F5F Stay Refreshed Hardware Desktop Screen goes dark unexpectedly, fans spin at maximum. Needs to restart using the power button.

Screen goes dark unexpectedly, fans spin at maximum. Needs to restart using the power button.

Screen goes dark unexpectedly, fans spin at maximum. Needs to restart using the power button.

O
OPMage_6115
Member
67
12-07-2024, 05:41 AM
#1
I've experienced a problem recently, but I'm unsure of its origin. My system setup includes a 5950x4TB NVMe WD Black boot drive and another 4TB NVMe drive from Gigabyte X570. I've tried various configurations: 32GB, 64GB DDR2, two sticks, four sticks. The EVGA 1300G2 PS MSI 4090 PBO supports XMP with both enabled and disabled. The issue manifests as random black screens requiring a power restart, accompanied by full fan speeds. This occurs unpredictably—sometimes while using Chrome, other times when idle or during program execution. It happens under stress and not under normal conditions. I've checked logs, hardware monitors, GPU stress tests, RAM diagnostics, and CPU temperatures; all tests came back normal. I've also reworked the CPU and monitored temperatures without any problems. Despite these efforts, I still face blackouts under different scenarios. I'm at a loss and would appreciate any guidance or testing suggestions. I prefer not to rush into replacements without understanding the root cause.
O
OPMage_6115
12-07-2024, 05:41 AM #1

I've experienced a problem recently, but I'm unsure of its origin. My system setup includes a 5950x4TB NVMe WD Black boot drive and another 4TB NVMe drive from Gigabyte X570. I've tried various configurations: 32GB, 64GB DDR2, two sticks, four sticks. The EVGA 1300G2 PS MSI 4090 PBO supports XMP with both enabled and disabled. The issue manifests as random black screens requiring a power restart, accompanied by full fan speeds. This occurs unpredictably—sometimes while using Chrome, other times when idle or during program execution. It happens under stress and not under normal conditions. I've checked logs, hardware monitors, GPU stress tests, RAM diagnostics, and CPU temperatures; all tests came back normal. I've also reworked the CPU and monitored temperatures without any problems. Despite these efforts, I still face blackouts under different scenarios. I'm at a loss and would appreciate any guidance or testing suggestions. I prefer not to rush into replacements without understanding the root cause.

R
Rosen05
Member
152
12-07-2024, 05:41 AM
#2
Looking for solutions? I'm considering DDU and Chkdisk. Besides the Kernel Power Event in Event Viewer, do you see any other details?
R
Rosen05
12-07-2024, 05:41 AM #2

Looking for solutions? I'm considering DDU and Chkdisk. Besides the Kernel Power Event in Event Viewer, do you see any other details?

S
Sparkle_Mage
Member
206
12-07-2024, 05:41 AM
#3
Thanks for the feedback. This appears to be a clean installation, so drivers shouldn't be a concern. I reviewed the logs and found only Kernel Power Event entries for the black screens, with no other patterns. I’ll run Chkdisk to check further. The problem likely relates to power, possibly a failing PSU that’s nearing the end of its life. It hasn’t been affected by surges or unusual conditions, and it happens in multiple locations, suggesting an internal issue rather than wiring problems.
S
Sparkle_Mage
12-07-2024, 05:41 AM #3

Thanks for the feedback. This appears to be a clean installation, so drivers shouldn't be a concern. I reviewed the logs and found only Kernel Power Event entries for the black screens, with no other patterns. I’ll run Chkdisk to check further. The problem likely relates to power, possibly a failing PSU that’s nearing the end of its life. It hasn’t been affected by surges or unusual conditions, and it happens in multiple locations, suggesting an internal issue rather than wiring problems.

P
PeskyBoy
Junior Member
8
12-07-2024, 05:41 AM
#4
Windows typically sets up some standard drivers that might lead to problems. This suggests the power supply isn't the main concern. Most power-related issues result in a complete shutdown, which feels more like a GPU failure rather than a power problem. It's possible a peripheral is malfunctioning. A 4090/5950X model is quite demanding on the system. If you can afford a replacement PSU, it wouldn't hurt. Before making any changes, check voltages with HWMonitor to confirm. Also, consider reseating the 4090 and its cables.
P
PeskyBoy
12-07-2024, 05:41 AM #4

Windows typically sets up some standard drivers that might lead to problems. This suggests the power supply isn't the main concern. Most power-related issues result in a complete shutdown, which feels more like a GPU failure rather than a power problem. It's possible a peripheral is malfunctioning. A 4090/5950X model is quite demanding on the system. If you can afford a replacement PSU, it wouldn't hurt. Before making any changes, check voltages with HWMonitor to confirm. Also, consider reseating the 4090 and its cables.

