8 Feb
2015
8 Feb
'15
4:50 p.m.
I am wondering if there were any magnetic tape filesystem?
I know that some micro you could give a name to your data/program on the
audio cassette and they had a catalogue feature
I also know about Linear Tape File System (LTFS)
Has anyone tried putting cp/m or fat filesystem on audio cassette?
8 Feb
8 Feb
5:16 p.m.
This may not be close enough to what you're looking for, but DEC put the RT11
filesystem on TU58 tapes. They could be used just like very slow floppies. The tape drive
took care of locating logical blocks on the tape, essentially providing a block type
device to the computer.
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
9 Feb
9 Feb
2:29 a.m.
On 2015-02-09 02:16, Mark J. Blair wrote:
This may not be close enough to what you're
looking for, but DEC put the RT11 filesystem on TU58 tapes. They could be used just like
very slow floppies. The tape drive took care of locating logical blocks on the tape,
essentially providing a block type device to the computer.
Very different.
The TU58 behaves like a disk, not a tape.
Anyway, ANSI is a "file system" for tapes. A "normal" file system will
be very hard, since that would require the ability to rewrite existing
blocks in the middle of the tape, which is a scary thing to try.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
12:26 p.m.
On 09/02/15 20:29, Johnny Billquist wrote:
On 2015-02-09 02:16, Mark J. Blair wrote:
that got me thinking about these Log-structured file system[1]
but all of them are too modern, I have not seen one designed for a 8-bit
CPU like the 6502 and they flash memory based :(
[1] http://en.wikipedia.org/wiki/Log-structured_file_system
This may not be close enough to what you're
looking for, but DEC put
the RT11 filesystem on TU58 tapes. They could be used just like very
slow floppies. The tape drive took care of locating logical blocks on
the tape, essentially providing a block type device to the computer.
Very different.
The TU58 behaves like a disk, not a tape.
Anyway, ANSI is a "file system" for tapes. A "normal" file system
will
be very hard, since that would require the ability to rewrite existing
blocks in the middle of the tape, which is a scary thing to try.
Johnny
8 Feb
8 Feb
5:18 p.m.
On 02/08/2015 04:50 PM, Tom Sparks wrote:
I am wondering if there were any magnetic tape
filesystem?
I know that some micro you could give a name to your data/program on the
audio cassette and they had a catalogue feature
I also know about Linear Tape File System (LTFS)
Has anyone tried putting cp/m or fat filesystem on audio cassette?
Magtape filesystems have been around since the 1950s at least, and
possibly the 50s. "ANSI tape labels" not only define file names, but
all of the other information you'd expect (system, creation date, file
name, etc.).
http://en.wikipedia.org/wiki/Tape_label
Problem arose when micro- and mini-computer software developers didn't
realize that a standard had existed for years. Some mini makers did
implement them, but by no means all.
--Chuck
6:03 p.m.
On 02/08/2015 05:50 PM, Jules Richardson wrote:
On 02/08/2015 07:18 PM, Chuck Guzis wrote:
Yeah, I had one of those--just an ISA card with some connectors. It was
S-L-O-W and not terribly reliable. I don't recall who made it.
--Chuck
Magtape filesystems have been around since the
1950s at least, and
possibly
the 50s.
Wasn't there at least one which used good ol' VHS video tape as the
storage medium? And by good, I mean bad. 
6:18 p.m.
Yeah, I had one of those--just an ISA card with some
connectors. It
was S-L-O-W and not terribly reliable. I don't recall who made it.
--Chuck
I remember reading a review of one back in the day. They were slow but
didn't they have massive (for then) capacity. Something like 1GB? I actually
saw a controller on sale on eBay a while back but it didn't have any SW do I
passed on it. It would have been cool to play with it if it was complete.
-Ali
6:28 p.m.
On 2/8/2015 6:03 PM, Chuck Guzis wrote:
On 02/08/2015 05:50 PM, Jules Richardson wrote:
Alpha Micro made a card that did backups to VHS.
Corvus had a thing called the bank, and I think they also had a gizmo to
do the tape backup to video tape. The Bank device was just a product
they did for whatever reason.
