Hello! 
 
This is my first post here, so please go easy on me.    I tried searching before posting. 
 
I'm looking for some guidance to see if you guys can help me figure out what is causing multiple card failures. 
 
Here is my sad story...
 
My original card, an EVGA GTX 470 (012-P3-1470-AR), lasted for three and a half years and failed in May 2014.  It abruptly started showing artifacts (lines all over screen) and indicated failure in windows hardware manager.
 
The replacement card was an identical model, worked perfectly for about two and a half months, and then failed in the same way.
 
Before installing the current card (second replacement...same model), I checked the voltage of all pins in both of the 6-pin molex connectors powering the card using a calibrated multimeter.  All six wire pairs had an identical voltage of 12.4V.
 
My computer had startup issues immediately after installing the current card.  I would get one good boot out of 5 or 6 attempts.  On failed attempts, the screen was blank after windows startup screen, but you could tell everything else was working normally (windows startup sound, etc.). 
 
On good starts, within a minute the screen would go blank for 1-2 seconds, and an error message box would appear stating "NVIDIA Windows kernel mode driver stopped responding and has successfully recovered" or "NvBackend.exe - corrupt file".  After the error messages would appear, the screen would act normal until shutdown.
 
I deleted and reloaded all video drivers with current and older versions but it had no effect.
 
This current card failed over the weekend (same as the first two cards with artifacts/lines after boot), about three weeks after I received it.
 
I was told by EVGA that all of these replacement cards are refurbished.  Is it possible that I received two bad RMA cards in a row, or is it more likely that there is something in my hardware that is causing the cards to fail?  At this point, I'm not sure what I should check, or how I would go about it. 
 
At a minimum, I am planning on installing the next replacement into a friend's computer to burn it in before installing it in mine.  
 
Is there something else I can do to check the PSU?  Could my motherboard be killing them?  If so, how would I check the motherboard? 
 
If any of you can give me some suggestions, I would really appreciate it!
 
Thank you for any ideas,
 
Mike
 
Here are the specs for my other desktop components, if it helps any: 
Motherboard - Gigabyte GA-X58A-UD3R
Processor - Intel Core i7-950
PSU - Corsair HX850
Memory - Mushkin Enhanced Ridgeback 2X2GB DDR3 1600 (996826)

I remember those problems with my GTX 480 and GTX 580's. For me, it was remedied by upping the voltage through EVGA precision or MSI Afterburner. Give it one to two bumps in voltage and it's no longer a problem. For me it would only happen on certain games... mainly ones that would put heavy load on the card.

maniacvvv
Welcome to the EVGA forums
 
Careful reading of your posted issues and computer specs points to the probable cause of your problems.
-->Limited system RAM and use of a low VRAM videocard
Then using the GeForce Experience with those issues!
 
Use of the GeForce Experience
*(should NOT be installed)
 
Use of RAM sticks known to have issues with x58 platforms -AND- use of only--> 4gigs
*(likely to be running out of memory AND incorrect RAM BIOS settings)
-----------------------------------------------------------------------------------
 
-->Completely uninstall the Geforce experience and read up on how to manually install your driver (only Graphics driver and PhysX)
-->Check your BIOS settings for your RAM sticks. XMP should NOT be used and the timings and voltage should be set manually.
 
You need to check your system RAM load on a clean boot (how much your system is using on the desktop)
Should you have to much loading at boot, this will impact how your video card will use resources and is most likely the cause of your problems.
 
Use of only 4gigs of system RAM and using a videocard with just 1280mb of Vram is underpowered for running ALOT of things.
It will require getting a very clean boot and paying alot of attention to VRAM loading (and the settings that cause such VRAM use)
 
Using the Geforce experience is going to cause serious problems with incorrect settings and overloaded VRAM.
The program is NOT setup to run correctly on a system like yours and will cause issues...
This appears to be the primary cause of your problems
 
 
 

+1. This is good troubleshooting to start with in my opinion

Thanks to all of you for your replies! 
 
maniacvvv,
 
What memory, and how much, would you suggest that I use with my hardware? 
 
Would the memory I have be okay if I had another 2 GB's of it?   I tried to choose my hardware carefully and was unaware of any conflicts.
 
It's funny that you mention GeForce Experience.  I'm running in safe mode and can't see....Is that the application that automatically loads with EVGA's driver download and sits in the background and looks for and installs new drivers?   When I think back, it seems that I installed this software right before my first card failure.  However, I was running everything at default and had no changes for fan speed, voltage, etc.
 
If what I just described is in fact GeForce Experience, why in the world would EVGA put out software with default settings that is bad for (and in my case, destroys) their own graphics cards? 
 
Can you point me to some instructions or links on how to manually install only the non-lethal drivers?
 
Thanks again,
 
Mike
 
 
 

I was running gtx480's in sli with a corsair hx1000.  When the corsair started going bad my system refused to boot.  I replaced it with an enermax platimax 1350w which booted my system right up until one of my 480's died, which I need to rma.  I would boot the system, if possible and then check the rails with them under load to see if there is voltage sag.