Z
zer0_porcento
Member
55
12-07-2024, 05:41 AM
#5
I performed DDU and updated the NVIDIA drivers after reinstalling Windows. The MSI Suprim Liquid X card is compact with an AIO, which is good for portability. I noticed it didn’t move much after installation. I only changed the system twice to test power supply stability in my home. Since then, it hasn’t moved much. I suspect the problem began when I added the extra 32GB DDR4 RAM. The computer became more prone to black screenings when I had four sticks installed. I removed two older ones and replaced them today, using XMP settings. This configuration worked fine for about a year without issues. I also had the same setup with a 3090 chipset for another year, mostly stable. My concern now is the two newer RAM sticks. It seems everything was fine before they arrived, but issues started around the time of installation. All other components remained stable for at least a year prior. It might be the GPU, especially since I ran it at 600W for a short while without any problems. That was an overclock, so I’m leaning toward something else being the cause. The PSU and thermal performance were perfect. I’ll keep monitoring and provide updates. Thanks for your advice and support.
Z
zer0_porcento
12-07-2024, 05:41 AM #5

I performed DDU and updated the NVIDIA drivers after reinstalling Windows. The MSI Suprim Liquid X card is compact with an AIO, which is good for portability. I noticed it didn’t move much after installation. I only changed the system twice to test power supply stability in my home. Since then, it hasn’t moved much. I suspect the problem began when I added the extra 32GB DDR4 RAM. The computer became more prone to black screenings when I had four sticks installed. I removed two older ones and replaced them today, using XMP settings. This configuration worked fine for about a year without issues. I also had the same setup with a 3090 chipset for another year, mostly stable. My concern now is the two newer RAM sticks. It seems everything was fine before they arrived, but issues started around the time of installation. All other components remained stable for at least a year prior. It might be the GPU, especially since I ran it at 600W for a short while without any problems. That was an overclock, so I’m leaning toward something else being the cause. The PSU and thermal performance were perfect. I’ll keep monitoring and provide updates. Thanks for your advice and support.

H
Henry2999
Junior Member
10
12-07-2024, 05:41 AM
#6
Great effort. I don’t recall doing this before, but a reseat wouldn’t hurt. Perhaps I’d be more cautious with it myself. It might just have been a bad luck call. Still, if it works out, it’ll hold steady.
H
Henry2999
12-07-2024, 05:41 AM #6

Great effort. I don’t recall doing this before, but a reseat wouldn’t hurt. Perhaps I’d be more cautious with it myself. It might just have been a bad luck call. Still, if it works out, it’ll hold steady.

M
mrscupcake564
Junior Member
23
12-07-2024, 05:41 AM
#7
Your RAM is faulty. Get a completely different one, just two sticks, and confirm it matches the motherboard specifications.
M
mrscupcake564
12-07-2024, 05:41 AM #7

Your RAM is faulty. Get a completely different one, just two sticks, and confirm it matches the motherboard specifications.

X
xXGoku997Xx
Junior Member
41
12-07-2024, 05:41 AM
#8
This post is about the 4090 issue I encountered. I was considering giving up and sticking to my laptop, since a crashing computer would be risky during client calls. The problem kept happening again and again. I checked the RAM thoroughly, tried different configurations, but nothing resolved it. In a final attempt, I ran Cinebench 2024 and Furmark 2 to see if crashes would be consistent. That worked out well—Cinebench performed perfectly for over an hour without any problems. Furmark, however, sometimes crashed the GPU within a few minutes after startup, though not every time. I updated the firmware and even tested both gaming and silent power settings, but nothing changed the outcome. I bought an RX580 on FB for $40 and it has run smoothly for over 12 hours without any issues. I also tested Furmark in the background for about six hours straight, with no interruptions. The GPU temperature stayed around 73°C, and it felt like the card was idle. MSI has already approved a replacement for the 4090. My plan is to order another 4090 and sell it while prices are still favorable. I don’t play games much, so I wasted most of the money on that card. I’ll keep the RX580 for running my monitor at 4K144 and meeting my needs. If I decide to upgrade, I can get a 7900XT for half the cost but expect around 80-90% of the performance. I’ll wait until I confirm there are no more crashes before making a decision. Thanks for all the advice!
X
xXGoku997Xx
12-07-2024, 05:41 AM #8

This post is about the 4090 issue I encountered. I was considering giving up and sticking to my laptop, since a crashing computer would be risky during client calls. The problem kept happening again and again. I checked the RAM thoroughly, tried different configurations, but nothing resolved it. In a final attempt, I ran Cinebench 2024 and Furmark 2 to see if crashes would be consistent. That worked out well—Cinebench performed perfectly for over an hour without any problems. Furmark, however, sometimes crashed the GPU within a few minutes after startup, though not every time. I updated the firmware and even tested both gaming and silent power settings, but nothing changed the outcome. I bought an RX580 on FB for $40 and it has run smoothly for over 12 hours without any issues. I also tested Furmark in the background for about six hours straight, with no interruptions. The GPU temperature stayed around 73°C, and it felt like the card was idle. MSI has already approved a replacement for the 4090. My plan is to order another 4090 and sell it while prices are still favorable. I don’t play games much, so I wasted most of the money on that card. I’ll keep the RX580 for running my monitor at 4K144 and meeting my needs. If I decide to upgrade, I can get a 7900XT for half the cost but expect around 80-90% of the performance. I’ll wait until I confirm there are no more crashes before making a decision. Thanks for all the advice!