Live data from Hacker News

Emulating the Early Macintosh Floppy Drive

thomasw.dev

41–43 of 43 posts

Re: Emulating the Early Macintosh Floppy Drive

#41
post #33

Earlier quoted context omitted.

Almost all digital formats die in obscurity. 3” floppies, 5v SmartMedia, Jazz disks, and DAT were good ideas in the moment that were not good ideas a moment later. The logistics of reading even popular formats like Qic 40 and ADAT today are hard.

The most ironic thing, IMO, is that at the time everyone seemed to be holding their breath for Castlewood's 'Orb drive', which promised the perfection of a fast and big storage media. It was delayed so long that by the time it actually reached the market it didn't get noticed.

Never heard of Orb, but once upon a time, I was waiting for Tandy’s THOR. Before that Flieschman-Ponds cold fusion.

https://en.m.wikipedia.org/wiki/Thor-CD

Re: Emulating the Early Macintosh Floppy Drive

#42
post #16

Does anyone make a usb floppy drive capable of reading Mac disks? I would be willing to pay quite a lot for one.

If you have important Mac disks to archive, then a Kryoflux is probably the way to go. It's a external USB attached floppy drive controller that uses a normal PC floppy drive. It reads the raw disk stream and software then decodes that for whatever disk format it was. It seems old Mac disks can be recovered with it: https://www.kryoflux.com/?page=home http://forum.kryoflux.com/viewtopic.php?t=135

KryoFlux is long obsoleted by the open-source GreaseWeazle.

Re: Emulating the Early Macintosh Floppy Drive

#43

It's interesting how much of the (latency sensitive) data processing is done on the main CPU; I've not used early Macs much, but given they have a multitasking GUI, would user input be noticeably delayed during disk activity, or could it cause problems to e.g. move the mouse rapidly while it was accessing the disk? 16 CPU cycles per bit sounds like it wouldn't leave much time for anything else.

The Macintosh would disable any interrupts that were not level 4 or higher when accessing the floppy disk to prevent data loss. This was basically everything other than the programmers switch. It did however have a check to see if the serial port received data while it was busy with the floppy disk so that could be serviced before a buffer overrun occurs.
Post reply on HN