Query re. 8" hard sectored floppies (on eBay)
john...@hotmail.com
Just saw some for sale on eBay:
http://cgi.ca.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1249475799
and was owndering if all 8" floppy disks were hard sectored, or were
they normally soft sectored?
I have no connection to the seller!
Regards,
John
Larry Bledsoe,MD
Most used the IBM standard soft sectored, 256K bytes,SSSD.
It was almost like the 1.44 floppy is now. My favorite was/is the
DSDD 1.2 Meg/floppy.
I think Altair and IMSAI used the hard sectored 8 inchers. Heath and Northstar used hard sectored 5 inchers.
Larry
Barry Watzman
Joe
wrote:
>
>--------------96D4F300B49E0D17EFE8695F
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
>IMSAI used soft sector. Most 8" systems used soft sector disks. Hard
>sector disks were mostly used by proprietary word processing systems,
>and a few short-sighted companies that didn't WANT to be compatible with
>any other systems.
It had nothing to do with short sightedness. Also all of the first
disk drive systems used HS disks! Disk controller ICs had yet to be
invented and hard sectoring made sector timing a LOT easier,
especailly when you had to do everything with small scale ICs. In
addition, hard sectored disks generally held more data since they
didn't have to have leave room for sector addresses, preambles and
other SW overhead. Furthermore, interchangability of disks between say
a Tektronix 4051 and an IBM simply was not needed. Only after the
developement of "standard" operating systems like CPM did
interchangability become important because then applications programs
could be written to run on multiple systems and for that it was highly
desirable for there to be a common disk format.
Just to add the list, the Tektronix 4051, Altair and some of the
Data Generals also used hard sectored 8" disks.
Joe
Fritz Chwolka
> Just to add the list, the Tektronix 4051, Altair and some of the
> Data Generals also used hard sectored 8" disks.
>
> Joe
An the Zilog MC-Z series.
The MC-Z 15 / 20 and 30 used hard sectored 8" disks.
Greetings from
Fritz Chwolka / collecting old computers just for fun\
Charles E. Fox
>On Fri, 29 Jun 2001 20:59:24, rig...@intellistar.net (Joe) wrote:
>
>> Just to add the list, the Tektronix 4051, Altair and some of the
>> Data Generals also used hard sectored 8" disks.
>>
>> Joe
>
>An the Zilog MC-Z series.
>
>The MC-Z 15 / 20 and 30 used hard sectored 8" disks.
>
>
Also Philips Micom.
Thomas Fischer
>
>
>--------------B2BD49FCCB70C705A21C5156
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
>
>
>
(snip)
>> Regards,
>> John
>
>Most used the IBM standard soft sectored, 256K bytes,SSSD. It was
>almost like the 1.44 floppy is now. My favorite was/is the DSDD 1.2
>Meg/floppy.
>
>I think Altair and IMSAI used the hard sectored 8 inchers. Heath and
>Northstar used hard sectored 5 inchers.
>
>Larry
>
FWIW, IMSAI was never hard-sectored. It followed the IBM 3740 and later
soft-sector format(s) on 8" and 5.25" drive products.
-Thomas "Todd" Fischer http://www.imsai.net
>--------------B2BD49FCCB70C705A21C5156
>Content-Type: text/html; charset=us-ascii
>Content-Transfer-Encoding: 7bit
>
><!doctype html public "-//w3c//dtd html 4.0 transitional//en">
><html>
>
><p>john...@hotmail.com wrote:
><blockquote TYPE=CITE>Hi,
><p>Just saw some for sale on eBay:
><p><a
href="http://cgi.ca.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=1249475799">htt
p://cgi.ca.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&i
><p>and was owndering if all 8" floppy disks were hard sectored, or were
><br>they normally soft sectored?
><p>I have no connection to the seller!
><p>Regards,
><br>John</blockquote>
>
><p><br>Most used the IBM standard soft sectored, 256K bytes,SSSD.
>It was almost like the 1.44 floppy is now. My favorite was/is the
>DSDD 1.2 Meg/floppy.
><p>I think Altair and IMSAI used the hard sectored 8 inchers. Heath
>and Northstar used hard sectored 5 inchers.
