Have you ever seen a broken CPU

[Random_G33K]

Hey, I was just wondering how many of you have seen a faulty CPU. I mean a CPU that died of an unknown cause (no spills and such accidents). Preferably CPUs after the year of 2000.

 

[thently]

?? Are you calling the whole computer a CPU or just the processor? The answer is yes several times and it is mostly electrical issues power supply or motherboard power regulators caps mostly. These can leave no visual reason for the whole setup to go pear shaped..

Thent

 

[Random_G33K]

I mean just the processor.

 

[Boy07]

I mean just the processor.

Processors are not easy breakable or damaged,but SKYLAKE processors can bend when installing aftermarket cooler.But otherwise they can only get damaged if overheated(extremely)

 

[Guest]

the reason behind the death of CPUs for which there is no known cause is usually electromigration.
http://www.tomshardware.co.uk/answers/id-2530767/electromigration-cpus.html

 

[Boy07]

the reason behind the death of CPUs for which there is no known cause is usually electromigration.
http://www.tomshardware.co.uk/answers/id-2530767/electromigration-cpus.html

Rarely,yes.But in general processors are pretty durable and strong.

 

[Guest]

the reason behind the death of CPUs for which there is no known cause is usually electromigration.
http://www.tomshardware.co.uk/answers/id-2530767/electromigration-cpus.html

Rarely,yes.But in general processors are pretty durable and strong.

not rarely, the transistor traces and nano circuits are being eroded in our CPUs even now as i type this. run a CPU benchmark on a newly purchased CPU and again after 3 years, you will notice that the CPU performs a bit worse...a bit. after some time the electromigration effect will be so bad that there will be frequent bluescreens and crashes...especially with CPUs that are primarily used for gaming.

 

[molletts]

I did have one about 5 years ago; I'd never had one before anywhere and it took me a while to pin it down. It was a Socket 604 Xeon processor in an ex-work server I had at home. I think it may have been caused by vibration - I "mothballed" the server while I was having electrical work done involving cutting channels in walls (sealed in a plastic bag to keep dust out and stored in a room that wasn't having any work done) but when I unpacked it afterwards, I got a peculiar error early in POST (something about the PIC timer not ticking, if memory serves me correctly).

I assumed, having reseated everything, that the system board was the likely cause so I replaced it but the error didn't go away. Eventually, I got desperate and took the machine back into work, arranged some downtime on the other still-in-use one and did a systematic parts swap which revealed the CPU to be the cause. One new (and amazingly cheap) Xeon CPU later and the server was fine. It later went back into service at work when we needed an extra machine. I've still got the duff CPU as a "souvenir" - the only faulty CPU I've ever had in 20+ years of working with computers.