig, he has no crashing issues at all so might help narrow it down. I can also run cinibench r23 fine and heaven benchmark 2009 no crashes.

 

[Vic 40]

Might want to look through some of these threads
https://www.google.com/search?client=opera&q=MSFT+XVDD&sourceid=opera&ie=UTF-8&oe=UTF-8
and see that you delete these,

Drive: MSFT XVDD, Disk drive
Drive: MSFT XVDD, Disk drive
Drive: MSFT XVDD, Disk drive
Drive: MSFT XVDD, Disk drive

Saw btw that your ram is not running at rated specs, did you enable XMP?

 

[kevmachine]

Thanks I will give it a read, I tried it on xmp and with xmp off just for testing, normally I have it on, but it was crashing either way. I have done alot of testing today. so this is what I kinda am starting to see. I normally have 3 nvme's on my mobo. 1 samsung 970 500gb and 2x 2tb seagate firecuda 520's. if I pull the 2 520s out and just leave the 1 nvme in it won't crash, I tried for almost an hour opening and closing various apps and programs all at the same time and downloading watching youtube playing games, nothing no crashes, but as soon as I installed one of the seagate nvmes it crashed, I tried the other one and it crashed, both of them can't be bad no way! I tested all the nvme slots with the single samsung 970 nvme just to see if it was possibly a bad slot or two, but all the slots worked without crashing after testing for 15-20 minutes each. but when I have the 970 and a single firecuda 520 in there it crashed on me. the 520's run at gen 4.0 speed and I think the 970 is gen 3? but not 100% sure. I don't know now if this is a bad motherboard maybe something is messing up with how gen 4.0 speeds are handled. or the psu again can't handel what ever rail feeds the nvme's. any input? all i know is with the single 970 with my os on it I was unable to get a crash after an hour of trying, and within 5 minutes of plugging in another seagate nvme it crashed.

 

[Vic 40]

Might want to look through some of these threads
https://www.google.com/search?client=opera&q=MSFT+XVDD&sourceid=opera&ie=UTF-8&oe=UTF-8
and see that you delete these,

What i meant is delete whatever is creating these.

 

[kevmachine]

They are all gone, I had to reformat everything last night because the pc crashed and corrupted windows to the point I couldn't fix it even after trying multiple recovery options. I honestly have it pinned down to if I have more then 1 nvme installed it will crash. Right now I have just my OS nvme and it hasn't crashed in over 14 hours. I ran disk check on my other nvme's and it came back fine no issues, these nvmes are only 6 months old and I don't think they have much wear. Yesterday I felt one while it was testing and it actually felt pretty hot to the touch which surprised me didn't know they got that hot, I put a fan on it but after 10 - 15 minutes the pc crashed with that nvme also installed. At this point I figure its something to do with something related to that. Pc runs fine with only my OS nvme installed. But if I install my other nvmes even just 1 it will crash randomly when doing pretty much anything. Just got my 500gb OS nvme right now which isn't great because I lost everything and have sooo much to download.

 

[kevmachine]

I might have to contact either my motherboard manufacturer or seagate on the drives. I am still not 100% sure if its the psu thou, perhaps when I have more then 1 nvme installed there isn't enought stable power to function properly. I am gonna lend my psu to a friend soon to test on his system.

 

[Vic 40]

Maybe set fixed pcie speeds for the drives in the bios if possible, make the C drive which should be in the top slot x4 and the others set those to x3. See if it helps.

 

[mdd1963]

Most NVME drives use about 3-5 watts peak, so, available power for a 2nd NVME drive from a particular PSU would seem to be an odd issue.

If you have one good NVME drive, then test each M.2 NVME slot's functionality with each NVME drive. (All of this assumes pf course there is no chaos/confusion caused the OS by having a cloned drive running with the original at the same time, which was exactly what caused a putrid mess just a day or two back for someone else..

 

[kevmachine]

Maybe set fixed pcie speeds for the drives in the bios if possible, make the C drive which should be in the top slot x4 and the others set those to x3. See if it helps.

Ok I will try and figure out how to do that, im sure its an easy google search thanks, ill update if i see any results.

 

[kevmachine]

Most NVME drives use about 3-5 watts peak, so, available power for a 2nd NVME drive from a particular PSU would seem to be an odd issue.

If you have one good NVME drive, then test each M.2 NVME slot's functionality with each NVME drive. (All of this assumes pf course there is no chaos/confusion caused the OS by having a cloned drive running with the original at the same time, which was exactly what caused a putrid mess just a day or two back for someone else..

I know its soo odd to me also, I already testing my OS nvme alone in each of my 3 slots on my mobo with 0 issues, each test was only about 20 minutes thou so it can't be 100% thou each test was me opening various apps or programs all at once/downloading/gaming/watching youtube all in a hasty manner which would normally 100% crash the pc with another nvme installed. OS nvme is in top slot at the moment. I have 1 of my 2tb firecuda 520's in atm and it crashed once in the last few hours thou I wasn't downloading to the 2tb drive in that time until about 15 minutes ago. I downloaded xbox app off windows store and started downloading forza 5 again lost everything last night due to corrupted boot and withtin 2 or 3 minutes of downloading it crashed. rebooted left everything as it was and its downloading now for over 30 minutes no crash yet even thou nothing changed. I bet if i open up some programs or start downloading on my OS nvme it would crash thou I won't do that at this time lol. in all this time I have not managed to get the pc to crash with just the OS nvme installed which to me must mean something.

I have also notice that sometimes with these crashes my bios will say its been reset back to factory, that in itself is alarming, why would my bios reset from a pc crashing? isn't that something that happens when you take the battery out and cycle all the power out.

 

[kevmachine]

went to bed last night left the computer on all night with 0 sleep or screen shut off, woke up and it was just as I left it, started doing stuff and boom shut down. when I booted it up there was a checksum error had to turn of xmp for my ram to get it to boot. just another day lol.

 

[DimkaTsv]

Are you sure your RAM is completely stable?
You might as well check it with memtest86. Because random crashes can be from either power problem, or some severe RAM problems. Even though RAM error will usually give BSOD before turning off.

started doing stuff and boom shut down. when I booted it up there was a checksum error had to turn of xmp for my ram to get it to boot

Because if you need to turn off XMP to even boot PC... It may be linked to RAM issues
Also reset to factory might mean that memory fails training and BIOS resets to default to make it at least POST so you can enter BIOS.
If i remember correctly 3600 (MCLK=1800) is on verge of FCLK stability for Zen 2. (Hell even with my 5600x i cannot pass FCLK=1800, even though my I/O seems trash-y), so you might just not be able to run XMP stable.

It still might be PSU problem, though. I just recommend to check RAM

 

[Vic 40]

Wonder with all of that if the motherboard is stable, might want to consider an rma.

 

[kevmachine]

I will run that ram test now. I hope not, that motherboard is only 6 months old and cost an arm and a leg sadly you may be correct.

- Ran Memtest for 1 hour 48 minutes it finished 13/13 tests with a mark of pass 0 errors. I know after doing some breif reading that you should run it overnight for multiple hours which I will do tonight but at least i can see the ram is at least not severely damaged. at this point im lost. I can only imagine its a hardware issue.