I have a case of Bank tapes somewhere and a backup device. The
interface was a Corvus network connection.
Jim
On 02/08/2015 07:18 PM, Chuck Guzis wrote:
Yeah, I had one of those--just an ISA card with some connectors. It
was S-L-O-W and not terribly reliable. I don't recall who made it.
--Chuck
Magtape filesystems have been around since the
1950s at least, and
possibly
the 50s.
Wasn't there at least one which used good ol' VHS video tape as the
storage medium? And by good, I mean bad.
6:54 p.m.
On 2/8/2015 6:28 PM, jwsmobile wrote:
On 2/8/2015 6:03 PM, Chuck Guzis wrote:
Following up with a link to some book with a reference about AMOS tape
backup:
https://books.google.com/books?id=F-Y0wdIIucgC&pg=PA63&lpg=PA63&…
Not sure how to get this mangled mess thru to people from the google
books link, but here is the search and the book reference:
alpha micro tape backup
"Encyclopedia of Microcomputers: Volume 1 - Access Methods to Assembly
Language and Assemblers", Allen Kent, James G. Williams October 1, 1987
http://www.crcpress.com/product/isbn/9780824727000
And a monument to the ridiculous way that books are priced, CRC
publishing wants $306.90US for a copy.
thanks
Jim
On 02/08/2015 05:50 PM, Jules Richardson wrote:
Alpha Micro made a card that did backups to VHS.
Corvus had a thing called the bank, and I think they also had a gizmo
to do the tape backup to video tape. The Bank device was just a
product they did for whatever reason.
I have a case of Bank tapes somewhere and a backup device. The
interface was a Corvus network connection.
Jim
On 02/08/2015 07:18 PM, Chuck Guzis wrote:
Yeah, I had one of those--just an ISA card with some connectors. It
was S-L-O-W and not terribly reliable. I don't recall who made it.
--Chuck
Magtape filesystems have been around since the
1950s at least, and
possibly
the 50s.
Wasn't there at least one which used good ol' VHS video tape as the
storage medium? And by good, I mean bad. 
9 Feb
9 Feb
4:40 a.m.
On Sun, 8 Feb 2015, jwsmobile wrote:
Alpha Micro made a card that did backups to VHS.
I had one of those for ISA bus. Sellam bought it from me about ten years
ago. Hope it didn't end up with the vultures who cleaned out his storage
building.
Corvus had a thing called the bank, and I think they
also had a gizmo to do
the tape backup to video tape. The Bank device was just a product they did
for whatever reason.
I have a case of Bank tapes somewhere and a backup
device. The interface was
a Corvus network connection.
The Bank used an oddball proprietary cartridge that operated in a
continuous loop like an 8-track tape.
I have three working (last time I checked) Bank drives and a number of
100MB capacity tapes. They also made a 200MB tape, but I've never been
able to get my hands on one.
A friend of mine once claimed that the Bank tape and transport was
originally designed by Akai as their entry into the VCR market. If true,
this would be the first analog video format I've ever heard of that
switched tracks at the end the loop :-).
--
7:29 a.m.
The MCM/70, an 8008 based APL machine released in late 1974, used dual
cassette drives as mass storage and swapping device. An amazing engineering
feat for the time and technology. Worked well, albeit s-l-o-w-l-y.
http://www.cse.yorku.ca/~zbigniew/mcm_add.html
Lee Courtney
On Mon, Feb 9, 2015 at 4:40 AM, Steven Hirsch <snhirsch at gmail.com> wrote:
On Sun, 8 Feb 2015, jwsmobile wrote:
Alpha Micro made a card that did backups to VHS.
I had one of those for ISA bus. Sellam bought it from me about ten years
ago. Hope it didn't end up with the vultures who cleaned out his storage
building.
Corvus had a thing called the bank, and I think they also had a gizmo to
> do the tape backup to video tape. The Bank device was just a product they
> did for whatever reason.
I have a case of Bank tapes somewhere and a backup device. The interface
> was a Corvus network connection.
The Bank used an oddball proprietary cartridge that operated in a
continuous loop like an 8-track tape.
I have three working (last time I checked) Bank drives and a number of
100MB capacity tapes. They also made a 200MB tape, but I've never been
able to get my hands on one.
