🚀 go-pugleaf

Michael Wong <nospam_mikwon@ebtech.net> wrote in article
<34933e67.166479364@news.ebtech.net>...
> >> >Actually the Ultrastar 9ZX has started shipping and it sustains about
> >> >20MB/sec.
> 
> >> Not under real-world conditions.  You're talking about
> >> contiguous-block transfers on a nearly empty hard drive.  In the real
> >> world, drives are partially full and fragmented.
> 
> >Contiguous blocks, yes.  Empty/full has nothing to do with transfer rate
> >except the drives transfer files located near the OD faster than those
near
> >the ID.  Most workstation hard drives are not very fragmented.
> 
> First you say that empty/full has nothing to do with transfer rate,
> and then you admit that it makes drives slower as they fill up (since
> they move toward the inner tracks), proving yourself wrong before you
> even finish the sentence.

You don't read English very well.  Empty/full has nothing to do with the
transfer rate.  Where on a drive information is recorded does effect the
transfer rate of that information.  A full drive reads the information
recorded at the OD just as fast as an empty drive.  Now do you understand
it.

> 
> Then, you say that workstation hard drives are not very fragmented,
> with absolutely ZERO evidence to support your claim.  Your penchant
> for unsupported and just plain indefensible statements is as strong as
> ever, Ron.

That statement needs no defense as all the competent high performance
workstations users that I know of periodically defrag.  All their DLLs and
EXEs and swap file..... are defraged and only late installs or other recent
files have the possibility of being significantly fragmented.  I guess that
you don't realize that much of a workstations I/O involves reading
relatively fixed files base.  You need to get more extensive real world
experience before you spout in these NGs.

Regards,

Ron Reaugh