18 Sep
2006
18 Sep
'06
5:06 a.m.
Subject: RE: CompuPro floppy controller differences
From: "Chuck Guzis" <cclist at sydex.com>
Date: Sun, 17 Sep 2006 11:32:01 -0700
To: cctalk at classiccmp.org
On 9/16/2006 at 4:30 PM Allison wrote:
Many can. The original IBM PC version was a an abortion. I'd always felt
they worked hard making that mess.
As to the A/nonA part thats less a problem than often thought. If it
really was a problem you could change out the part.
I have a few mid range ISA-8 FDC boards that use the 9216 data sep and
they do read SD 8" and a bunch more.
Ah, it's using the definitive PC FDC. It was
developed well before
PCs standardized. PCs however don't offer much choice and if they did
keep all the FDC features you could put a 8" on a PC and read SSSD!
You could?!? (if I'm parsing your statement correctly). IIRC, the original
PC FDC card had IBM's proprietary data separator on it (several hybrid
packagaes) and couldn't read SD/FM to save its life. It also had the 8272
(non "A") part, so it had other problems with "alien" formats. Unless, of course, we're talking about PCs in the
generic sense and not the
5150 per se. There, it's a crapshoot as to what capabilites any given FDC
card has.
In my original statement I was refering to the generic and as "Standard"
its still anything but. When you consider the number of drives and
configurations the floppy whent though on PCs over the years it's a big
mess. It's only that it's all behind us that was know where the bodies
lie and we can assemble it from known history. but it lead to 3.5" drives
with out drive select jumpers/switches and other "innovations".
Over the lifetime of PCs I've worked with:
5.25 180k single sided
5.25 360k 2 sided
5.25 720k 2 sided 80 track
3.5" 720k
3.5" 1.44mb
3.5" 2.88mb
Syquest, Zip, Jazz and other higher density removeables
The only thing they didn't do stock was 8" SSSD and 8"DSDD, and
hard sector! The aftermarket added those!
Standard, really, more like a computer version of a leatherman.. ;)
Allison
18 Sep
18 Sep
9:08 a.m.
On 9/18/2006 at 8:06 AM Allison wrote:
5.25 180k single sided
5.25 360k 2 sided
You've forgotten the original DOS 1.x 160K SS and 320K DS. I wonder if the
conservative 8-sector approach on the part of IBM/Microsoft was the
miserable track record of the original (IIRC, Shugart) drives. The one
that came with my original 5150 eventually died when an inductor in the
motor tach circuit opened up. I replaced it eventually.
Cheers,
Chuck
5:51 p.m.
On Mon, 18 Sep 2006, Chuck Guzis wrote:
You've forgotten the original DOS 1.x 160K SS and
320K DS. I wonder if the
conservative 8-sector approach on the part of IBM/Microsoft was the
miserable track record of the original (IIRC, Shugart) drives. The one
While there
might be regional variations, I NEVER saw IBM use a
Shugart in the 5150. For the first year or so, it seems like they were
all Tandon TM100-1, and later IBM contracted to have drives made with IBM
front panels by CDC? and others.
But not Shugart. OTOH, TRS80 and Apple][ both used Shugart mechanisms
(SA400 for TRS80, SA390?? for Apple(with Apple's own board on them))
Most of the early Shugart SA400s were 35 track, instead of 40.
7:49 p.m.
On 9/18/2006 at 5:51 PM Fred Cisin wrote:
But not Shugart. OTOH, TRS80 and Apple][ both used
Shugart mechanisms
(SA400 for TRS80, SA390?? for Apple(with Apple's own board on them))
Most of the early Shugart SA400s were 35 track, instead of 40.
I had a Shugart SA450 on a PC, but for the life of me, I don't recall if I
bought it as a second drive or if it was the first one that came with the
machine. An unremarkable drive in any case. Perhaps the original drive
was a TM-100-1; I don't recall since I migrated to half-highs soon after
purchasing the system and added a Micropolis quad density (full height) in
the opened-up space.
Cheers,
Chuck
9:22 a.m.
On 9/18/2006 at 8:06 AM Allison wrote:
3.5" 720k
There were two "standard" versions of the 720K 3.5 format, differing in the
cluster and FAT size. One was the definition put out by IBM sometime
around MS-DOS 2.1 and the other, by Microsoft in MS-DOS 4.00 (which was an
abomination).
So, you haven't used 3.5" 1.3MB DOS diskettes? (2x8x1024)? Many USB
floppies and Superdrives support the DOS-V format under Windows 2K and XP;
though FORMAT doesn't appear to understand how to initializae blanks.
..and let's not forget the short-lived "special" distribution formats from
IBM and Microsoft that got around 1.8MB per 3.5" floppy by using more
sectors per track+interleave+skew(MS DMF) or differing sector sizes (IBM
XDF). I suppose you could call those "standard", since they were
officially-endorsed formats.
