On Fri, 8 Sep 2017, John Foust via cctalk wrote:
When attemping to read an old Amiga floppy, the magnetic media
separated from the metal hub. Were these originally glued?
Assuming I can find a way to re-affix the media to the hub, how many
guesses as to the proper alignment might I need to make? After all,
the metal hub has an index slot and the media was formatted in
relation to that.
Is the answer different for a PC floppy?
Partially.
The PC has a gap after index, and then a smaller gap between each
sector.
With the PC, the index is only needed during formatting, BUT, even during
reading, the NEC-style FDC goes "blind" momentarily right after index.
Thus, any sectors too soon after index are "not found". However, if you
encounter that problem with disks that are otherwise PC readable, you can
"mask"/interrupt the index line of the floppy cable. On many old 5.25"
drives (NOT including Teac 55), you can even just put a write-protect tab
over the index access hole of the floppy jacket (on Teac and a few others,
the drive is dependent on seeing index to be "ready"). Interrupting the
cable is not without a few minor complications. For example, if you ask
for a specific sector, the PC (INT13h) normally waits through a few
revolutions, then reports error code 4 ("Sector Not Found"); but, if
you've interfered with the index, then it will come up with error code 128
("Drive Not Ready") instead.
Reading PC sectors without masking will work fine, except for any that
straddle, or are too soon after, index. Some disk formats, such as
Cromemco have very tight inter-sector gaps, and reading with PC can be
less unreliable.
On the Amiga, there are little or no gaps between sectors. It essentially
does a track read/write and then parses in memory, instead of searching
for individual sectors to read. It probably relies on index to decide
when to start reading. It may be possible to write some code on the Amiga
to read a track, and then splice from end to beginning to patch it.
What are you imaging it with?
If you could read the entire track (such as with a WD 179x), ignoring
index, and then splice material from the end to the beginning as
necessary, then you would be OK.
At least SOME flux-transition boards (I KNOW CP Option Board does) require
the index pulse to operate. (THAT made it trivial to create a disk that
would choke Copy-II-PC-OptionBoard software, but could be copied [losing
the protection] with DISKCOPY)
If you can read "more than a track", then you can manually find and mark
where the start of track is.
The weird hacks to fool the NEC into reading track (in order to read Amiga
on PC) might or might not work with index masked. Without it masked, your
placement of index will have to be accurate.
--
Grumpy Ol' Fred ci...@xenosoft.com