><p>Larry</html>
>
>--------------B2BD49FCCB70C705A21C5156--
>
Michael J. Mahon
>: It had nothing to do with short sightedness. Also all of the first
>:disk drive systems used HS disks! Disk controller ICs had yet to be
>:invented and hard sectoring made sector timing a LOT easier,
>:especailly when you had to do everything with small scale ICs.
>
> Just so. Soft-sector disks need a complex controller to handle flags,
>address-marks, etc. That would be a whole PCB in discrete logic. It's
>quite possible (I did it) to run a hard-sectored disk direct from a
>Z80 (2.5MHz), with one CTC channel to count sectors and time head-step
>pulses, some PIO pins and half an SIO for the data. All else is
>software.
And Steve Wozniak figured out how to do it with a 256-byte PROM
and 6 TTL parts--using nothing else but software running on a 1MHz
6502! The use of the PROM as a 2MHz state machine with the head
signal going directly to an address pin is one of the most brilliant
hardware hacks that I know of!
He was motivated by the high cost of the Western Digital controller,
and by the desire to get better than single-density data capacity.
The 6-data-bits-in-8-on-the-disk GCR scheme he developed provided
143KB on single density 5.25" floppies, with good robustness and
error detection.
-michael
Email: mjm...@aol.com
Home page: http://members.aol.com/MJMahon/
Bill Marcum
Michael J. Mahon wrote in message
<20010701235851...@ng-md1.aol.com>...
and nowadays hardly anything except an Apple II can read Apple II disks.
Once single chip floppy controllers became available, everybody used them
except Apple and Commodore, and eventually even they switched to MFM.
Michael J. Mahon
Yes, but with minimal changes, and even a little less software complexity,
the same scheme could have read and written the lower-density IBM
standard (which was not nearly so "standard" then). It was a conscious
choice to go with GCR to obtain higher density, rather than implementing
the single density standard.
With another state machine ROM, an Apple II controller can read any
soft-sectored 5.25" disk with single density bit rates, regardless of track
format, so it stands as an existence proof of a simple, soft-sectored,
software controlled approach requiring only more cleverness than the
average bear.
I agree that this is academic, now that FD controllers are not only
pervasive and cheap, but practically obsolete! But at the time, it
was a major triumph of intellect over economics. ;-)
Clarence Wilkerson
>>>The Heathkit H-89 used HS disks for both CP/M and HDOS, as well.
>>>
>>>Bob Mason
The UCSD Pascal for H-89 also used hard sectored drives, ...
but
Heathkit only used HS disks in th 5.25" format. With DS QD
drives, one could actually get 440k on the HS 5.25" disk.
The H-89 had a couple of 8" drives ( H-47 and H-67 I think)
options, but these were soft-sectored formats ( and used
an intelligent controller in the drive package.
--
Clarence Wilkerson \ HomePage: http://www.math.purdue.edu/~wilker
Prof. of Math. \ Internet: wil...@math.purdue.edu
Dept. of Mathematics \ Messages: (765) 494-1903, FAX 494-0548
Purdue University, \
W. Lafayette, IN 47907-1395 \
B'ichela
<wil...@peano.math.purdue.edu> wrote:
>The H-89 had a couple of 8" drives ( H-47 and H-67 I think)
>options, but these were soft-sectored formats ( and used
>an intelligent controller in the drive package.
had a Heathkit H88 with those! (god do I miss that noisy box). Btw the
interface that connects the drives to the computer. was that a
variation of SASI? and what was that 34 pin connection on the back of
the H47 box for? More slave remex drives? how many could one REALLY
have anyway? the H-67/Z-67.. that I saw too, it had a 10Mb Hard drive in
it. 8 inch in size. the one I saw had a clear casing on the hard
drive assembly.
My beautiful H88 got toasted by a power surge (sigh...) anyone
know who still has one looking for a home. Sadly I got rid of all of
the parts to a CP/M pack rat, he was not able to build me another
working system from all of the parts laying in his cellar. (some
problems with ROMS and board mods or something along those lines.) All
I ask if I can get another is... Can someone please send me the
manuals on building/repairing the system? I especially loved the "flip
top" case style that mine had.
