AMD 280X Crashes at medium temperatures

vvf3

Commendable
Jul 29, 2016
3
0
1,510
What's up people,

I have had to deal with my 280X by MSI (Gaming OC @1020MHz) acting up recently.

To preface, as it has overheated in the past, I have recently disassembled the card to replace the thermal paste on it and completely dust it off.

Now the card is running at smooth temperatures, as in it does not exceed the 75°C mark, which I think is okay for card in use for over 2 years. Furthermore, I have upped my case cooling game with 2 case fans sucking power in at the front and 3 cards, that being 2 on the top and 1 at the back, venting (hot) air out of the case.

Albeit running at temperatures within the ordinary, it is still acting up. It sometimes crashes when I'm only 5 minutes into csgo @60°C. However, the symptoms are the same as they have been in the past: I neither get a black screen or bluescreen but a coloured one (mostly green or blue but I can tell it's not a bluescreen) with vertical stripes on it.

Here's my specs:

GPU:AMD 280X Gaming OC by MSI running @1020MHz
PSU: BeQuiet System Power 7 500W (Silver 80+)
CPU: Intel Xeon 1230v3
Mobo: Asrock H87 Pro4 (1150 Socket)
5 Case fans, each one of them illuminated (3 Xilence ones, 1 Cooltek and the other one I can't tell right now)
16Gigs of ram
Prolimatech Lynx for my CPU cooler

On a final note, this problem seems very peculiar to me but it may just as well be the card suffering from the damage it has sustained.

Looking forward to your responses and help
 

vvf3

Commendable
Jul 29, 2016
3
0
1,510
Ive been digging recently and here's what I've done:

DBAN'd my pc and reinstalled all drivers - partially related to my problem, there was just too much junk on my pc to keep track of
As stated above, update the driver
Flashing the Bios (to a recommended one that is supposed to prevent the card from overheating)

Run Unigine-Heaven 3.0: I gave it eight runs of benchmarking lasting 260 seconds each:

First "error", as in coloured stripes, occured after 16mins. It looked more like a single frame. GPU load was 99% and max temp was 72°C
The second time an error occured I had just initiated the 7th run. In fact, the 7th run was rather full of errors: After a minute (26mins in total, approx.) the screen started flickering, with said errors and the generated picture alternating for over 10 seconds.
I also got errors when I had been 90 seconds and 130 seconds into the 7th run.
Peculiarly, I did not get any errors in the 8th benchmark.

I gave it a run in Furmark 2 mins after testing the card by using unigine heaven. Everything was fine until I passed the 3:30 minute mark. Having worked fine for 3:30 minutes, the card started screwing up, with both vertical and horizontal stripes coloured green or blue appearing all the time. The GPU load was at 99% all the time and the temperature did not exceed 75°C

So as already stated in my previous post: It's not a temperature issue. Now we know it's neither a BIOS nor a driver issue.
Can I permanently change my voltage in MSI afterburner?
Do you think my card is overvolted? Edit: My GPU Voltage is 1.200V
I'd appreciate suggestions to resolve this problem for as things are right now I don't have much use for this card anyway.