Re: Reading Pascal 400kb Disk

[mjmahon@aol.com (Michael J. Mahon)]

In article <F7ScnXhcHboyxvejXTWcrg@comcast.com>, "Joel"
<joelbuckley54.nospam.@hotmail.com> writes:

>"Michael J. Mahon" <mjmahon@aol.com> wrote in message
>20030308145558.09900.00002271@mb-fc.aol.com">news:20030308145558.09900.00002271@mb-fc.aol.com...
>
>> The issue of reading a single-sided 3.5" disk on a double-sided
>> drive is only an issue of software.  It is only necessary to avoid
>> reading sectors that are on the unused side of the disk, and then
>> to interpret the single-side sectors with appropriately translated
>> block numbers to correspond to its file system.
>>
>> As Wayne has pointed out, GSOS can interpret single-sided
>> ProDOS 3.5" disks.  What you are looking for is a program that
>> can interpret a single-sided Pascal file system.
>>
>> If all else fails, you can write such a program.  The job is made
>> much simpler because the Pascal file system stores files as
>> contiguous ranges of blocks--no index lists to worry about.
>>
>> I believe that someone posted here a while back that the physical
>> sector ordering of a double-sided 3.5" disk was alternating sides
>> by cylinder, so every N sectors (N/2 blocks) should be on the
>> formatted side, and readable by any ProDOS block editor.
>> (Don't use GSOS if GSOS can't "mount" the disk.)
>>
>
>Two things:
>
>As I recall, the tech note for two 400k volumes on one 3.5 disk, uses the
>first 800 blocks for volume 1 and the second 800 blocks for volume 2. They
>do not use separate sides. Both volumes use both sides.

Agreed.  That's why I suggested reading every other blah blocks, where
"blah" is the number of blocks on the track.  If the outer zone is 18 blocks,
then the first 18 (say) should be on one side, then skip 18 and the next
18 will be on the same side, etc.  Sorry for not spelling it out...

>My memory is that GS/OS's Pascal FST did NOT recognize this format. My
>recollection is that it didn't recognize all 3.5 format Pascal disks,
>either. I seem to remember that a number of multi-os file copy programs did
>not support Pascal on 3.5 disks even when they did support 5.25 Pascal
>disks.

A multi-OS copy will not be sensitive to the file system, and will just
copy sectors from one disk to another.  The only problem would be that
the copy program, expecting a two-sided disk, will get a read error as
soon as it tries to read a sector on the uninitialized side.

>Years ago I had written a Pascal attach.driver for a 3.5 drive under Pascal
>1.2 (Apple didn't support it until 1.3). I've never heard of a version 1.4.
>The Pascal blocks are in the same order as ProDOS. I have a vague memory
>that at some point they had boot blocks for Pascal that would work with
>either 5.25 or 3.5 disks.
>
>A minor difficulty here, if I remember correctly, is that the 3.5 drive
>doesn't have the same number of blocks on each track. I don't remember the
>exact layout, but the outer tracks had more blocks per track than the inner
>ones. That's why we have 800k instead of 720k.

Right--there are five "zones".

>Assuming you can read the blocks on one side of the disk with a sector
>editor, the easiest thing might be to write a disk copy program which copies
>the 800 blocks on one side to the first 800 logical blocks on a standard
>double sided disk. The disk might not be bootable, but any data on it might
>be recognizable either with Pascal 1.3 or through one of the multi-o/s file
>transfer programs.

This would be a good start, since it might just make the disk mountable
by most Pascal file system-capable programs.  Of course, someone has
to write the program that reads only the sectors on the formatted side.
With patience, it could be done manually (all 80 tracks...).

-michael

Check out amazing quality 8-bit Apple sound on my
Home page:  http://members.aol.com/MJMahon/