1 Oct
1998
1 Oct
'98
6:31 p.m.
< > this for all conventional formats, but not Apple disks. It would be
< > nice to connect one of these and read the disk image into a file.
< > The main reason why I ask is for rescuing messed up floppies...
Apple used GCR coding for their drives and there are no commonly
available chips that do. In fact the apple does not use the traditional
floppy disk controller chips at all. The FDC logic on the Apple is
rather simple and the CPU provides the brains as a software FDC.
It was part of the idea of using software to replace hardware to make
the machine simpler and less expensive.
< PCs use a disk controller chip based on an NEC 765 design that can not
< read ANYTHING other than MFM (some can also do FM), with IBM 3740 style
Excuse me! That is patently incorrect! If you qualified that as *most
PC controllers cannot read some formats* that would be true.
The 765 and all of the chips based on the 765 core can read FM encoded
disks. It's not always possible with PC based designs as some are
hardware crippled (sloppy data sep circuits) and most of the PC bios code
is marginally brain dead.
However the 9266, 37c65 and 92c665 have been seen doing FM.
In the mean time my S100 card design and the compupro DISK1A both happly
read 8" SSSD (FM) disks and even single density 5.25 media.
< The Western Digital disk controller chips (179X, before they changed ove
< NEC style) could do a reasonably real track read of anything MFM
< regardless of sector header structure.
Try the 765 read diagnostic command. Hex 02 (fm) or 42 for MFM.
<They were used in TRS-80s and a LOT of machines OTHER THAN PC.
The TRS80 used the 1771 that was FM only.
the WD 177x or WD179x still cannot read an apple disk.
Allison
1 Oct
1 Oct
8:10 p.m.
On Thu, 1 Oct 1998, Allison J Parent wrote:
Apple used GCR coding for their drives and there are
no commonly
available chips that do. In fact the apple does not use the traditional
floppy disk controller chips at all. The FDC logic on the Apple is
rather simple and the CPU provides the brains as a software FDC.
It was part of the idea of using software to replace hardware to make
the machine simpler and less expensive.
And more hackable. Imagine being able to create your own disk formats!
Clever developers did this. Probably the most interesting was the
18-sector DOS that Broderbund used on some of its later Apple ][ games.
This served two purposes: 1) it allowed more storage space for programs on
a single disk (normal Apple DOS only had 16 sectors) and 2) it foiled
software pirates as you now had to figure out ways to cram 630 sectors
worth of data onto a 560 sector normally formatted disk so it could be
copied with a regular disk copy program (I solved the problem for myself
my writing an 18-sector disk copier using the RWTS [DOS] of the game).
Sam Alternate e-mail: dastar(a)siconic.com
-------------------------------------------------------------------------------
Ever onward.
Coming in 1999: Vintage Computer Festival 3.0
See http://www.vintage.org/vcf for details!
[Last web site update: 09/21/98]
2 Oct
2 Oct
4:58 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 2332 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19981002/97353885/at…
9:20 a.m.
On Fri, 2 Oct 1998, Cameron Kaiser wrote:
You could be just as tricky with Commodore peripherals
because the Commodore
drives had their own RAM, ROM and 6500-series CPU. Enterprising programmers
created 40-track DOS for it instead of 35 (some 740-odd sectors free instead
of 683) and Commodore fastloaders have long been based on this, usually using
the CLK, SRQ or ATN line as a second DATA line on the serial bus and custom
software on either end.
You could easily hack Apple DOS to 40 tracks, but the limitation was some
hardware (drives) couldn't seek that far. The Apple Unidisk (or 5.25"
drive as it was later generically called) was only capable of 39 tracks.
I can't remember what the original Disk ]['s could do. I remember being
able to get some drives up to 40 tracks.
One of the most original uses for the Commodore disk
drives, though, was
nothing data transfer related. A math demo for the 64 used the 1541 as a
co-processor for calculating the Mandelbrot set. Unfortunately, this didn't
yield the speed increase one would think because of the slug-like speed of
the Commodore IEC serial bus, but it was still significant, and a neat idea.
