New GTX 970 Killed in action, looking for replacement

nonakosa

Reputable
Jun 14, 2015
2
0
4,510
Hello, I recently built a new machine and have experienced a catastrophic gpu failure. I bought an MSI GTX 970 Gaming 100ME, and everything was running fine until yesterday afternoon. I loaded the Witcher 3, (which came with the card), and after playing for about 3-5 minutes the computer shut down and I could smell burnt electronics from under my desk. The gpu was hot, but not skin-melting hot, and the pc would only boot up if I removed the power connectors from the card. I had previously been able to play this game on this hardware without issues, so I am a little confused as to how this problem manifested. Everything in the machine is new, so the cost will be refunded and I will replace the card, now I am looking for a replacement that I can feel confident buying.
I have been doing research as to what may have happened and all I can find is that 1) the Witcher 3 seems to push some cards pretty close to overheating, 2) bad Nvidia drivers that caused the fans not to spin, or maybe 3) bad batch of cards that were destined to fail anyway. I prefer to think that it is option 3 as the card came with the game and so should be able to run it, and because Nvidia drivers should never kill Nvidia cards, but I am also a bit out of my element with this problem. I would prefer to replace my GTX 970 with the same card - it was great while it lasted - but I don't want to take the risk of cooking another card and having to replace it again. Is it likely that I just got a bad card? Should I skip the 100 ME and go for the one without the backplate, (I have read that the backplate can cause higher temps)? Am I wrong in thinking that an MSI card should work well with an MSI motherboard?
My full specs follow:
MSI Z97-G45 Gaming Motherboard
Intel i5 4690k 3.5GHz CPU
Corsair Vengeance Pro 32GB RAM
Corsair CX Series 750w ATX PSU
Samsung 850 EVO 500GB SSD
WD Blue 1TB HDD
Samsung SATA 1.5 Optical Drive
MSI GTX 970 Gaming 100 ME GPU

Thanks in advance for your help..I am eager to get this machine back up and running.
 
Solution
Let me clear a few things up.
The GTX 970 can run the Witcher 3 on ultra 60fps.
I know this because I have a Strix 970 and it works perfectly fine and stays under 70c.
Now, one of a few things happened.

1 -The card was faulty from the factory (Probably this one)
2 - The PSU killed your card (unlikely if it's been working fine with other card)
3 - You have exceptionally bad airflow in your case and the card couldn't stay cool enough (unlikely considering these card are designed to shut off if it gets too hot)

It's probably just a faulty card. Get a new one and monitor while you use it. If it happens again, it's your system killing the card.
Why did you even unplug the pcie power connectors. That was a mistake. Witcher 3 is quite graphically intensive so its not surprising if it is pushing your card to work hard.
GTX 970 is designed to have a 0db mode when under low load/temps.
So what temperature did you card go up to ? If it is 80C, then that is fine as it is safe for it to run that high for gaming.
 
If The Witcher 3 has caused your GPU to overheat, then once you replace the card, you should run the game either locked at 30 fps or on medium settings to prevent the same fate from happening to the next card. Either two GTX 970s in SLI or a GTX 980 Ti can better cope with the game on ultra settings.
 


think they mean after the pc shut down it wouldnt come on till they removed the pic-e connectors
 


that card should run it without a 30fps cap

maybe not everything on ultra but certainly above medium

mine runs at about 55fps--60fps with most settings at high or above at 2560 x1440

 
Let me clear a few things up.
The GTX 970 can run the Witcher 3 on ultra 60fps.
I know this because I have a Strix 970 and it works perfectly fine and stays under 70c.
Now, one of a few things happened.

1 -The card was faulty from the factory (Probably this one)
2 - The PSU killed your card (unlikely if it's been working fine with other card)
3 - You have exceptionally bad airflow in your case and the card couldn't stay cool enough (unlikely considering these card are designed to shut off if it gets too hot)

It's probably just a faulty card. Get a new one and monitor while you use it. If it happens again, it's your system killing the card.
 
Solution
Thanks to everyone who replied, and apologies if I am not using this forum correctly..this is my first post to this forum and I think that I should have set it up for a discussion instead of solution when I initially posted. Since I do not see a reply button I am posting this as an "answer".
To Joshua Martin and others who replied that it may be a faulty card, this is kind of what I am also assuming. I do have plenty of airflow in the case and the psu has been working fine with everything up until this point. I have since this morning tested the card in another machine with the same results, so I guess that for now I am going to assume that it was a faulty card, re-order another one and go from there. I still have my apprehensions, but just because it all seems to have been so random. The card ran great for the few weeks that I had it in there..until it didn't. I suppose that is just the nature of faulty hardware. What would you all suggest that I use to monitor the new one when it arrives?

Thanks again
 


 

TRENDING THREADS