Introduction
------------

These two programs originate from Steve Gibson, an InfoWorld columnist. The
purpose of the programs is to measure how quickly your hard disk can read
individual tracks, and specifically how it pertains to hard disk interleaving.
The "interleaf factor" of a hard disk is the spacing of the sectors along a
track. If the interleaf factor is 1, then sequential sectors are contiguous.
If the interleaf factor is 3, then sequential sectors are positioned every
three places. If the interleaving of a hard disk is too high or too low, its
performance can degrade significantly. According to Steve Gibson, MOST hard
disks are not interleaved properly.

System Requirements
-------------------

DOS 2.0 or up is assumed and required for both programs. The hard disk must
be drive C: (the first hard disk), have its controller plugged into a slot, be
bootable, and not require any special drivers. Disk caching software should
be disabled or removed. Memory resident programs that "steal" computer time
should be disabled or removed. It is suggested that the computer is booted up
from a plain DOS disk, minus all programs and device drivers that might make
themselves resident.

The Programs
------------

SPINTEST measures the average number of revolutions the hard disk requires to
read a single track. It should take between about 14 and and 98 seconds to
run.

SPINTIME determines if the results returned by SPINTEST are valid. It should
take exactly 60 seconds to run and the number of RPM's returned should be
close to 3600. If either of these conditions are not met, the results
returned by SPINTEST are not valid.

The Results
-----------

The fewer number of revolutions the hard disk takes to read a track, the
better. If the interleaf factor is too low (i.e. the sectors spaced too close
together), it means the hard disk (and/or controller/computer) is not fast
enough to read sequential sectors in a single revolution. For instance, if a
hard disk has its interleaf factor set to 3 when it should be set to 4, three
or more times the number of revolutions might needed to read an entire track.
If the interleaf factor is too high, then the sectors could be closer together
and the disk would not have to spin as much to get to the next sector. The
original IBM PC/XT's hard disk had an interleaf factor of 6, and its
performance can be moderately improved by changing it to 5. Many hard disks
nowadays come with an interleaf factor of 3, even though most controllers
and/or computers would work better at 4.

My Experience
-------------

I have an old IBM PC into which I installed a 20 MG hard disk. Since nothing
told me to do otherwise, I formatted the hard disk using the default interleaf
factor of 3. When I ran SPINTEST, I found out that my hard disk rotated an
average of 12 times to read a single track. Suspecting that by increasing the
interleaf factor I could improve my hard disk performance, I changed the
interleaf from 3 to 4 by backing up my hard disk, doing a low level format
with an interleaf factor of 4, repartitioning the disk, doing a high level
format, and finally restoring the disk from its backup copy (whew!). Now,
when I ran SPINTEST, I found that my hard disk needed to rotate only an
average of 4 times to read a single track. By changing the interleaf factor
from 3 to 4, I had speeded up my hard disk about 300%.

What to do Next
---------------

If you know what your doing and have a good idea what your interleaf factor
should be, you can try what I did (DO NOT try it if you DON'T know what you're
doing). Your best bet is probably to get the interleaving information
yourself (see below). Included with the information is a promotion for a
soon-to-be-released commercial program that will determine the optimum
interleaf factor for a hard disk and automatically change it without
destroying the hard disk data.

You can request Steve Gibson's free interleaving diagnostics (which consists
of the two programs and an information brochure) by sending a self-addressed,
stamped business-size envelope to:

Gibson Research Corp.
Box 6024
Irvine, CA 92716

This document is brought to you by:

Eric O. Tauck
72457,1557

I am in no way affiliated with Steve Gibson or Gibson Research Corp. You can
ask me about your hard disk if you want, but I don't know that much about them
and probably will not be able to answer you questions.