Re: [LI] [too-offtopic] SDRAM

[Bhyrava Prasad <prasadb@xxxxxxxxxxxx>]

Rohit wrote:
> 
> > > How long is a processor supposed to last for NO overclocking?
> > No one knows. I don't think that the people at Intel test a
> > processor for 10 years before they release.
> 
> So there IS wear and tear there. I remember a prof of mine telling
> me that processors NEVER fail. How could that be? I had been wondering
> since..

My info is at least a decade old now, but I believe
that the metallization layers on the chip expands at a
different rate from the silicon and continuous heat
cycling (switching on and off) can eventually cause
failure. There are other failures, which a good VLSI 
book will probably describe.

However, it is much more likely that your CPU will get outdated
before failing , or some power spike will kill the machine. 

If you are overclocking , make sure you use a large heatsink
and a good fan. Use heatsink compound or a thermal forming
material (high end heatsinks) between the heatsink and CPU. 
There are plenty of
sites describing overclocking. There are even active heatsinks
(using peltier effect devices for cooling , and extreme ones
such as liquid nitrogen). Essentially (AFAIK), electron mobility is
more at lower temps , as the lattice vibration is less. However, 
this is probably way offtopic , however interesting.

- Bhyrav Prasad
--------------------------------------------------------------------
The Linux India Mailing List Archives are now available.  Please search
the archive at http://lists.linux-india.org/ before posting your question
to avoid repetition and save bandwidth.