fried gtx 680m

My story is a sad one...I installed a new gtx 680M 3 days ago...everything was fine. I only tested it 10 minutes in crysis 3, 10 minutes in witcher 2 and a game of dota 2. Temps barely reached 65 degrees. I saw that the core clock reached 627mhz under full load, 0.925V. I think I should have tryed to flash it with svl7's vbios flashes...but instead I trye to see if nvidia inspector did anything, I raised the core clock by about 50mhz and started furmark to see if the clock indeed goes beyond 627mhz. After almost 10 seconds the temp reached 67 degrees and then screen went black. 3 days used gtx 680m fried. After reboot it seemed to boot windows (heard the windows noise) but screen was still black. Tryed to see if video port works but nothing, tryed the vga cable on an old monitor and success. Had image. From nvidia control panel I checked the laptop monitor but it would not work. After 1 minute, screen went black and now laptop does not start properly...I have the blinking leds lights for "The video card is preventing the system from completing POST.". I think it's pretty futile right now...but I still ask it...any advice? Anything more I can check? There are no visible burns on the card or anything.

 

Just got your message, poor chopsy and poor gtx 680m at least now you know next time not to mess around with furmark and flashing dodgy files onto the expensive GPU. Are you going to purchase a new 680m?

 

I have not flashed it. Did not even overclock it. Core clock was at 627mhz when the screen turned black. I can't belive the thing is so fragile!

 

Sorry about those issues with the 680m. It seems like the only route now is to send it back. I assume it should have a warranty?

 

I sure hope so. I bought it from widezu69, he said that if I send it back to him he could try to get the card replaced. I'm waiting to see if I have any other options of getting it fixed.

 

In my opinion , I don't think a m15x has the right cooling for a 680m.

 

It ran cooler than my gtx 260m.

 

The only temp sensor on nvidia cards are located on the die, so even though it is correctly cooled down it does not necesarily means that all the other components normally covered by thermal pads are - and this will cause the card to die.
The die can run over 100 degrees without getting damaged, caps, ICs, etc cannot.

widezu69 is a nice guy, I'm sure he will sort you out in the end