BTW the H88 did not have the 5 1/4" bay populated by a drive.
the plug for power was precent but the drive bay was covered by a
"dummy panel".
Speaking of the 8" soft sectored disks (double sided/double
Density) who still carries these inexpensivly? I wish I saved a fw so
I could show the local kids just how big floppies USED to be.
Clearly a H-47 and a H-88 or H-89 box with the needed CP/M
disks and manuals would be greatly appreciated, I am not interested
in the hard sectored 5 1/4" disks I have grown quite fond of the
big 8" media.
I have a surplus H-77 5 1/4" Dual Full Hight case used on my coco3
for TWo 20MB half hight Hard drives! those Linear Power Supply units
were BEASTY! lots of brute force! Turns the drives over without a
growl! Was the H-77 external cases used with the soft sectored or hard
sectored disk controllers? I got one of the original drives in the
case just being used as a space filler (the case has holes for 4
half hight drives! Loved those "garage doors".
Question of the H-47 box. Just how many Watts of power could
be provided for the drives? I know that the drives I had used a
120VAC 60 Cycle Syncronous motor (I am in the United States btw) 5VDC
and 24VDC for the drives? I would assume that one could easily power a
small array of half hight 5 1/4" floppys off one of those. Say perhaps
8? the Filter Cap ALONE was the size of a Oil Filter on a small car!
--
B'ichela
Axel Berger
BM>but it didn't follow the IBM standard, and nowadays hardly anything
BM>except an Apple II can read Apple II disks. Once single chip floppy
BM>controllers became available, everybody used them
Quite, but the question can and should be asked, whether a coding
system not requiring critical and dedicated specialized hardware is not
more reliable and better than any that does, even if and when that
hardware becomes cheap and ubiquitous.
We old have old disks with read errors and I don't know but suspect,
that Wozniaks way makes rescueing residual information more feasable
than the MFM way.
Lee Hart
hard-sectored 5-1/4" disks, never 8" hard-sectored. The hard-sector
5-1/4" disk controller was also very simple, with no controller chip. It
used a Signetic USART chip to serialize the R/W data, and a few TTL
gates for everything else. Software did most of the work.
B'ichela wrote:
> Ah yes! that dreaded H-47 box with those Remex drives
Yes. Remex seemed to think they were clever to make a precision disk
drive out of plastic and sheet metal. It didn't work. I could never keep
mine aligned and working for long.
> Btw the interface that connects the drives to the computer. Was that a
> variation of SASI? And what was that 34 pin connection on the back of
> the H47 box for?
Yes, it was a SASI interface, but crowded into a 40-pin ribbon cable
connector. The 40-pin connection on the H47 allowed up to 4 drives to be
connected (two H47 boxes, each with two 8" floppy drives).
> I had a Heathkit H88 with those! (god do I miss that noisy box)...
> All I ask if I can get another is... Can someone please send me the
> manuals on building/repairing the system?
I have several H89s; more than I can use. If you would like one to play
with, I can send you one for $25 plus shipping. It is complete and
working with the H17 hard sector controller and an internal disk drive.
I have all the manuals and other accessories you are likely to ever need
as well. Contact me directly for further details.
> BTW the H88 did not have the 5 1/4" bay populated by a drive.
> the plug for power was present but the drive bay was covered by a
> "dummy panel".
Or a disk-drive-sized bin that held about two boxes of floppies.
> Was the H-77 external cases used with the soft sectored or hard
> sectored disk controllers?
Both. The disk drives were the same; only the data format differed
depending on which disk controller you plugged it into.
> Question of the H-47 box. Just how many Watts of power could
> be provided for the drives? I know that the drives I had used a
> 120VAC 60 Cycle synchronous motor (I am in the United States btw) 5VDC
> and 24VDC for the drives? I would assume that one could easily power a
> small array of half hight 5 1/4" floppys off one of those.
The H47 used non-standard power and data connectors (a consequence of
the @#$^*! Remex drives), but you could install standard 8" drives in
the case with a little work. There was no -5v supply, which is needed by
some old 8" drives. You wouldn't want to put 5-1/4" drives in the H47
because it has a 24v supply, not 12v as is used by 5-1/4" drives.
