[SOLVED] Display blacks out randomly ?

[nymikemet]

So a few weeks ago, My computer started to have an issue where the display would go off at random, sometimes it was watching a YouTube video, sometimes it was right on startup, sometimes it was a few hours into a discord call, and every time i'd have to restart the computer.
additionally, I would start up the PC, and after the POST screen, my display would look like this

I tried a different DisplayPort Cable, then I tried switching to the other DP input on the monitor, another DP output on the graphics card and then I tried the HDMI input, all to no avail, though the aforementioned startup screen only happened using DP, but weirdly the POST screen never had issues showing up
I also tried to turn off HDR, I turned off all startup apps, and Updated all my drivers and still nothing

I'm 99% sure it's either the Monitor (Samsung Odyssey G7) or the Graphics (RTX 3080) card (I'll link the parts list below), both were purchased in April of this year, but i'm truly stumped on which it is

Any help is highly appreciated, thank you in advance

PCPartPicker Part List: https://pcpartpicker.com/list/qmpcLP

CPU: Intel Core i9-10850K 3.6 GHz 10-Core Processor ($409.00 @ B&H)
CPU Cooler: Corsair iCUE H100i ELITE CAPELLIX 75 CFM Liquid CPU Cooler ($139.99 @ Newegg)
Motherboard: Gigabyte Z590 UD AC ATX LGA1200 Motherboard ($169.99 @ Amazon)
Memory: G.Skill Trident Z RGB 16 GB (2 x 8 GB) DDR4-3000 CL15 Memory
Storage: Western Digital Blue SN550 1 TB M.2-2280 NVME Solid State Drive ($104.99 @ Lenovo)
Storage: Samsung 860 QVO 2 TB 2.5" Solid State Drive
Storage: Sabrent Rocket 2 TB M.2-2280 NVME Solid State Drive ($299.98 @ Amazon)
Storage: Crucial BX500 2 TB 2.5" Solid State Drive ($194.99 @ Amazon)
Video Card: MSI GeForce RTX 3080 10 GB GAMING X TRIO Video Card ($1413.99 @ Adorama)
Case: Fractal Design Meshify 2 Compact TG Light Tint ATX Mid Tower Case ($119.98 @ Newegg)
Power Supply: Mushkin 1000 W 80+ Gold Certified ATX Power Supply
Operating System: Microsoft Windows 10 Home OEM 64-bit ($108.78 @ Other World Computing)
Monitor: Samsung Odyssey G7 LC32G75TQSNXZA 31.5" 2560x1440 240 Hz Monitor ($699.00 @ Amazon)
Total: $3660.69
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2021-07-20 00:30 EDT-0400

 

[nymikemet]

I'm currently trying a different monitor, so if it still happens, it has to be the GPU, right?
It couldn't be a Windows issue, could it?

 

[LinuxDevice]

I'm currently trying a different monitor, so if it still happens, it has to be the GPU, right?
It couldn't be a Windows issue, could it?

In the past my issues with this were always the power supply of the monitor dying. I've replaced power supplies in several different monitors over the years, and never had the issue be from a video card failure. On the other hand, I have heard of many people having some similar issue due to video card, but my first guess is that it is the monitor. This is especially true if the monitor has a built in supply and does not use an external power supply (capacitors tend to go bad, but not all at once...I've only had this issue with monitors which have their own internal power supply).

 

[nymikemet]

In the past my issues with this were always the power supply of the monitor dying. I've replaced power supplies in several different monitors over the years, and never had the issue be from a video card failure. On the other hand, I have heard of many people having some similar issue due to video card, but my first guess is that it is the monitor. This is especially true if the monitor has a built in supply and does not use an external power supply (capacitors tend to go bad, but not all at once...I've only had this issue with monitors which have their own internal power supply).

Well it just happened twice with a different monitor, so there's that
both the card and monitor were purchased in April, so either way, something crapped out 3 and a half months of honestly not very heavy usage

 

[LinuxDevice]

Well it just happened twice with a different monitor, so there's that
both the card and monitor were purchased in April, so either way, something crapped out 3 and a half months of honestly not very heavy usage

Are the monitors different models? It does sound like either PC PSU or video card if both monitors are different models. If they are the same model, then they might have the same weakness in an internal power supply for the monitor.

 

[nymikemet]

Are the monitors different models? It does sound like either PC PSU or video card if both monitors are different models. If they are the same model, then they might have the same weakness in an internal power supply for the monitor.

The normal monitor is a Samsung G7, which has a external power supply, the one I swapped to test it was a 27 inch Dell Gaming monitor that has a internal power supply.

the power supply on my computer is a no name brand that I got with most of the build (I got Cyberpower PC to build the pc, so I didn’t have a say on it), I plan on replacing it with a Corsair PSU soon, but do you think that it could be the root of the problem?

 

[nymikemet]

I found this in the Event viewer following another display issue

Event 17, WHEA-Logger

A corrected hardware error has occurred.

Component: PCI Express Root Port
Error Source: Advanced Error Reporting (PCI Express)

Primary Bus: Device:Function: 0x0:0x1:0x0
Secondary Bus: Device:Function: 0x0:0x0:0x0
Primary Device Name: PCI\VEN_8086&DEV_1901&SUBSYS_50001458&REV_05
Secondary Device Name:

Followed by this one

Event 4101, Display

Display driver nvlddmkm stopped responding and has successfully recovered.