Cheers,
Chuck
5:56 p.m.
New subject: Ah, the weirdities of PC formats
On Mon, 18 Sep 2006, Chuck Guzis wrote:
There were two "standard" versions of the
720K 3.5 format, differing in the
cluster and FAT size. One was the definition put out by IBM sometime
around MS-DOS 2.1 and the other, by Microsoft in MS-DOS 4.00 (which was an
abomination).
First official support for 3.5" in PC-DOS was V3.20.
First official support for 3.5" in MS-DOS (OEM'd) was V2.11.
A few machines added support for 3.5" on their own (HP, Gavilan, etc.)
Several versions of Windoze FORMAT (COMMAND LINE!) support /T:80/N:8
which gives a 640K format!
/F:x support seems to have been dropped completely.
9:40 p.m.
New subject: Ah, the weirdities of PC formats
On 9/18/2006 at 5:56 PM Fred Cisin wrote:
First official support for 3.5" in PC-DOS was
V3.20.
First official support for 3.5" in MS-DOS (OEM'd) was V2.11.
A few machines added support for 3.5" on their own (HP, Gavilan, etc.)
Any DOS (PC or MS) could support 720K with a DEVICE= driver after version
2.0. And if you were lucky enough to have the Microsoft OEM tech
bulletins, you could do it with DOS 1.x. As a matter of fact, didn't
Godbout initially offer support for PC-DOS by taking a stock $40 copy of
PC-DOS 1.0/1.1 and supplying his own IBMBIO.COM/IO.SYS? I seem to recall
some elaborate ritual with it.
I still have the OEM customization guides for MS-DOS 1.25 and 2.00. The
preface to 2.00 states that eventually it would support the entire set of
Xenix functions. Uh-huh.
But there was a glorious mixup in PC/MS 720K formats. I'm not sure, but
I think they even used different media bytes.
One could in PC-DOS 3.?0 use DRIVPARM in CONFIG.SYS but you had to suffix
the DRIVPARM with a string of control-A's.
Several versions of Windoze FORMAT (COMMAND LINE!)
support /T:80/N:8
which gives a 640K format!
/F:x support seems to have been dropped completely.
AFAIK, the code's still in the kernel driver, but FORMAT's just gotten
dumb, probably intentionally. And if you hook up a 360K 5.25" drive and
set the BIOS up for it, XP still won't format the blasted thing, although
it'll read and write just fine.
I think MS would love to kill off floppy support entirely.
Cheers,
Chuck
19 Sep
19 Sep
12:25 p.m.
New subject: Ah, the weirdities of PC formats
>First official support for 3.5" in PC-DOS was
V3.20.
>First official support for 3.5" in MS-DOS (OEM'd) was V2.11.
>A few machines added support for 3.5" on their own (HP, Gavilan, etc.)
On Mon, 18 Sep 2006, Chuck Guzis wrote:
Any DOS (PC or MS) could support 720K with a DEVICE=
driver after version
2.0.
I was thinking of support WITHOUT an additional device driver.
(DRIVPARM was NOT a device driver, and didn't occupy extra RAM)
I still have the OEM customization guides for MS-DOS
1.25 and 2.00. The
preface to 2.00 states that eventually it would support the entire set of
Xenix functions. Uh-huh.
any hope of a copy?
One could in PC-DOS 3.?0 use DRIVPARM in CONFIG.SYS
but you had to suffix
the DRIVPARM with a string of control-A's.
3.20
I think MS would love to kill off floppy support
entirely.
1:06 p.m.
New subject: Ah, the weirdities of PC formats
On 9/19/2006 at 12:25 PM Fred Cisin wrote:
I was thinking of support WITHOUT an additional device
driver.
(DRIVPARM was NOT a device driver, and didn't occupy extra RAM)
I suspect that the OS could easily be patched to accomodate 720K without a
device driver as early as PC-DOS 1.0.
any hope of a copy?
My paper file copy's a faded yellowed tractor-feed listing, but I'll dig
around and see if I can find the file it was printed from.
Anyone remember the procedure of getting stock PC-DOS to work with the
Compupro 85/88 CPU board?
Cheers,
Chuck
18 Sep
18 Sep
9:27 a.m.
You could?!?
(if I'm parsing your statement correctly). IIRC, the original
PC FDC card had IBM's proprietary data separator on it (several hybrid
packagaes) and couldn't read SD/FM to save its life. It also had the 8272
(non "A") part, so it had other problems with "alien" formats.
Many can. The original IBM PC version was a an abortion.
12:09 p.m.
Many can. The
original IBM PC version was a an abortion.
It was way worse than that.
Calling it "an abortion" is an affront to those who are Pro-Choice.
6670
days inactive
6671
days old
11 comments
4 participants
participants (4)
-
ajp166＠bellatlantic.net -
cclist＠sydex.com -
cisin＠xenosoft.com -
spectre＠floodgap.com