personally if my chip dies I would RMA as long as no physical damage was done to it. I remember when I went through a bad batch of XP 1800+s and I wasn't overclocking them I put them in and they would break down during a linux kernel compile. The hard thing about stress testing dual cores is things that work on single cores don't even come close to stress testing a dual core even if you run more than one instance of the program and force it to run on a certain core. it's a good orthos blend works well. I hope intel just RMAs the X chips and doesn't check them, because that's their publicly OCed chip.
nice guess but if they went that route they might as well put some sort of bar code in the chip that would have the info in it so they could just scan it rather than having to plug it in. I think the extra cost of 128 gates would be too much and raise the price of the chip.