 

[LinuxDevice]

The normal monitor is a Samsung G7, which has a external power supply, the one I swapped to test it was a 27 inch Dell Gaming monitor that has a internal power supply.

the power supply on my computer is a no name brand that I got with most of the build (I got Cyberpower PC to build the pc, so I didn’t have a say on it), I plan on replacing it with a Corsair PSU soon, but do you think that it could be the root of the problem?

It is possible that the GPU issues are related to the power since unstable power can cause GPU failures. The other error:

A corrected hardware error has occurred.

Component: PCI Express Root Port
Error Source: Advanced Error Reporting (PCI Express)

Is specific to the PCIe bus...the communications with the GPU, and not the GPU itself. Once again it is hard to say, but this could be related to unstable PSU. I hate to say it, but probably changing the PSU as a test is the next step, and it might not solve anything (or it might).

 

[nymikemet]

It is possible that the GPU issues are related to the power since unstable power can cause GPU failures. The other error:



Is specific to the PCIe bus...the communications with the GPU, and not the GPU itself. Once again it is hard to say, but this could be related to unstable PSU. I hate to say it, but probably changing the PSU as a test is the next step, and it might not solve anything (or it might).

I ordered a new PSU that should arrive today. in the event that it is not the PSU, does it look more like damaged hardware, or a communications issue between the Motherboard and GPU?

 

[LinuxDevice]

I ordered a new PSU that should arrive today. in the event that it is not the PSU, does it look more like damaged hardware, or a communications issue between the Motherboard and GPU?

If a PCIe error is not due to power, then it is unusual to have bus errors (such an error doesn't care if the end point is a GPU or anything else, it is the PCIe bus which is in error). In this case though the error was "corrected" (possible only with the AER, Advanced Error Reporting), and in theory should not matter, but in reality (if this repeats), then there is probably a hardware problem. Corrections should prevent a failure, but the need for even one correction is highly suspicious.

There is no guarantee that such a problem is a definitive hardware error of either the motherboard or video card. Such errors can be caused by signal quality, which in turn is a combination of the layout of both the motherboard traces and the PCIe card itself...meaning it could be that signal quality errors would only show up with this combination of motherboard and GPU. Even if neither component is defective this would still be a hardware failure. Yes, motherboard or GPU could be at fault, but so could the combination with an otherwise good motherboard and GPU. No way to tell without a PCIe bus analyzer (a very very expensive test tool).

One thing I would consider when installing the new PSU is to reseat the GPU and any connectors on it. Simple changes to make sure the socket is correctly seated can affect signal quality.

You're basically going to have to install the new PSU and try it out to know.

 

[nymikemet]

If a PCIe error is not due to power, then it is unusual to have bus errors (such an error doesn't care if the end point is a GPU or anything else, it is the PCIe bus which is in error). In this case though the error was "corrected" (possible only with the AER, Advanced Error Reporting), and in theory should not matter, but in reality (if this repeats), then there is probably a hardware problem. Corrections should prevent a failure, but the need for even one correction is highly suspicious.

There is no guarantee that such a problem is a definitive hardware error of either the motherboard or video card. Such errors can be caused by signal quality, which in turn is a combination of the layout of both the motherboard traces and the PCIe card itself...meaning it could be that signal quality errors would only show up with this combination of motherboard and GPU. Even if neither component is defective this would still be a hardware failure. Yes, motherboard or GPU could be at fault, but so could the combination with an otherwise good motherboard and GPU. No way to tell without a PCIe bus analyzer (a very very expensive test tool).

One thing I would consider when installing the new PSU is to reseat the GPU and any connectors on it. Simple changes to make sure the socket is correctly seated can affect signal quality.

You're basically going to have to install the new PSU and try it out to know.

So barring it being a lucky few days, I think I fixed my PC and I believe it was in fact the PSU. The Apevia PSU that came with my PC only had 2 8 pin PCI-E leads, and the 3080 Requires 3, so the builder daisy-chained the 2 connectors from one lead. I replaced it with a Corsair RM850, and used 3 separate 8 Pin PCI-E power connectors when setting it up.

Additionally, I was one version of Windows 10 behind, and updated my computer to version 21H1

Also I had a optional update for Intel - Other hardware - Intel(R) Xeon(R) E3 - 1200/1500 v5/6th Gen Intel(R) Core(TM) PCIe Controller (x16) - 1901

so one of them, or some combination of the 3 seems to have fixed the computer

 

[LinuxDevice]

So barring it being a lucky few days, I think I fixed my PC and I believe it was in fact the PSU. The Apevia PSU that came with my PC only had 2 8 pin PCI-E leads, and the 3080 Requires 3, so the builder daisy-chained the 2 connectors from one lead. I replaced it with a Corsair RM850, and used 3 separate 8 Pin PCI-E power connectors when setting it up.

Additionally, I was one version of Windows 10 behind, and updated my computer to version 21H1

Also I had a optional update for Intel - Other hardware - Intel(R) Xeon(R) E3 - 1200/1500 v5/6th Gen Intel(R) Core(TM) PCIe Controller (x16) - 1901

so one of them, or some combination of the 3 seems to have fixed the computer

It seems likely the cause was the earlier inadequate power supply.