Re: Apple ][ plus 13 sector! Problems...

[mjmahon@aol.com (Michael J. Mahon)]

Eric Smith wrote:

<snip>

>Reading a data mark works similarly.  The data mark pattern for 16-sector
>ormat is D5 AA AD.  There is a self-sync pattern before the data mark.
>The software (RWTS or equivalent) doesn't attempt to find the data mark
>until immediately after it has decoded the desired address mark, so it
>doesn't try for nearly as long.  It would not be acceptable for it to
>read the data field of the wrong sector.

It is worth pointing out that no self-sync bytes would be needed before
data marks if data fields were never re-written, since the state machine
would stay in sync between the address and data fields.

When a write to disk is done, the software reads, searching for the
correct address mark and address, as when reading, but then it
switches on the write head and (re-)writes the data self-sync nibbles,
the data mark, the data, and the data epilog nibbles.  (Actually, because
of a bug, it only writes the first two epilog nibbles correctly, so the
third epilog nibble is ignored on read.)

Because switching on the write head cannot be guaranteed to be
"coherent" with the previously written nibbles, new "lead-in" self-
sync nibbles must be written to ensure that, when reading, the state
machine will be correctly synchronized.  More than the minimum
required number of self-sync nibbles are written to create a "gap"
to accomodate the write head switching and disk speed variations.

-michael

Check out parallel computing for 8-bit Apples on my
Home page:  http://members.aol.com/MJMahon/