Question GPU Blackscreen Crash Under Load

[Sparky1780]

Hi all,

For the last month I've been having issues with my Gigabyte 2080 Super. On stock settings the card will backscreen crash once its under load. All fans go to 100% and the pc is unresponsive and I have to do a hard reset. then when booting back up the card is completely fine until its under load again.

I thought it could be a power issue so I purchased a new 1000w Corsair RM1000e, I doubt the unit is faulty but I guess its a possibility. I am using stock cabled and have 2 separate 6+2 Pin connectors into the GPU.

What I've tried:

• Testing RAM with windows memory tester and trying each stick individually
• Re-pasting the GPU
• Undervolting to both 1900Mhz at 900mv and 1900mhz at 875mv - this provides a temporary fix that makes the card useable but will still crash after a few hours.
• Updating Motherboard and GPU bios.

PC Specs

• AMD Ryzen 3600
• Asus B450-F Gaming
• Corsair RM1000e
• Gigabyte RTX 2080 SUPER WINDFORCE OC 8G
• 32gb DDR4 Ram (16gb Corsair kit, 16gb crucial balistix kit)

It used to log Kernel error 141 in event viewer but recently I've been getting no error logs in reliability monitor and 0n event viewer.

I only bought this card a few months ago from eBay and it worked perfectly fine for those two months now is only starting to crash. Will I just have to accept the card is dying and buy another.

 

[Colif]

event 41 (63) is the reaction to a restart. Odd you don't get any if its still black screen crashing. If you have to restart, it should at least create that. Anyway, seems to show its a hardware problem

I assume its not heat? what temps you getting?

tried it in another PC?

How long you been using missmatched ram? Tried it with just 1 stick?

 

[Sparky1780]

event 41 (63) is the reaction to a restart. Odd you don't get any if its still black screen crashing. If you have to restart, it should at least create that. Anyway, seems to show its a hardware problem

I assume its not heat? what temps you getting?

tried it in another PC?

How long you been using missmatched ram? Tried it with just 1 stick?

That's what I assumed with those errors too, just due to me hard resetting.

I have been using the RAM kits for a couple years now on a different GPU.

I have just tried using 1 stick of RAM and did a few more tests on Heaven Benchmark. Again with default speeds the crash still happens. I will attach some of the GPU-Z Logs but from what I can see I don't think it points to an issue. The temperature is around 60-65 and never hits more than 72.

I did try the GPU in another PC and it still crashed but the PSU was only 500w in that system so the issue might have occurred regardless.

Sometimes the event logs error Liver Kernel Event 1b8 but yeah very strange the error logging has seemed to stop.

GPU-Z Logs - https://filebin.net/qacpkb48du72vxt8

 

[puw2]

When repasting the gpu did you check the thermal pads also? Could be that one is out of place or slightly creased/loose causing this issue. .