A friend of mine once claimed that the Bank tape and transport was
originally designed by Akai as their entry into the VCR market. If true,
this would be the first analog video format I've ever heard of that
switched tracks at the end the loop :-).
--
--
Lee Courtney
+1-650-704-3934 cell
10 Feb
10 Feb
10:05 a.m.
On 9 February 2015 at 03:28, jwsmobile <jws at jwsss.com> wrote:
Alpha Micro made a card that did backups to VHS.
AFAIK that was the standard software distribution format for the
machines. I never did much AM support but my 1st job was at a company
which, along with IBM, Apple, Compaq & other more mainstream makes,
also resold Alpha Micro kit. Fascinatingly weird -- a sort of tiny,
rather proprietary minicomputer, based around the 68000 and
microcomputer technology.
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Cell/Mobiles: +44 7939-087884 (UK) ? +420 702 829 053 (?R)
9 Feb
9 Feb
3:51 a.m.
On Sun, Feb 08, 2015 at 07:50:54PM -0600, Jules Richardson wrote:
[...]
Wasn't there at least one which used good ol'
VHS video tape as the storage
medium? And by good, I mean bad.
For the Amiga, there was the "Video Backup System":
http://www.bigbookofamigahardware.com/bboah/product.aspx?id=1456
It was an "ingenious" hack in that it generated a monochrome video signal that
was a valid(-ish) serial stream could be hard-clipped and fed back in through
the serial port. I've never seen it work successfully though.
In a similar vein of misusing the interpretation of signals to produce cute
effects, I offer this which I discovered last night:
https://www.youtube.com/watch?v=J1ylMJwfLmM
Now, there are a number of vector demos which run on oscilloscopes in X-Y mode,
but that one is a bit more special than most in that it's not just some
beam-drawing plus some random music, but a reasonable chiptune that can *also*
be fed into a scope to do vector graphics.
Non-compressed audio can be downloaded from here:
http://www.pouet.net/prod.php?which=58317
... although I couldn't discern any real difference on my scope between that
and the Vorbis-compressed audio in the Youtube video.
7:39 a.m.
On 02/08/2015 07:50 PM, Jules Richardson wrote:
On 02/08/2015 07:18 PM, Chuck Guzis wrote:
I picked up a Digi-Data VHS drive with Pertec formatted
interface at a surplus
shop. It ALMOST worked, but was pretty clear the heads were
badly worn.
I did a couple backups with it, then it basically just
stopped working at all,
went to infinite retries.
Jon
Magtape filesystems have been around since the
1950s at
least, and possibly
the 50s.
Wasn't there at least one which used good ol' VHS video
tape as the storage medium? And by good, I mean bad.
7:48 a.m.
On Mon, Feb 09, 2015 at 09:39:40AM -0600, Jon Elson wrote:
On 02/08/2015 07:50 PM, Jules Richardson wrote:
> On 02/08/2015 07:18 PM, Chuck Guzis wrote:
>> Magtape filesystems have been around since the 1950s at
>> least, and possibly
>> the 50s.
I'm just surprised no one mentioned stringy floppy.
http://www.trs-80.org/exatron-stringy-floppy/
Diane
--
- db at FreeBSD.org db at db.net http://www.db.net/~db
9:01 a.m.
On 02/09/2015 07:48 AM, Diane Bruce wrote:
I'm just surprised no one mentioned stringy
floppy.
http://www.trs-80.org/exatron-stringy-floppy/
I did--twice.
Just didn't mention the awful poetry, as in
"I love my stringy floppy.."
From an early Exatron newsletter.
--Chuck
8 Feb
8 Feb
5:19 p.m.
On 2/8/2015 4:50 PM, Tom Sparks wrote:
I am wondering if there were any magnetic tape
filesystem?
I know that some micro you could give a name to your data/program on the
audio cassette and they had a catalogue feature
I also know about Linear Tape File System (LTFS)
Has anyone tried putting cp/m or fat filesystem on audio cassette?
I am sure you are thinking for tapes in smaller computers, but the
Univac III originally had only UniServo tape drives, and used tapes for
the OS. All tapes written had the OS written on the front end of the tape.
