3.5" floppy failure mode? Hub not attched to media.

Peter Corlett abuse at cabal.org.uk
Fri Feb 27 07:54:17 CST 2015

On Thu, Feb 26, 2015 at 02:26:05PM -0800, Chuck Guzis wrote:
> It isn't particularly the track pitch that gets to be critical, but the
> radial positioning of the hub that defines the index position. If you can
> read disks without relying on the index pulse, you may have better luck.

The original poster is reading Amiga disks. Although the Amiga hardware can
sense index pulses, the native format ("trackdisk.device") does not make use of
them. When the Amiga writes a sector, it actually does a read-modify-format of
the whole track. This is why the sectors are all bunched together with a single
large inter-sector gap per track, and also why "You MUST replace Volume foobar
in DF0:" due to ejecting a disk mid-write, as the whole track is now corrupted
and needs to be re-written.

I've not disassembled mfm.device to see if it does the same thing with MS-DOS
disks, but I'd be surprised if it didn't.

As to the disintegrating disk, there's probably some cute bit-nibbling trick
that can be performed by bit-banging the Amiga's disk countroller to retrieve
the data even if the media is misaligned due to re-gluing, but it's not worth
the effort for a disk that's clearly badly-made and was probably unreadable
even before it fell apart.

More information about the cctech mailing list