Originally posted by: BentValve
There is a global positioning sensor on the CPU and they can track what you are doing with it at all times, sorry man you cannot get away from the big AMD in the sky.
They should have a thermometer stuck inside the CPU package, kind of like when you buy a frozen turkey. If the sensor "pops out" - you're done! No RMA for you!Originally posted by: V00D00
It says on the package, if you run it at any speed higher than it is rated for the warranty is voided.
That's not exactly true. The CPU mfg states guidelines for both normal operating temp ranges, and additionally at what temp ranges operation will damage the chip. If the chip ever hits those temps (on the high end), it's operating outside of the valid range, and is therefore not covered by warranty. The only way that it would hit those temps, when not overclocking, is when it wasn't properly installed. In any case, I've never heard of a CPU that was otherwise working, non-overclocked, within normal temp ranges, suddenly "melting down" and exceeding safe operating temps. In other words, it's not the CPU's job to remain in that temp range - it's the job of the system integrator installing the CPU to ensure that it always stays within the safe operating temp ranges.Originally posted by: Crimson
Here's the problem with some of the 'solutions' that people have presented to the problem.. If they put in some sort of detector if someone has overclocked the cpu, such as temp sensors, the problem lies in that a REAL cpu malfuction could cause it to overheat thus tripping the sensor.. There is no way to differentiate between a real malfuction and one caused by overclocking.
Originally posted by: VirtualLarry
They should have a thermometer stuck inside the CPU package, kind of like when you buy a frozen turkey. If the sensor "pops out" - you're done! No RMA for you!Originally posted by: V00D00
It says on the package, if you run it at any speed higher than it is rated for the warranty is voided.
(Bizarrely, that's actually a completely viable low-tech solution for being able to detect if the CPU core was ever run beyond mfg's specified thermal limits, assuming that you chose the melting-point of the fusable-link alloy properly. I wonder how much that would increase packaging/production costs for CPUs? They could even put it in the base of the heatsink instead, and pass the cost on to the heatsink vendors instead.)
Originally posted by: JonathanYoung
What's punctation?