fightinfilipino
there's like 10 posters on a web forum loudly complaining. that's hardly exposing anything, that's about the rate of manufacturing error expected for ANY complex electronics!
good lord.
And a very good point, looking at it statistically, I'd imagine there are no more or less failures or vocal customers than any other forum for any other GPU. It's just the nature of the issue is likely to claim many more cards over a longer period of time. I don't have a copy of furmark nor do i overclock, (pascal seems to be good enough to sort that out on its own) however I do believe depending on the build quality of the specific unit, temperature degradation will play a causal factor to failure, and it would seem this specific design is quite poor at managing VRM/memory temperatures. as per the temperature reviews looking specifically at the power delivery section of the board.
In my job, I do the opposite, and focus on trying to keep the heat in, and where the use of a solid medium isn't possible, air works pretty well, especially when non ventilated. So it comes as little surprise that when we see the photos of the VRM surrounded by air pockets between the board and support plate that temperatures will be high, and that no amount of fan speed is likely to really aid the VRM temp unless it is in direct contact