Graphic Cards Screen randomly goes black

V

Vaibz

Recruit
So I have a gigabyte 1660Super(out of warranty) I'm facing this issue where the screen randomly goes black and the fans run at full RPM. I can do normal browsing and stuff but whenever I run a game or any benchmark, this happens again. Ive changed thermal paste, my HDMI&PCIe everything is all good. My VRAM is also stable while playing games at about 45%
 
Last edited:
Vaibz said:
So I have a gigabyte 1660Super(out of warranty) I'm facing this issue where the screen randomly goes black and the fans run at full RPM. I can do normal browsing and stuff but whenever I run a game or any benchmark, this happens again. Ive changed thermal paste, my HDMI&PCIe everything is all good. My VRAM is also stable while playing games at about 45%
Click to expand...
Try reinstalling nvidia drivers. Use DDU first.
 
nRiTeCh said:
Temps ?
How old is the gpu ?
Tried low intensity games ?
Click to expand...
Temperatures about 60°c

I bought the GPU in 2019 November end (my dumbass didnt register it for extended warranty)

Yes I did try low intensity games. This is happening to me even in CSGO at ~medium settings
 
1. First use DDU and remove all the old drivers and install a latest one.
2. Monitor the gpu temps and power draw while running gpu benchmarks (before they crash) . Also do same for cpu. This will help determine what causing the issue temp or power draw .
 
Vaibz said:
Ive changed thermal paste
Click to expand...
Which thermal paste did you use?
And
1) Did you change any component that important that might lead to this?
2) from how much time are you experiencing this?

Questions might sound silly but better to rule out basic things first before going advanced.
Also do what @SkyWreak said in above post.
 
Last edited:
SkyWreak said:
1. First use DDU and remove all the old drivers and install a latest one.
2. Monitor the gpu temps and power draw while running gpu benchmarks (before they crash) . Also do same for cpu. This will help determine what causing the issue temp or power draw .
Click to expand...
I did do fresh installation using DDU. Temperatures before crash were about ~60-62°c

DEVRAJ BARIA said:
Which thermal paste did you use?
And
1) Did you change any component that important that might lead to this?
2) from how much time are you experiencing this?

Questions might sound silly but better to rule out basic things first before going advanced.
Also do what @SkyWreak said in above post
Click to expand...

I used Arctic MX-4 carbon based thermal paste. I'm experiencing this from about a month now.

Also i did not change any component, I tried using another GPU the same one (1660s but from MSI) it worked fine, so im pretty sure something is wrong with my GPU only
 
Vaibz said:
I used Arctic MX-4 carbon based thermal paste. I'm experiencing this from about a month now.

Also i did not change any component, I tried using another GPU the same one (1660s but from MSI) it worked fine, so im pretty sure something is wrong with my GPU only
Click to expand...
And did you also tried running your gpu in another system ?
 
Vaibz said:
I used Arctic MX-4 carbon based thermal paste. I'm experiencing this from about a month now.

Also i did not change any component, I tried using another GPU the same one (1660s but from MSI) it worked fine, so im pretty sure something is wrong with my GPU only
Click to expand...
I assume your gpu requires 8 pin power so what psu you using and how much old it is ?
 
nRiTeCh said:
Then test gpu on another pc. I think gpu is dying...
Click to expand...
i did and faced the exact same issue. im pretty sure the gpu is indeed dead,
can it be sold for spares like fan or body at all?
also what could be the reason or its the tendency of gpus to start being problematic after a certain period of time, i have a gt210 bought in 2012 which is still rock solid and working, no idea what went wrong with this one :(
 
Fully uninstall the Nvidia GPU drivers with DDU once again.

Then open up Control Panel go to "View by" on the top right and select "Small Icons", then go to "Power Options" and then click on "Choose what the power buttons do" and then click on "Change settings that are currently unavailable" to activate the inactive options and the finally uncheck "Turn on fast start-up (recommended)".

While you're at it also follow the instructions on this Nvidia forum post on disabling Multi-plane Overlay (MPO) which are known to cause GPU related issues on WIndows 10/11 such as stuutering, random crashes etc.

Once that's done reboot the PC, then reinstall the drivers and then let me know if that did the trick or not.
 
Skewel said:
Fully uninstall the Nvidia GPU drivers with DDU once again.

Then open up Control Panel go to "View by" on the top right and select "Small Icons", then go to "Power Options" and then click on "Choose what the power buttons do" and then click on "Change settings that are currently unavailable" to activate the inactive options and the finally uncheck "Turn on fast start-up (recommended)".

While you're at it also follow the instructions on this Nvidia forum post on disabling Multi-plane Overlay (MPO) which are known to cause GPU related issues on WIndows 10/11 such as stuutering, random crashes etc.

Once that's done reboot the PC, then reinstall the drivers and then let me know if that did
Click to expand...

Skewel said:
Fully uninstall the Nvidia GPU drivers with DDU once again.

Then open up Control Panel go to "View by" on the top right and select "Small Icons", then go to "Power Options" and then click on "Choose what the power buttons do" and then click on "Change settings that are currently unavailable" to activate the inactive options and the finally uncheck "Turn on fast start-up (recommended)".

While you're at it also follow the instructions on this Nvidia forum post on disabling Multi-plane Overlay (MPO) which are known to cause GPU related issues on WIndows 10/11 such as stuutering, random crashes etc.

Once that's done reboot the PC, then reinstall the drivers and then let me know if that did the trick or not.
Click to expand...
Just tried everything you mentioned, no success
 
Last edited:
Vaibz said:
Just tried everything you mentioned, no success
Click to expand...
If that doesn't work, try downloading MSI Afterburner and just clicking on the reset button and see if your clockspeed resets to default, maybe there might be some auto under/overclocking at play. While you're at it, I recommend you to change your Windows power plan to "High Performance" and try again.

If both those options, don't work - then you should possibly consider RMAing your PSU.
 
Skewel said:
If that doesn't work, try downloading MSI Afterburner and just clicking on the reset button and see if your clockspeed resets to default, maybe there might be some auto under/overclocking at play. While you're at it, I recommend you to change your Windows power plan to "High Performance" and try again.

If both those options, don't work - then you should possibly consider RMAing your PSU.
Click to expand...
tried no success, unfortunately my PSU is out of warranty too (Corsair provides 3 YEAR warranty on VS series right?)
i'll try with another psu once im done with exams next month. :(
 
Last edited:
can you try lowering the vram freq by increments of 100 and see if it makes any differece?
also try reducing gpu frequency by 300-400 and try it.
 
awestorr said:
can you try lowering the vram freq by increments of 100 and see if it makes any differece?
also try reducing gpu frequency by 300-400 and try it.
Click to expand...
to be done through curve editor in MSIAfterbuner? if not, can you link me to a tutorial on how to do that it'd help out alot. thanks!
 
Back
Top