I would think similar things have been done with Atari 8-bit drives.
This is indeed clever. Hans Franke was telling me about some guy in
Europe (Germany?) who hooked up a bunch of 1541's to a single C64 and then
parted out a Mandelbrot calculation to each to do extremely fast Fractals.
I think the coolest C64 disk drive program was the one that made the drive
play "Daisy" by recalibrating the drive head at different intervals; the
drive head would be hitting the stop at different frequencies and created
a violin like sound. I was fascinated with it. I attempted the same on
the Apple with a Unidisk drive, but the best I could achieve sounded
closer to a Tuba under water. Or a goose being choked.
Sam Alternate e-mail: dastar(a)siconic.com
-------------------------------------------------------------------------------
Ever onward.
Coming in 1999: Vintage Computer Festival 3.0
See http://www.vintage.org/vcf for details!
[Last web site update: 09/21/98]
11:58 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 1061 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19981002/2927eb6a/at…
1 Oct
1 Oct
9:29 p.m.
On Thu, 1 Oct 1998, Allison J Parent wrote:
< PCs use a disk controller chip based on an NEC
765 design that can not
< read ANYTHING other than MFM (some can also do FM), with IBM 3740 style
Excuse me! That is patently incorrect! If you qualified that as *most
PC controllers cannot read some formats* that would be true.
We appear to differ only on the definition of "PC".
The PC disk controller, AS IMPLEMENTED BY IBM, did NOT provide access to
all of the capabilities of the 765. In particular, FM can NOT be done
with that board unless you make a few modifications, or replace it with
any of the better designed after-market boards.
The 765 and all of the chips based on the 765 core can
read FM encoded
disks. It's not always possible with PC based designs as some are
hardware crippled (sloppy data sep circuits) and most of the PC bios code
is marginally brain dead.
However the 9266, 37c65 and 92c665 have been seen doing FM.
With the 37C65 and the like, the switch between MFM and FM can be done
entirely in software, not requiring the extra hardware gate needed for
the original 765 as used by IBM. Thus, any of the FDC boards using those
chips are generally a better choice than the IBM ones. And some, such as
the MicroSolutions Compaticard were specifically marketed for their
versatility.
Almost any of the controllers with 1.2M capability can handle 8" MFM (for
obvious reasons). Most of those with 37C65 or the like, or modified to
permit FM switching can also do 8" FM.
In the mean time my S100 card design and the compupro
DISK1A both happly
read 8" SSSD (FM) disks and even single density 5.25 media.
An excellent choice for the task.
Try the 765 read diagnostic command. Hex 02 (fm) or
42 for MFM.
Very useful. But I'd STILL rather have a real MFM (and FM) raw
track read.
<They were used in TRS-80s and a LOT of machines
OTHER THAN PC.
The TRS80 used the 1771 that was FM only.
The TRS-80 model I did indeed use the
1771 (FM). Data separation was poorly
done, and there were after market boards to patch it.
The Percom Doubler for the model 1 used either the 1791 or 1793 (I forget
which).
So did the Lobo expansion interface,
and so did the TRS-80 model 3 and 4. But Radio Shack made some WEIRD
design decisions regarding use of Data Address Marks.
the WD 177x or WD179x still cannot read an apple disk.
Absolutely correct. But they CAN read some MFM formats that do NOT have
IBM style sector headers, unlike the 765 series.
There were several add-on boards for PCs for dealing with Apple ][ disks,
including the Apple Turnover and Matchpoint. They installed between the
existing FDC and the drives, similar to the Option Board. Reliability
was less than ideal, and some drives worked better than others.
--
Fred Cisin cisin(a)xenosoft.com
XenoSoft http://www.xenosoft.com
2210 Sixth St. (510) 644-9366
Berkeley, CA 94710-2219
9430
days inactive
9431
days old
5 comments
4 participants
participants (4)
-
allisonp＠world.std.com -
cisin＠xenosoft.com -
ckaiser＠oa.ptloma.edu -
dastar＠ncal.verio.com