Question Computer randomly restarts with demanding tasks

[Wissaib]

Hello everybody

About a week ago I bought a new desktop PC. The problem is that when using the computer for moderately demanding tasks such as gaming, the screens go to black without warning and the PC reboots. In the event viewer there is no indication of any error. There is only a message that states that the computer wasn't turned off properly. The problem occurs about 1-2 hours into the games.

It's important to say that all the settings in the BIOS (which is up to date) are set to default. None of the components are overclocked.

The kind of tasks that cause the reboot of the compute put an average load on the CPU of around 40%, with some peaks in between.

I understand that computers can reboot if the CPU temperature becomes too high, however, it stays at an average of 65 degrees celsius when measured by using the Core Temp application. It never peaks at more than 75 degrees.

All the drivers are up to date, including the GPU ones.

I took the computer to the shop where I built it. However, they were unable to replicate the problem and according to them everything seemed to be fine. However, I can't really trust the competence of those technicians because of some stories that I have heard. The problem is that for the moment I can't really take the computer to some other place beause it would void my warranty.

A few minutes before I decided to write this post I noticed that the USB 3.1 Gen 2 ports of my motherboard (the green ones) seem to be dead. Yesterday they were working fine. All the ports are turned on in the BIOS. Also, I have an audio connector in the green jack of the motherboard. From time to time the computer asks me if I connected a new device there. The connector is firmly connected. Could this be some problem consisting on the system not receiving enough power?

All my system is connected to an APC Pro 1500 UPS, which has a maximum output of 865 W. Also it's important to say that I tested the RAM with the Windows Memory Diagnostic tool included in Windows. It showed no errors.

The specs are the following:

Intel® CPU Core i9-9900K 5.00GHz (1151-v2)
Asus M/B Intel TUF Z370 Plus Gaming (1151-v2)
Asus Video NVIDIA Geforce GTX1660Ti O6GB Dual Fan
Crucial DDR4 16GB 3000MHz Ballistix Sport LT (Red) (2 sticks for a total of 32 GB)
WD Unidad SSD 1TB Sata3 2.5" Blue 3D NAND
Seasonic Power Supply 1000W Focus Plus 80Plus Gold - Full Modular
Corsair WaterCooling Hydro H100X 2018

Also I have Windows 10 Professional 64-bit (which is up to date), as well as three other conventional hard drives of 2, 4 and 5 TB.

Also, is there any tool that could log the BIOS activity? For example, something that could detect if some component is suddenly losing power or something like this?

Any ideas?
Thanks

 

[bignastyid]

Given the weak VRM setup on that motherboard and the hefty power requirements of the 9900k, the VRMs overheating would be my top suspect. Grab a monitoring app like HWinfo64 and see if you can monitor the VRM temps(not sure if that board has a VRM temp sensor), if it does run a cpu stress test while moniotring the VRM temp and see how hot they get.

 

[Wissaib]

Given the weak VRM setup on that motherboard and the hefty power requirements of the 9900k, the VRMs overheating would be my top suspect. Grab a monitoring app like HWinfo64 and see if you can monitor the VRM temps(not sure if that board has a VRM temp sensor), if it does run a cpu stress test while moniotring the VRM temp and see how hot they get.

Thank you for your answer. I am trying both HWMonitor and HWinfo64. Neither of those show an explicit VRM temperature. However, they both show several temperature parameters. Could any of those represent the VRM? I attach a screenshot. The left one is HWMonitor and the right one is HWinfo64.

Maybe I could have an option to swap my motherboard for an MSI® M/B Intel Z390 Tomahawk (1151-v2). Would this help if I confirm that it's effectively a VRM overheating issue, considering that I don't pretend to overclock anything?

 

[bignastyid]

Looks like that boards lacks VRM temp sensors. That MSI board does have a better VRM setup that should handle the 9900k, atleast at stock settings.

 

[Wissaib]

Looks like that boards lacks VRM temp sensors. That MSI board does have a better VRM setup that should handle the 9900k, atleast at stock settings.

Thank you for your reply. Looks like I will go ahead and swap the motherboard. In any case, I don't pretend to overclock my hardware, even if I get the MSI motherboard