--
Lee A. Hart Ring the bells that still can ring
814 8th Ave. N. Forget your perfect offering
Sartell, MN 56377 USA There is a crack in everything
leeahart_at_earthlink.net That's how the light gets in - Leonard Cohen
B'ichela
>Yes. Remex seemed to think they were clever to make a precision disk
>drive out of plastic and sheet metal. It didn't work. I could never keep
>mine aligned and working for long.
alligned. as did my Commodore 1541.
>Yes, it was a SASI interface, but crowded into a 40-pin ribbon cable
>connector. The 40-pin connection on the H47 allowed up to 4 drives to be
>connected (two H47 boxes, each with two 8" floppy drives).
thing too. Big Noisy and HEAVY! ;)
>I have several H89s; more than I can use. If you would like one to play
>with, I can send you one for $25 plus shipping. It is complete and
>working with the H17 hard sector controller and an internal disk drive.
just want the Big Monster 8" disks and a matching computer and H47
drive box.
>> BTW the H88 did not have the 5 1/4" bay populated by a drive.
>> the plug for power was present but the drive bay was covered by a
>> "dummy panel".
>
>Or a disk-drive-sized bin that held about two boxes of floppies.
liked it the way it was. But then a 8" disk I doubt would fit in
there Unless one FOLDED them ;)
>> Was the H-77 external cases used with the soft sectored or hard
>> sectored disk controllers?
>
>Both. The disk drives were the same; only the data format differed
>depending on which disk controller you plugged it into.
door is broken. I just use it for a "spacer" as the two MFM hard
drives fit in one of the bays. I use it for my Coco-XT Burke and
Burke-RTC hard drive interface.
>The H47 used non-standard power and data connectors (a consequence of
>the @#$^*! Remex drives), but you could install standard 8" drives in
>the case with a little work. There was no -5v supply, which is needed by
>some old 8" drives. You wouldn't want to put 5-1/4" drives in the H47
>because it has a 24v supply, not 12v as is used by 5-1/4" drives.
regulator chip to bring the 24vdc down to 12vdc :) But still... Just
how much "Juice" can one of these brats provide?
Speaking of those Remex drives. Was there any after market
replacement kits to install a more generic 8" unit inside this box?
--
B'ichela
Richard Erlacher
was an error, without any inquiry as to whether to retry or what.
I seriously doubt that anything other than cost was considered by the
Apple folks. The cost reduction that Woz managed in the controller
and drive did give them a tremendous advantage, and that was what they
cared about. Unfortunately, they really didn't put enough reliability
into their OS to make it possible to do anything other than reboot if
there was an error.
Any rigorous encoding scheme allows for recovery of data if enough is
known about its failure modes. I doubt that Woz' GCR was any less or
any more reliable than FM or MFM implemented with a controller IC.
He (Woz) was clever enough to get the job done, though, and at a time
when the other "major" competitors, e.g. Radio Shack, etc, were still
struggling to get DD recording on their media to work at all.
Dick
Lee Hart
>> I have several H89s... I can send you one for $25 plus shipping.
>> It is complete and working with the H17 hard sector controller
>> and an internal disk drive.
B'ichela wrote:
> No thanks. I don't want to deal with hard sectored floppies. I
> just want the Big Monster 8" disks and a matching computer and
> H47 drive box.
I have the H47 interface boards, but no H47 or H67 boxes. But since it
was really a SASI interface, there were many other things people could
use. Several vendors made hard drives and 8" floppy drives for the H89
that used the H47 or H67 interface board in the H89, and some drive
controller board (DTC and OMTI in particular). A typical setup would
have the H47/H67 card in the H89, and a DTC controller, 5-1/4" hard
drive, and 5-1/4" floppy drive in an external H77 box.
> But still... Just how much "Juice" can one of these brats provide?
OK, I looked it up in the Heath H47 manual. It has a "360 watt" power
supply according to the specs! But checking the schematic reveals this
means it has a 3 amp fuse and runs on 120vac. The +24v and +5v
regulators are rated at 5 amps; they probably limit the actual power
output.
