[SOLVED] Black Screening Under Load, Multiple Restarts Needed to Boot

[tsndx]

Hello,
Over the last month I have begun to experience increasingly worse problems on my PC. I have tried to figure out how to fix this but to no avail. Hopefully somebody on here who knows more than I do will be able to help.

The problem being faced is that the computer will often times restart while under load. By this I mean that the screen will go black, the motherboard manufacturer's logo screen will appear (where you have the option to boot into the bios), and I will then be brought to the log in screen. It gets worst however.

Sometimes, the computer will freeze and then black screen permanently. This happens both under load, and sometimes a minute or two after closing an intensive program that was running for hours without problem. It will then take multiple restarts to get it to the bios or log in screen. Earlier today it took over 10 restarts to get anything to display on screen. When I press the power button, the fans spin and system lights are powered, but there is no output to the screen. My keyboard has a back light and that is not powered when it fails to boot. I have even tried using a cable directly from my motherboard to my monitor to bypass the GPU and there is still no display on the monitor. I have read online that this can be caused by an HDD being unable to spin up and load the OS, but my OS is loaded on an SSD so this should not be the case here. After many restarts, it eventually boots normally.

All of this, and my searches online has led me to believe it to be a PSU problem. This computer has run fine for about 5 years now. Near the beginning of the year I switched my gpu to a 1660ti, and my psu is 650w, which is more than enough. I have used this GPU for many months without issue. All overclocks have been reverted to stock speeds to rule out that as the problem. There are no thermal issues with the computer. I just ran a benchmark for multiple minutes to monitor the thermals to confirm this and my gpu reached a max of 68C and my cpu reached a max of 58C before I black screened and froze.
I would just buy a new power supply but I am now worried that the issue may stem from my GPU. I have recently done a driver update and I am now experiencing blue screens under load with the error nvlddmkm.sys . Looking online this appears to be an nvidia GPU problem. I don't think this problem is PSU related, though I am hoping to get other people's feedback. I find it a bit odd that my psu would be potentially failing less than 5 years out. I did not get a cheap PSU.

Any replies are greatly appreciated. Thanks for taking the time to read. If any further information is needed I can provide anything needed.

 

[Vic 40]

Enlighten us with more knowledge about parts used in the pc. For psu that means also maker+model.

Still got the old gpu to test with?

Did a complete reinstall of Nvidia drivers? Do use DDU to get rid of the old drivers first.

The bios of the motherboard is up to date? Drivers as well?

 

[tsndx]

Hi, thanks so much for replying. I should have included more system information in my original post you're right.

The psu is an EVGA supernova GS 650w.
My motherboard is an msi z97 gaming 5, and my cpu is a 4690k. I have 16gb ddr3 ram.
I do have the old gpu (r9 390) to test with. I have reverted the drivers this morning to see if it got rid of the blue screen. I am going to do more testing with my current gpu before reverting to the 390. In truth I probably should have tried this before posting. It can be very difficult to troubleshoot this as sometimes I can go an entire day of the system being under load with no issues. Other days I cannot be under load for more than 3 minutes without facing a frozen system and black screen requiring multiple restarts to even get to the bios.

I don't think the black screen / freezing aspect is a bios version or driver version problem. I'm sure you know more than I do, but I had been on the same drivers for many months and this only started 1 month ago. I had very recently updated (after the black screens regularly occured) the gpu driver in attempts to see if that fixed it, thought it seemed to have introduced the new blue screen problem. I have since reverted to the previous version, and no I had not used DDU in fact I had not known about it.

The bios is not up to date though I see nothing in the updates that would indicate a fix for any sort of related problem. Maybe you think it is worth a try?
I know I am not really giving much to work off of here, I was just hoping someone would have been able to point to any one component and say that is the likely culprit. I really do appreciate your reply.

Edit: I had just done some more testing and got a freeze after closing out of a program. I just let it sit for many minutes and it eventually restored the display. I checked event viewer to try to see if I could gather any information and I got multiple issues related to nvlddmkm which is an nvidia gpu related issue. I then took your advice and ran DDU and have installed the newest driver for my gpu. The event viewer errors lead me to think it is solely a gpu problem, however I cant imagine my gpu preventing my from booting / requiring multiple restarts to boot successfully.

 

[Vic 40]

You have an onboard gpu with which you can test if the gpu might be a problem.

Would myself update the bios probably using that onboard gpu to be sure the gpu can't be in the way. Take the gtx 1660 ti out and use motherboards video output for an image.

Ram is running rated specs? If updating bios ( and i strongly suggest) make sure that XMP is enabled after doing that.