My 750ti has a 6 pin connector, so it's getting it's juice from the PSU. The 750ti was running in there for over a year with no problems. Only after putting in the RX 480 did the slot die. Here's a pic of the 750ti: this is different than the pictures you showed the first time and the motherboard is different to, pcie slots in the first one is blue and this one black ?
So what is the out come? did you reach out to AMD and your mother board maker, too see if anyone will take ownership of the dead slot?
Can you please supply windows system logs of the crash as a start as proof that maybe the card did kill the board?
Yes, it is a different motherboard. That is my wife's computer. I put the 750ti in there after I got my RX 480. That picture was just to show the 750ti and the 6 pin connector.
I haven't reached out to anyone. As I had stated from the start, I was not looking for anyone to compensate me or replace anything. I'm perfectly content running in slot 2.
The event viewer simply shows a critical Kernel-Power Event ID 41:
An update, though:
Yesterday I had an 8 hour gaming session after installing the new drivers with the power fix (I did turn on the new comparability mode) and nothing blew up! So that's a good sign! This was with the very same +83MHz overclock on the core clock (nothing on the memory and no voltage or power target adjustments).
I was absolutely loving the RX 480 until it started smoking spontaneously and then died on me. Thankfully the PCI-E slot wasn't damaged so I was able to stick my old GTX 760 back in and I was up and running again. Can't help but be really disappointed though as up to this evening I was loving it (60fps on Doom maxed out with Vulkan).
I didn't overclock it or tamper with the cooler and I've a a Fractal Tesla 650w PSU so I highly doubt that had anything to do with it 😕
Did you not update the drivers to the new 16.7.2? They fixed the overdrawing issue in that driver (especially if you turn on compatability mode, which I did).
Yes, I did update to the latest driver release but I didn't turn compatibility mode on. I assumed that as I had a half decent MB and PSU and I had no intentions of overclocking I'd be OK to not enable compatibility mode. To be fair, the fault may not be related to the power draw issue and may be related to my particular card - I was loving the performance boost over my GTX 760 and needed a bit of a vent
There's nothing on the back of the card showing any damage but the smoke and burning smell was most prominent near the HDMI port so I tried to take a photo of the board through the vent. This is the best I could do as I don't want to undo any screws as I've an RMA sorted with OCUK:
Disclaimer: I'm a programmer, not an electronics expert, but it looks pretty scorched to me.
My first reaction was to go for a GTX1070/80 but before I'm accused of being exclusive to team green, I cooled down a bit, remembered the price/performance ratio, DX12, Vulkan and async compute advantages, and decided to replace it with 2 x Sapphire RX 480 Nitro+ (I'm a firm believer that multi GPU is the future).
I'm a self employed software developer and I'd just started work on a Vive project and to be honest, when I couldn't work on that and had to drop back to < 30fps on Doom with my GTX 760 I needed a bit of a vent - hence my post. As I mentioned to roquen22, it is probably just my individual card.