Can a Motherboard kill a GPU ?

Mar 8, 2018
2
0
10
I hope you guys can help me with this one.


So, two weeks ago my computer started shutting down and rebooting with a message saying : " Power surges detected during previous power on. ASUS Anti-Surge was triggered to protect system from unstable power supply unit ! " whenever I tried launching a game. I did some research online and came to the conclusion that the PSU was most likely at fault so I changed for an EVGA Supernova 650W G3 80+ Gold. After installing the new power supply, the computer booted as usual and everything worked just fine. I even ran Unigine's superposition benchmark and everything went well. I then proceeded to try and launch a game and the computer shut down once again. After several minutes I tried to start it back up by pressing the power button and I saw a spark and smoke coming out of the case. I immediately turned off the PC from the PSU switch. The next day I looked inside and there was a burnt spot on one of the motherboard's metal components (I believe they are for heat dissipation) It clearly came from the GPU since once I removed it, I noticed that the  card's PCB was burned. After entirely removing the card I tried booting the computer and once again everything worked just fine, I used the PC without a GPU for about a week and had no issues. I then proceeded to RMA the card which was still under warranty. While I was waiting for the new card to arrive I bough a cheap 1050 off of Amazon and ran several benchmarks with it, mainly Unigine stuff and the same games that used to cause my PC to shut down. This time everything went well, no crashes, no shut downs, nothing. I also tried some AIDA64 benchmarks and stress tests. Today the RMA'ed card arrived and I'm not sure if I should put in and test it. I'm afraid that it could get damaged. Any suggestions on what to do or insight on the subject are greatly appreciated.

Google Drive Link for Images : https://drive.google.com/open?id=1Ckui_THtKyGGaTRWUEfKTM4R1WGDtnj6




  • The case was clean from dust, it didn't experience high temperatures.
    I don't believe the GPU dying can be attributed to a overheating problem since it died when I tried powering on the computer and the card was cool.
    Old card (980Ti) was using an 8+6 pin setup for power the 1050 only gets power from the PCI-E Slot.
    Right now I can't really change the card slots since the top PCI-E slot is obstructed by the CPU Cooler.
    Changing the whole thing (Motherboard/Cpu/Ram) isn't really an option right now unless I buy another Z-87 motherboard.


System specs :



Motherboard : ASUS Z-87 Pro LGA1150

GPU : GeForce GTX 980 Ti Superclocked+ ACX 2.0+     New Card : EVGA GeForce GTX 1080 SC GAMING

Power Supply : Seasonic M12II 620W 80+ Bronze          New PSU : EVGA Supernova 650W G3 80+ Gold

Storage : Samsung 850 EVO 500GB / WD Black 1 TB

RAM : 16GB Kingston Hyper X DDR3 1600

 
Solution
Here's my best guess. Your video card was bad....which caused the first PS voltage to fluctuate...giving you the error. The second PS was more robust so the voltage didn't fluctuate so you didn't get the error.

Once you launched the game the GPU really started sucking power and failed big time.

So I think it was the GPU the whole time.

I would install the replacement GPU.
Here's my best guess. Your video card was bad....which caused the first PS voltage to fluctuate...giving you the error. The second PS was more robust so the voltage didn't fluctuate so you didn't get the error.

Once you launched the game the GPU really started sucking power and failed big time.

So I think it was the GPU the whole time.

I would install the replacement GPU.
 
Solution
Mar 8, 2018
2
0
10


I installed the new GPU yesterday and ran some stress tests / Benchmarks also tried the games that used to make the computer shut down and everything seemed normal. I monitored it with msi Afterburner and haven't see anything out of the ordinary in the graphs. I think it was the old GPU that was faulty as you said. Thank you for your feedback.