> Speaking of those Remex drives. Was there any after market
> replacement kits to install a more generic 8" unit inside this box?
Yes. The Remex "slave" drive was an ordinary 8" floppy drive with
non-standard power and data connectors. The Remex "master" could drive
one master and 3 slaves. So there were kits that kept the "master" logic
board, but replaced the two physical Remex drives with generic 8"
Michael J. Mahon
>I'm not sure I buy that one. The Apple way was to go TILT if there
>was an error, without any inquiry as to whether to retry or what.
DOS 3.3 automatically retried about 2x10 times, then recalibrated
the head and retried some more. I have never seen a case where
additional retries would (correctly) recover a sector not recovered
by this procedure. In other words, soft errors are handled pretty
well, and hard errors are well detected.
There were numerous patches to temporarily "relax" the error
checking, to permit grabbing a bad sector for manual data recovery.
>I seriously doubt that anything other than cost was considered by the
>Apple folks. The cost reduction that Woz managed in the controller
>and drive did give them a tremendous advantage, and that was what they
>cared about. Unfortunately, they really didn't put enough reliability
>into their OS to make it possible to do anything other than reboot if
>there was an error.
Cost was certainly key, but density was, too. The original release
of the controller ROM and software supported 13 256-byte sectors
per track. A little over a year later, DOS 3.3 was released together
with a ROM upgrade that upped that to 16 sectors per track. This
resulted in 143K usable bytes on a single-sided, single-density disk,
about 50% more than the normal FM single-density would give.
An extra cost saver was eliminating all the seek logic from the
disk controller board. The only thing left in the disk drive is the
motor speed board, the analog R/W electronics, and the seek
stepper drive transistors!
>Any rigorous encoding scheme allows for recovery of data if enough is
>known about its failure modes. I doubt that Woz' GCR was any less or
>any more reliable than FM or MFM implemented with a controller IC.
In fact, since the formatting and format recognition was all in software,
Woz could have a more robust code than the usual hardware controller
would provide. For example, address marks and data marks are delimited
by an 8-bit code on the disk which can never occur in data or elsewhere
in marks. (There are 66 valid 8-bit codes, of which 64 encode 6-bits of
data, leaving two for unique address and data marks.) After the marks,
there are two additional required marker bytes prior to address/data.
And following the address/data field, there is an 8-bit checksum byte
and two epilogue bytes. This is a pretty robust coding.
Since it is all in software, many game manufacturers made "modifications"
to the codes to create not-easily-copyable disks (which, of course were
all ultimately copyable ;-). Many of these modifications reduced the
level of redundancy or speed- or mechanism-tolerance of the disk, relative
to the native format, but even these "weakenings" were typically quite
reliable in practice.
B'ichela
>I have the H47 interface boards, but no H47 or H67 boxes. But since it
>was really a SASI interface, there were many other things people could
>use. Several vendors made hard drives and 8" floppy drives for the H89
>that used the H47 or H67 interface board in the H89, and some drive
I conld connect my Adaptec ACB-4070 bridgeboard to connect two MFM
hard drives to this unit. How hard would it be to use one of those on
a H47 host adaptor... My board is SASI/SCSI compatible. I have the
full manual on the board also. Right now it is connected to my
486DX2/66mhz Linux box. (I know Legacy hardware on legacy computers
(NOT!) ;)
>> But still... Just how much "Juice" can one of these brats provide?
>OK, I looked it up in the Heath H47 manual. It has a "360 watt" power
>supply according to the specs! But checking the schematic reveals this
>means it has a 3 amp fuse and runs on 120vac. The +24v and +5v
>regulators are rated at 5 amps; they probably limit the actual power
>Yes. The Remex "slave" drive was an ordinary 8" floppy drive with
>non-standard power and data connectors. The Remex "master" could drive
>one master and 3 slaves. So there were kits that kept the "master" logic
>board, but replaced the two physical Remex drives with generic 8"
>drives.
drives/kits cables (oh my!) got any you want to dump with a working
H47 host adaptor and the needed H89 "mods" to support it? including a
boot disk.
