Macintosh disks are funky.

It is fairly well known that Macintosh low density disks are not compatible with IBM PC computers. Macintosh 400k single sided and 800k double sided low density use a different and more finicky low-level format encoding - GCR, like the Apple II, and different sector organization. But on top of that, they use variable bitrates.

On an IBM PC style drive, the floppy controller blast bits to and from the drive at a consistent 250kbps or 500kbps, and the disks always spin at a constant 300RPM (or 360RPM in the case of 5.25" 1.2mb floppy drives). This has the side effect that physically on the disk, bits are crammed closer together towards the center of the disk.

So someone at apple got the brilliant idea that changing the drive's motor speed as it reads/writes different tracks would let them cram more data on disk. This actually worked, and worked OK, except for the fact that almost NOBODY else did it that way.

A nice trick was developed with the Central Point Deluxe Option Board/Transcopy card for the IBM PC. Instead of changing the drive speed, the hardware changes the bitrate that it reads and writes at. Devices like the Kryoflux and SuperCard pro can essentially do the same trick as they work at the flux level and can encode/decode any bitate.

Unfortunately, this trick is not compatible with all floppy drives. Floppy drives have filters that remove some analog "noise" received by the drive head, and automatic gain control that can actually create noise when there are large gaps with no flux transition.

In the case of IBM PC style floppy drives, this normally restricts frequencies to those normally present on an "MFM" or "FM" encoded low or high density floppy disk.

When reading a Macintosh disk in such a drive, the effect is that either the inner or outer tracks will appear garbled and mostly unreadable. A few drive models, mainly from Teac and Toshiba, are not as restrictive and may permit all tracks to read or write.

But it is still a "hack", as these drives were NOT designed for this.

What this means, is that when you attempt to read or write a Macintosh low density disk with a flux-level device using a standard floppy drive, you will almost certainly have to experiment and try multiple drives. This is especially true of any disk that is badly worn or that was written to in different drives - which is the case for most early Macintosh disks.

I have been trying to read in some badly worn floppy disks. The data is there, but the wear combined with differences between the drive the disk was mastered in and the user's drive, made these extra difficult.

I have had to flip between a pile of drives, and mix-and match different track reads from different drives. Some will barf on inner tracks, some will barf on outer tracks, some seem to not like other tracks for alignment or stability reasons. The behavior of these drives also differs depending the disk itself. One drive that normally barfs on Macintosh disks got a perfect read of one of these disks. What?!

It makes very little sense, really, but Macintosh disk flux dumps usually show quite a bit of "noise", so even the slightest read issue turns in to a total read failure depending on which way the wind is blowing.

Heck, for some reason writing a brand new 400k/800k floppy on a perfectly fine floppy disk using a Power Mac, is always completely unreadable by my Kryoflux. Yet everything else works fine.

The point is, Macintosh disks can be a real bitch. :P