Question Windows shutting down 45 seconds after booting

brianbotkiller

Distinguished
May 26, 2011
14
0
18,510
Hi all,

I have a brand new, clean build, using an Asrock z390 Pro 4 mobo, with a 9700KF CPU -- this is my second time building the same build with the same hardware.

The system just never stays booted -- it boots into Windows 10, installed clean from a clean ISO from MS, and then just shuts off -- no BSOD, no error messages -- just shuts off. Then, it reboots, and does the same thing, until Windows repair comtes up, at which point, it doesn't shut itself off. It only dies if it's booted into Windows. This ISO is Windows 10 1903, the latest build from MS via media creator tool, installed via USB onto bare metal.

I've tried different cases, entirely new hardware -- always happens.

It DOESN'T happen in Safe Mode, and if the system sits in BIOS, Windows installer, or etc, it doesn't shut off.

So, this has to a driver issue in Windows, but, I haven't even been able to get to the point that I'm installing drivers -- the system just dies too quickly. So, it's using all native stuff in Win 10, and dying.

I've also disabled networking in the BIOS, updated the BIOS, and checked for erroneous stuff that could cause issues -- nothing -- same issue.

Yes, I've tested every component, and they're all fine. I've had two iterations of this same build, with different hardware, and nothing.

I'm out of ideas. Does anyone have any ideas about how I could troubleshoot this?
 

brianbotkiller

Distinguished
May 26, 2011
14
0
18,510
Sure.

Mobo: Asrock Z390 Pro 4
CPU: Intel 9700KF
GPU: Radeon HD5450
m.2: WD Blue 500GB installed in m.2 middle slot
RAM: Crucial Ballistix 2400 DDR 4
Cooler: Cooler master 212 EVO
Windows: Build 1903 of Home, from MS Media creation tool, sent to ISO and burnt to USB using Rufus
PSU: Corsair CX750M

To be clear -- this is the second iteration of all of this hardware -- I've RMA'd all of it, trying to suss this thing out -- so this is the second run of the same hardware, with the same issue happening.

I've also tried a clean boot, disabling all services, and, there's no third-party drivers on this thing, as it doesn't even touch the net.
 
It's not a driver issue necessarily - drivers are needed to run hardware - safe mode doesn't load anything but the most basic drivers - so safe mode it's loading whatever hardware is causing the issue.

How old is the 5450? I don't know how you RMA'd that as it is a VERY old card, and probably the first thing I would look at as far as parts to replace.
 

brianbotkiller

Distinguished
May 26, 2011
14
0
18,510
Nah, it's not the HD5450. I've built literally thousands of systems with the HD5450, and this behavior is the same with built-in GPU on the mobo, and a totally different discrete GPU.

And at this point, yes, it is either a driver issue, or something inherently wrong in Windows. I understand what Safe Mode is -- that's the reason that safe mode exists, to run a limited driverset. My issue and confusion is how a totally clean ISO of Windows 10 could be so unstable. The point is that this is a totally clean ISO, no third party drivers, and the system simply blips out 45 seconds after booting, and Windows error logs aren't even throwing anything useful.
 

brianbotkiller

Distinguished
May 26, 2011
14
0
18,510
I'm not trying to be an ass to you, I have thousands of builds that have never had this issue, as well, so I'm simply looking for help, since this is the first time in 25 years I've seen something this inane thanks
 
OK - let me put it to you this way (even though it 's probably not what you want to hear):
1 - Windows 10 has more OS-level hardware limitations than any OS before it, so even though a 9-year-old video card may work perfectly fine in Windows 8, Windows 10 says no.
More info here: https://community.amd.com/thread/185720
2 - So in a way, yes it's a driver issue, in that as soon as Windows determines what the card is, the OS says "no."
 
In your short 45 or so seconds running in standard mode, install a temp monitoring software, like CPUID Hardware Monitor and watch the temp on all the cores. What temp are they getting up to?

Also, what case are you using? I didn't see it listed, but may have missed it.
 

brianbotkiller

Distinguished
May 26, 2011
14
0
18,510
Yeah, not temps, either. 40c on average. Cooling is fine.

Case is a Fractal Define C -- and yes, I've tested with a different power switch to ensure that the case's switch isn't randomly shutting the system off.

Thanks,
 

brianbotkiller

Distinguished
May 26, 2011
14
0
18,510
A 9700KF on a 212 EVO running at 40? That's pretty amazing. What speed was the CPU at?


I dunno offhand, man, it would just be stock in this case, I haven't had time to tweak this thing. I've really been focused on the rest of this thing. I know that the CPU wasn't overheating based on the BIOS and a heat sensor gun. The issue isn't CPU heat so I'm not focusing my attention on that.

And yeah, the 212 EVO is more than capable, I water cool these CPUs from time to time but it all depends on the build.
 
Jan 22, 2020
1
0
10
I know it's a couple months old but did you ever figure it out? I have the same issue, about 15 seconds for myself and then Windows reboots on an MSI motherboard, ISO 1909 downloaded from the Microsoft website. Stays up in BIOS, initial lock screen and safe mode. I suppose it's a driver somewhere.