>--
>Lee A. Hart Ring the bells that still can ring
>814 8th Ave. N. Forget your perfect offering
>Sartell, MN 56377 USA There is a crack in everything
>leeahart_at_earthlink.net That's how the light gets in - Leonard Cohen
>
>
--
B'ichela
Lee Hart
> If someone made a 40pin to 50pin SCSI-1 centronics cable.
> hard drives to this unit. How hard would it be to use one of those on
> a H47 host adaptor...
Hackers just sliced some wires out of the middle of the 50-pin SASI flat
cable, and moved the two edges inward to fit a 40-pin connector. The
actual circuit is shown in the Heath schematics (Heath did it with a
little PC board mounted to the back of the H47/H67).
The stock Heath boot ROMs and stock Heath CP/M had the code to talk to
the Remex, DTC, and OMTI SASI controller boards. The only drawback was
that it only supported 1 or 2 single/double density single/double sided
8", and one 10megabyte hard drive. If you don't have these drives, then
you have to modify the boot ROMs, CP/M BIOS, FORMAT, and MOVCPM
utilities to get it all to work.
> Ok, now the question for the tinkerer with some of these
> drives/kits cables (oh my!) got any you want to dump with a working
> H47 host adaptor and the needed H89 "mods" to support it? including a
> boot disk.
As I say, I have the H89s, H47 interface cards, and Heath CP/M. How
about $35 plus shipping?
Barry Watzman
they were architecturally unique and there is no substitute. They had a disk
controller on the drive, and talked to the host through a host adapter card
and what amounted to a parallel port. It was very much like SASI / SCSI, but
it wasn't either, and since the controller was actually in the drive and not
separate, and since the drives themselves were not even the same (there was a
"master" drive and a "slave" drive), there is no way to substitute something
else easily.
The only good thing that came out of this fiasco was the H-89-37 controller,
but it was too little too late. And because of the slow CPU speed of the H89,
there was no way to do double density 8" in a stock H89 with out resorting to
some form of external controller.
Barry Watzman
Clarence Wilkerson
H89 are "dumb", in the sense that the software has to
to generate the handshakes. On the other hand, this doesn't
seem to be time critical. I have done programs in Turbo
Pascal to exercise the interface and drives. I've used
the H-67 interface board with several different "bridge
boards", and the software modifications were pretty minor.
Basic read/write were std., but setting disk parameters seemed
to vary a lot from one bridge board to another brand.
My memory is that the DTC 510 bridge to 5.25" MFM hard
drivers was almost a drop-in
replacement software wise for the DTC board in the
H-67 box that contolled
the original 8" hard drive.
If you want to get an idea of what's involved, there's
a copy of a later Heath CP/M bios in source form with
the drivers for the H-47 and H-67. Quite a bit simpler
than the floppy code!
B'ichela
>As I say, I have the H89s, H47 interface cards, and Heath CP/M. How
>about $35 plus shipping?
and on a fixed income with a $350 vet bill to pay.
However perhaps someday when I get back on my feet.
The vet bill was for my cat that was chomped on by a pit bull.
He is fine, Dog attacked in self defense as my cat attacked first.
--
B'ichela
Jeff Jonas
>> Data Generals also used hard sectored 8" disks.
>The MC-Z 15 / 20 and 30 used hard sectored 8" disks.
I guess I was kinda late on the scene:
the Zilog MCZ2/50 used soft sectored 8" floppy disks.
I was suckered to trying RIO and RIO/CP with the PL/Z language.
Life was better when we upgraded to CP/M and "C".
Sadly, I didn't know back then that the DSDD format was non-standard
so I have disks that are hard to read due to the Zilog strange
sector interleaving.
Back in college, we had a Floppy Flinging contest:
build a machine to fling a Vydec Exxon Office Systems floppy the furthest.
That word processor / office system used 8" floppy disks with
hard sectors along the outer edge and a unique envelope with
notched sides to prevent wrong insertion!
--
Jeffrey Jonas
jeffj@panix(dot)com
The original Dr. JCL and Mr .hide
dwight elvey
Wow, I wish I'd looked at this message earlier.
I could have used these on my Nicolet 80.
It is a rare machine that uses hard sectored
but the Nicolet is a rare machine.
Dwight