>>Have either Sean or Dave ever put their methodology in writing here
Yep, both of them. They merely installed NT in a small
partition (the only one on the drive) then grew the
partition with Partition Magic to the full drive size.
In both cases the drives were 13GB IBM IDE drives, and I
believe both were 7200 rpm (I only remember
that explicitly from Sean's posts, but I'm reasonably
sure Dave had the same drive).
I tried the same thing on a Maxtor 17G drive and failed,
as reported here at the time (and recounted painfully
to you a few messages back <gg>). For all I know, it's
also a matter of drive geometry.
FWIW, PC Mag recently benchmarked these exact two drives,
the IBM 13.2GB 7200 vs the Maxtor 17.4GB 5400. PC Mag,
as you all know, is not my most trusted rag, but here
they described their methodology in detail even to
my satisfaction, and the maxtor beat the IBM in all
benchmarks. They even showed you why (which, if you
read back here far enough, you will find my discussion
on how that could be). A very satisfactory verification.
But I still can't boot from a large partition on the
Maxtor ( <G> + :( ).
BTW, not a benchmark, but one area the IBM will always
beat the maxtor (or any 7200 drive will always beat a
5400 drive) is in random access of a small amount of
data. That's limited to the current physical cylinder
(physical head position) if the drives have widely
unequal seek times, but as most recent drives are within
a few percent of each other on that, it really means
over the whole drive. That's because the rotational
delay is independent of the data; it depends only on
the rotation speed.
Spots |
