"Fred Cisin (XenoSoft)" <cisin(a)xenosoft.com> wrote:
Did they have the same number of sectors on each track
("constant
rotational speed"), or did they have different numbers of sectors on
different tracks ("constant linear velocity")?
The Twiggy disk uses a zoned-CAV system like modern hard drives, except
that it changes the spindle motor speed and keeps the data rate constant.
Thus the number of sectors per track varies.
There are 46 tracks for data (numbered 0 to 45), and track -1 which is
used for speed calibration.
I'm assuming that track 0, the outermost track for the lower head, has
a radius of 2.25 inches, the same as a standard 5.25 inch drive. The
Twiggy uses a track pitch of 62.5 TPI.
The zone information is given in this table, which is compiled from
the Lisa Hardware Manual; I've added the peak flux density which I
computed in an Excel spreadsheet.
512-byte peak
sectors rotation flux
per rate density
track track (rpm) (fcpi)
----- ------- --------- ---------
0-3 22 218.3 9721-9933
4-10 21 228.7 9550-9989
11-16 20 240.1 9588-9973
17-22 19 252.7 9552-9955
23-28 18 266.8 9509-9931
29-34 17 282.5 9463-9907
35-41 16 300.1 9414-9981
42-45 15 320.1 9453-9749
Note that for Apple's (6/8) GCR format, equivalent to (0,2) RLL,
the minimum flux density is 1/3 of the maximum flux density. For MFM, the
minimum is 1/2 the maximum. This means that the media has to work for the
range of 3138-9989 FCPI.
Standard 5.25 inch floppies spin at a constant rate, and transfer data
at a constant rate, so they have a much lower flux density on the outer
tracks.
For 360K format, the maximum flux density ranges from 3537 FCPI on side 0
track 0 to 5877 fcpi on side 1 track 39, for an overall range of
1768-5877 FCPI.
For 1.2M format, the maximum flux density ranges from 5895 FCPI on side 0
track 0 to 9870 fcpi on side 1 track 79, for an overall range of
2947-9870 FCPI.
Obviously, then, my inital guess that high-density media is required
would appear to be correct.
GCR would tend to provide a flux transition density of
about 1.5 times that
of FM, and slightly less than that of MFM.
Apple's (6/8) GCR is exactly 3/4 as efficient as MFM.
Eric