Discussion R9 280X dead three days after installing 22.6.1 Driver

[CardellB]

Thread for the public record, in case more people with HD7000, R# 200/300 - series have a similar issue. Maybe we can find a common cause.

Brief description of the situation:

1. Screen suddenly went black while using the computer. Monitor was still on, as indicated by the continued presence of backlight and no monitor standby indication (typical of GPU failure).
2. Computer turns on, but won't boot (not even to the BIOS splash screen) for a few minutes, unless the computer is plugged off and left for a few minutes.
3. After waiting with the computer unplugged, the computer boots normally again.
4. As soon as the GPU usage reaches around 40% usage for a few seconds, or the temperature reaches 50ºC (cannot clearly identify which one, or if both), we are back to step 1 (GPU failure).

Already tried:

1. Using DDU and then reverting back to previous drivers - same situation.
2. Using the GPU in another slot - same situation.
3. Using the onboard graphics - works.
4. Using another Graphics Card (Strix 1080Ti OC, which should draw more power than the 280X) and running GPU, CPU and RAM intensive tasks for prolonged periods - worked.

Imo, 2. and 3. confirm the GPU is the culprit

NOTES:

• Exact model: Sapphire R9 280X Vapor-X OC (Dual-fan).
• GPU was running normally until the point of failure - no artifacts, noticeable weird noises, performance losses, flickering, throttling or anything else could be noticed.
• Max temps were never going past the peak of ~71ºC, which were normal since 2013.
• Only OC being run was the Factory OC from Sapphire.
• The PSU is a 850W Corsair RM850 PLUS Gold 2021 (and the 1080Ti OC works fine with it) - so it shouldn't be a problem.

(Irrelevant, further) notes:

• OS: Windows 10 with all the latest updates.
• CPU i7-7700k
• MOBO: Maximus IX Code
• RAM 16GB Corsair Vengeance LPX

 

[Deleted member 362816]

I have that same card on a shelf, Let me run it for a few days on the new drivers and I will report back.

 

[CardellB]

I have that same card on a shelf, Let me run it for a few days on the new drivers and I will report back.

Thanks, mate! 💪

 

[Deleted member 14196]

It’s most likely a coincidence

 

[CardellB]

It’s most likely a coincidence

I seriously hope the short time between the driver installation and the GPU dying was a coincidence...
Though, if considerably more people have such issue, then we should start to question such events. Yet, taking the lack of answers to this topic into consideration, it was most likely an unhappy coincidence.

Still, I wish it lasted me longer, as the modern-day equivalent cards (3070, 6700) cost over 2+ times as much as I paid for my (new, undiscounted) 280X back in the day...

Cheers!

 

[spentshells]

Thread for the public record, in case more people with HD7000, R# 200/300 - series have a similar issue. Maybe we can find a common cause.

Brief description of the situation:

1. Screen suddenly went black while using the computer. Monitor was still on, as indicated by the continued presence of backlight and no monitor standby indication (typical of GPU failure).
2. Computer turns on, but won't boot (not even to the BIOS splash screen) for a few minutes, unless the computer is plugged off and left for a few minutes.
3. After waiting with the computer unplugged, the computer boots normally again.
4. As soon as the GPU usage reaches around 40% usage for a few seconds, or the temperature reaches 50ºC (cannot clearly identify which one, or if both), we are back to step 1 (GPU failure).

Already tried:

1. Using DDU and then reverting back to previous drivers - same situation.
2. Using the GPU in another slot - same situation.
3. Using the onboard graphics - works.
4. Using another Graphics Card (Strix 1080Ti OC, which should draw more power than the 280X) and running GPU, CPU and RAM intensive tasks for prolonged periods - worked.

Imo, 2. and 3. confirm the GPU is the culprit

NOTES:

• Exact model: Sapphire R9 280X Vapor-X OC (Dual-fan).
• GPU was running normally until the point of failure - no artifacts, noticeable weird noises, performance losses, flickering, throttling or anything else could be noticed.
• Max temps were never going past the peak of ~71ºC, which were normal since 2013.
• Only OC being run was the Factory OC from Sapphire.
• The PSU is a 850W Corsair RM850 PLUS Gold 2021 (and the 1080Ti OC works fine with it) - so it shouldn't be a problem.

(Irrelevant, further) notes:

• OS: Windows 10 with all the latest updates.
• CPU i7-7700k
• MOBO: Maximus IX Code
• RAM 16GB Corsair Vengeance LPX

Heat gun that card

 

[Deleted member 362816]

Heat gun that card

Don't understand why this was downvoted as he has nothing to loose at this point, I have toaster ovened quite a few cards in my day still have a 1070ti from 2+ years ago that is still working after (vram issue originally)

 

[CardellB]

Heat gun that card

Even though it seems simple, my knowledge about the intricacies of hardware is close to zero 😆, but I'll try to get the courage and do it 💪

Tbh I had never heard about this "home-tooled" reflow method, so thanks for the tip!
(and I'm unable to perform "conventional" reflow as I lack both the expertise and tools needed )

Peace!