David Wilson

On Nov 24, 1:29 pm, David Wilson <mcs6...@gmail.com> wrote:
> The problem with patching it to support 80 or 160 tracks is fitting
> the code into the space available - it is optimized for testing $01xx
> blocks.

Good news. I checked my copy of the Beneath Apple ProDOS supplement
and it turns out that the Disk II driver occupies $D000..$D6FF and the
last 22 bytes ($D6EA..$D6FF) {directly following the command/block
number check} are unused. If this is not enough space, I suspect that
converting the driver from 6502 to 65C02 code might free up more
space.

ferdimh@gmx.de

On 26 Nov., 23:48, David Wilson <mcs6...@gmail.com> wrote:
> On Nov 24, 1:29 pm, David Wilson <mcs6...@gmail.com> wrote:
>
> > The problem with patching it to support 80 or 160 tracks is fitting
> > the code into the space available - it is optimized for testing $01xx
> > blocks.
>
> Good news. I checked my copy of the Beneath Apple ProDOS supplement
> and it turns out that the Disk II driver occupies $D000..$D6FF and the
> last 22 bytes ($D6EA..$D6FF) {directly following the command/block
> number check} are unused. If this is not enough space, I suspect that
> converting the driver from 6502 to 65C02 code might free up more
> space.
OK then I know now that it's free. I thought it would be free as it
was all zeroes, but I didn't know for sure.
When checking for any other number, the code would be one byte longer
(DEX is repaced by CPX).
I need more space to fit my track number translation routine. It has
11 bytes, so it should fit.
I wouldn't like to convert it to 65C02 code since that would mean a
complete reassembly.
At the moment I'm working on the formatter code to have some disks to
work with. I wrote the formatter for DOS3.3 since there is the init
routine easily available. I think I could copy the ProDOS Global Page
to another location and load RWTS under ProDOS.
There is another problem I encountered when patching RWTS: The read
routine uses the variables from the seek code causing a lot of
trouble. I hope this is not the case in ProDOS, but I don't know.

ferdimh@gmx.de

I finally got it working now.
Because David Wilson told me that there's a ProDOS "RWTS" disassembly
on Apple Assembly line, I managed to fully understand the code and to
make a first running patch. But I couldn't access track 0 on the
second side (i.e. logical track 80). Today I figuered out this was a
fence-post error (the tracks are numbered 0-79 so track 80 and not 81
is the first track on the other side.) Now I'm making plenty of copies
of the disk with an (also self-written) 160-track disk copier and hope
at least one of them will survive whatever happens.
I noticed another interesting phenomenon: Writing to disk 1 is ok, but
writing to disk 2 is extremely slow. I have no clue what the cause
could be. This does not apply to all programs. ADT uses both drives at
high speed.
I really can't think why. Maybe you can...?

---
Ferdinand Meyer-Hermann