Random Black Screen Issue

Before you say it, Yes i understand my hardware is old LOL... I got a lot of hate on nvidia's forums for talking about the hardware, But i'm using a GTX 550 Ti, I get random black screen as if my driver has crashed, But windows doesn't say it crashed.. It happens daily 20-30 times a day, I tried reinstalling drivers, underclocking the graphics card, And no results.. I tested with a different card a GTX 1060 and it does the same thing.. Just not as often Any idea?.. i'd like to get back to gaming again but i just cant.. And it even does it siting idle or even at the lock screen of windows when i first turn on..

Thanks..

 

I would try another monitor first - just to rule that out. If you don't have one, most TVs in recent years support computer input.

Then, since everything inside the computer depends on good clean, stable power, I recommend swapping in a known good power supply to see if the problem continues.

 

I use dual monitors, Both go off and the one is brand new, The power supply is only a couple months old as well.. I will try a different power supply and see

 

Alright, Different power supply didn't fix the issue.. Any other ideas?

 

You have not expressly stated this but the system as it was set up was running with no issues previously? So this is something recent?

When you tested with the GTX 1060 (or GTX550ti) did you test with one monitor connected or both? If I were you I would try with just one monitor connected and see if same. Also how are you connecting the vga to the monitor(s), DVI, HDMI, other.....

Re the GTX550 Ti, does the card have one or 2 power connectors?

What psus have you tried? specs please.

Lastly, when you say random black screens, what do you mean exactly ie does the monitor go black monetarily and resume or do you have to reboot as monitor remains blank?

 

The system was perfectly fine for a long time, About a week or two ago it started doing this, I use DVI and a DVI to HDMI for the one monitor, I've tried on just one monitor and it does the same issue, The 550 Ti has one power connector.. My main psu is Rosewill Hive-750s (750 watt), The other psu is EVGA 1000 GQ (1000 watt) And the display when it goes black, It acts like a driver crash, It goes black and comes back. But doesn't say the driver has crashed.. I've changed drivers old and new drivers but nothing changes.. the PC is usable but when it black screen's it gets very frustrating.. Games crash, Even sometimes my web browser crashes... But i don't know what could be causing this issue

 

I can see that this type of issue would be very frustrating.

For clarity until the crashes started about 2 weeks ago, you did not manually update any drivers whatsoever. BTW, what Nvidia driver version were you, are you using? Also for clarity, the monitor drop outs happen not just in games but at anytime.

What o/s are you using? and do you know whether there was an MS update about 2 weeks ago?

From a HW point of view, I think that we can rule out the PSU and the VGA since you tried with the GTX1060.

If you had cpu issues, then the pc would just freeze and only a hard reboot would get the system going again.

How much ram do you have? how many sticks? Have you tried running memtest to see whether the ram is fine?

Also prior to running memtest what I would do is just try to run your rig with only 1 ram stick

Do you have any addon cards on your mobo ie sound card, ethernet card or using the on-board ones? What mobo are you running?

 

The monitor drops out no matter what i'm doing, Can be in a game or just idle on the desktop and it goes black, I'm using windows 10 (64bit)
Haven't had any updates since, Was using the latest nvidia driver i manually installed from nvidia, Months prior to the issue.. Then i reinstalled to see if that was the issue.. (obviously wasn't)

Driver version is 391.35

Motherboard is ASRock z390 pro4,

Memory 2x 16gb

Memtest doesn't show any problems..

Using onboard Ethernet and sound.

 

Run WhoCrashed Free Home Edition and upload the log.
https://www.majorgeeks.com/files/details/whocrashed_free_home_edition.html

 

Eldon's suggestion is definitely worth taking up.

At this stage we are no closer to knowing whether the matter at hand relates to issues with hardware, software or perhaps some heat related problem ie northbridge overheating perhaps as a result of paste degradation? (especially if the monitor blackouts have coincided with a heatwave where you live).

Other than Eldon's suggestion, I would suggest that you boot your rig with a Linux Live CD/USB and see whether the monitor blackouts still manifest themselves in the same way. In the first instance, this is what I would do if I were in your shoes.

If, after having booted with Linux (Linux Mint a good choice) that issue does or does not manifest itself, whichever is the case, then at least you will have narrowed down the possibilities.

Unfortunately, no easy route that I can see and it can be a pig to get to the bottom of such a problem.

It reminds me when I had a faulty cpu years ago during which I systematically replaced the vga, psu, ram and motherboard (same as original one) and it was only when I replaced the cpu that I resolved this issue. This was confirmed when I rebuilt the pc using all the original components but with the new cpu in situ.

 

My thoughts exactly.

Speccy will show the temperatures.
https://www.majorgeeks.com/files/details/speccy.html

 

Running the whocrashed program it tells me there's no valid crash dumps?

"Crash Dump Analysis
Crash dumps are enabled on your computer.

Crash dump directories:
C:\WINDOWS
C:\WINDOWS\Minidump

No valid crash dumps have been found on your computer
"

 

My temperatures at idle hover around 40-50c, Under 100% CPU it hits 80c, Which still isn't enough for it to kick off. Its not shutting off the display is just going black for about 4-5 seconds or so

 

Not sure if posting links is allowed, if it isn't i'm sorry, But this is a video of what it is doing..

 

If there is an overheating issue causing the black screens it may not necessarily be associated with the cpu. I presume that you are running an Intel cpu as those temps (especially at load) are way too high for it to be an AMD cpu.

Why don't you take the side of the pc case off and see whether this makes a difference.

Other than my suggestion advocating for you to try running a Linux Live CD/USB, I have nothing else to suggest since I have already stated previously, we are no nearer to isolating whether this is a hardware, software or possibly overheating issue.

If you were to try the Linux Live option this would, imo, give some pointers otherwise you are just looking for a needle in a haystack.....

Good Luck

 

Alright.. Ill test with linux and see what it does.. If linux doesn't have the issue, Could it be a software problem maybe?

 

I think that this would be an almost inescapable conclusion

BTW, you have provide scant info about your system, eg mobo, cpu, ram, etc

You have also not specified whether you are running everything at stock or if any / all components are overclocked.

See how you go with Linux but depending on the result any user on this forum trying to help you will require complete and detailed info, unless some users on this forum are hiding mind reading skills....!!!

 

Linux after about 4 hours hasn't had the issue, Going back to windows and no issue yet?... And the cpu has never been overclocked, Gpu i overclock to squeeze out the best performance But obviously it wasn't the gpu if two gpu's had the issue

 

Check the GPU temperature.
https://www.majorgeeks.com/files/details/gpu_z.html

 

Now it's beginning to smell like a heat issue as Eldon's keen 'wolf like nose' already suspects also !!

Other than posting GPU temps, have you tried running your rig with the side cover off.

Try to run your rig under load ie run 3DMark (in loop mode) and see if the issue manifests itself again. Monitor the GPU temp and also try to touch the Northbridge heatsink to see if its warm / hot

 

ps. I know that you also tried with the GTX1060 and got the same issue but do this and let us know what you find.

Don't forget to touch the Northbridge heatsink

 

I'm not convinced.
Not knowing what software you're using for the overclocking, and with similar graphics cards, it's not impossible for the program to apply the settings to the different card.

Why not?
Diagnosing hardware problems on a forum is difficult at best. Trying to do it without knowing the hardware specs is near impossible.
Speccy's Summary might help us.
https://www.majorgeeks.com/files/details/speccy.html

 

Guessing he is using MSI Afterburner.....

I would agree that all overclocks should be disabled for problem solving