On Wed, 5 Sep 2012, Tothwolf wrote:
The first person who comes up with a way to reliably
predict drive failure
would become an overnight billionaire.
Your spinning rust will fail!
[Am I the first? USA dollars or facebook shares?]
Oh.
You want to know WHEN? Prob'ly in less than 40 years. plus or minus 40.
Statistically, drives tested with Spinrite fail sooner than those that
never get tested at all. (likewise, people in hospital critical care
wards are more likely to die than the general population of people at home.)
How many people remember when Spinrite BY DEFAULT would take any KNOWN BAD
tracks, and return them to use if they happened to pass the Spinrite
tests? (Spinrite testing is substantiqally less thorough than the analog
testing that manufacturers do.]
It consistently could not find fault with the track that Windoze 3.10
setup consistently died on. Thanks to DUMBDRV's delayed writes, there
were no choices of Abort, Fail, or Ignore, and Windoze did not give a
choice about it installing the cacheing software and misconfiguring it.
I told MICROS~1 that they had a problem; they said "hardware problem, not
interested"; they had to do a free replacement of MS-DOS 6.00 with 6.20
(to configure the cacheing in a less suicidally agressive setup)
Mmmm popcorn...
They will last longer in a cool dry environment with no magnets and no
popcorn.
--
Grumpy Ol' Fred cisin at
xenosoft.com