Re: Slot 'extender' dev cable

[pausch@saaf.se (Paul Schlyter)]

In article <1170506846.660794.180760@v33g2000cwv.googlegroups.com>,
David Wilson <mcs6502@gmail.com> wrote:

> On Feb 3, 8:12 pm, pausch@saaf.se (Paul Schlyter) wrote:
>> If ProDOS is as good as DOS 3.3... :-) ...then ProDOS too should be
>> able to fully access a ProDOS floppy with extra tracks at the end
>> (assuming that the ProDOS disk contains correct info about its
>> disk geometry).
> 
> The ProDOS Disk ][ driver "knows" how many blocks are present on a 35
> track floppy and validates the block number passed in before
> attempting to read or write the block.

That's the way CP/M worked too: it merely assumed a particular disk
geometry - and if the disk you inserted had another geometry, well,
tough luck!

> The practical upshot is that you need to patch the kernel (either on
> disk or in RAM after loading) as well as the disk formatter program
> (see Understanding Apple ProDOS for the addresses to patch [you will
> need to adjust these addresses for more recent versions of ProDOS 8]).

CP/M had a *documented* way of acheiving this, described in the CP/M
alteration guide: each logical disk had a Disk Parameter Table, where
you could specify things as the number of logical sectors (or "blocks"
in ProDOS jargon - a CP/M "block" was only 128 bytes though), the
number of entries in the directory (there were no subdirectories), the
interleave factor, the number of system tracks, the cluster size, etc.
Nowhere on the CP/M disk there was any info about the disk geometry of
that disk.

> As I have a 40 track drive attached to a UDC card I simply updated the
> ROM from 35 tracks (280 blocks) to 40 tracks (320 blocks) and I can
> run completely standard ProDOS and applications with full 40 track
> support.

....but what happens if you then insert a 35-track floppy?  I assume
reading it would be OK, but if you tried to write it, could your patched
ProDOS then attempt to write beyond the last (35'th) track?

In CP/M you could set up two different CP/M volumes to refer to the
same physcial disk - the two volumes could then specify different disk
geometries for that physical disk.

-- 
----------------------------------------------------------------
Paul Schlyter,  Grev Turegatan 40,  SE-114 38 Stockholm,  SWEDEN
e-mail:  pausch at stockholm dot bostream dot se
WWW:     http://stjarnhimlen.se/