Loren said:
It's a matter of language. There's no offical standard, how do you
prove which one is right?
Do we use the normal metric approach where k = 1000, M = 1,000,000 and
G = 1,000,000,000??
Or do we use the computer engineer's approach where k = 1024 and so
on?
Uh, as a long-time (i.e., aged) computer engineer, I recall that the
K=1024 notion was the de facto standard *only* for RAM; and that, because
RAM was nearly always offered in binary sizes. For other computer
hardware items, including HDs, tapes, communications widgets, and CPUs,
the K=1000 notion was -- and is -- far more common.
And, there are standards (SI and IEEE, probably others) supporting the
K=1000 notion, but no standards supporting the K=1024 notion. So, the
HD vendors can't IMHO be found wrong by rational people when using
decimal notions of size prefixes (k/M/G/T/P/E/Z/Y) -- particularly when
they explicitly note on their ads and labels that they do so.
That said, I appreciate M$'s use of both methods to show the properties
of a HD (in Explorer, right-click on any HD then click on Properties),
to at least alleviate confusion; it is probably too late to eliminate the
confusion by switching universally to decimal size prefixes.