Re: Commodore 64 kills APPLE II

[dempson@actrix.gen.nz (David Empson)]

Michael J. Mahon <mjmahon@aol.com> wrote:

> Roger Johnstone wrote:
> 
> >The Apple UniDisk 3.5 is programmable. It has it's own processor and
> >enough RAM to hold a program sent to it from the host computer.
> 
> It is my understanding that the UniDisk 3.5 is a smartport device.
> So the question is whether there is a command implemented in
> its firmware that allows transfer of control to a program stored in
> its SRAM.

Yes, there is.  It was specifically designed to support copy protection
mechanisms - you can download code into the drive (and execute it) but
cannot upload the code back to the computer.

> >The Apple IIgs is fast enough to control the 3.5" disk drive directly,
> >so the Apple 3.5 Drive doesn't have a processor. However, the IIgs can
> >store the same routines in it's own memory and run them in the same way
> >as the processor in the UniDisk 3.5 does.
> >
> >That said, does anyone know if any software used this ability, rather
> >than just using the routines in ROM? Copy protection schemes perhaps?
> 
> All the usual tricks permitted by the SWIM chip, which has capabilities
> and limitations very similar to the Disk ][ controller, though at a higher
> data rate.
> 
> I'm unaware of the full panoply of copy protection tricks employed
> on 5.25" disks being applied to 3.5" disks, however.  Is there some
> limitation on direct control of the head positioning motor on the 3.5"
> drive?  (Or had the industry simply outgrown most of that by the
> time the IIgs and 3.5" drive became common?)

The 3.5" drive mechanism is inherently "smarter" than the 5.25" one.  It
probably has a microcontroller on the drive itself, which is responsible
for details like stepper motor control.  The only head movement
supported is complete tracks.  (The drive mechanism uses a command
interface for operations like moving the head, with something like a
three or four bit command code and one data bit.  It has a similar
mechanism for reading status information like the write protect sensor.)

Code downloaded to the UniDisk 3.5's on-board 65C02 processor is able to
change details like the address marks, but there isn't enough RAM in the
drive to do things like entire raw track dumps.

-- 
David Empson
dempson@actrix.gen.nz