There were several variants of the OS until disk units were added, and
then the production run ended.
Most tape sets were written with specific jobs as the target, and the
tapes had instruction decks added after the OS portion to control the
run of the job. However most of the tapes had the OS written at the
front of the tape, since many operations required rereading the OS in
various incarnations, and no single tape was tied up being the tape that
did all the OS reads.
Systems would have many drives. I believe the one I have has 6 or 8
drives, as an example in addition to using punched card for input.
thanks
Jim
6:02 p.m.
Disks and drums were a luxury on old systems; there were many versions
of tape-only operating system. Even S/360 had TOS.
Anyone know of a port of IBSYS or FMS to modern gear?
One of the problems of using tape as a working medium is that, with a
few exceptions, rewriting of data in place can be problematical. (with
tape speed variations, you aren't guaranteed that overwriting a block on
tape will occupy the same amount of tape as the old one.
So you're essentially limited to skipping to a filemark, then creating a
new end of information by writing after that.
However, many mainframe systems automatically archived inactive files to
tape, retrieving them back to disk when requested. There were tape
robots for regular 10.5" reels, as well as interesting mechanisms for
storing data on special tapes, e.g. IBM 3850, 2321.
Archive and a few other PC QIC-based systems allowed for block overwrite
and could simulate a disk, albeit pretty slowly.
And then there's the Exatron Stringy Floppy.
--Chuck
10 p.m.
On Sun, Feb 8, 2015 at 5:50 PM, Tom Sparks <tom_a_sparks at yahoo.com.au> wrote:
I am wondering if there were any magnetic tape
filesystem?
[by which you presumably mean true random-access tape, rather than
e.g. 9-track which is normally append-only]
Yes. Obvious examples are DECtape/LINCtape. Different operating
systems used different file systems, but they generally appeared to
software as a small, slow disk.
Has anyone tried putting cp/m or fat filesystem on
audio cassette?
Most instances of tape file systems I've ever dealt with used a file
system specifically designed for efficiency on tape. For instance,
the TOPS-10 DECtape filesystem normally used every nth block for a
file, rather than consecutive blocks. This is somewhat like sector
interleave, except that it's done at the filesystem level rather than
being hidden by either hardware or a low-level block device driver.
Some members of the 6502 Group (established 1975, still meeting
weekly) developed the XPL0 programming language (similar to ALGOL and
BCPL), and an operating system partially written in XPL0, called TFS
("Tape File System") for use on homebrew or Digital Group 6502-based
systems with digital cassette drives. TFS eventually evolved into FFS
("Floppy File System"), and then APEX, which was sold for use on the
Apple II by Apparat (better known for NewDOS/80).
11:44 p.m.
On 09/02/15 16:00, Eric Smith wrote:
On Sun, Feb 8, 2015 at 5:50 PM, Tom Sparks
<tom_a_sparks at yahoo.com.au> wrote:
I am wondering if there were any magnetic tape
filesystem?
[by which you presumably mean true random-access tape, rather than
e.g. 9-track which is normally append-only]
Yes. Obvious examples are DECtape/LINCtape. Different operating
systems used different file systems, but they generally appeared to
software as a small, slow disk.
Has anyone tried putting cp/m or fat filesystem
on audio cassette?
Some members of the 6502 Group (established 1975,
still meeting
weekly) developed the XPL0 programming language (similar to ALGOL and
BCPL), and an operating system partially written in XPL0, called TFS
("Tape File System") for use on homebrew or Digital Group 6502-based
systems with digital cassette drives. TFS eventually evolved into FFS
("Floppy File System"), and then APEX, which was sold for use on the
Apple II by Apparat (better known for NewDOS/80).
I think I found what you are talking about at
http://bytecollector.com/dg_phideck.htm
tom
9 Feb
9 Feb
12:09 a.m.
I wrote:
> Some members of the 6502 Group (established 1975,
still meeting
> weekly) developed the XPL0 programming language (similar to ALGOL and
> BCPL), and an operating system partially written in XPL0, called TFS
> ("Tape File System") for use on homebrew or Digital Group 6502-based
> systems with digital cassette drives. TFS eventually evolved into FFS
> ("Floppy File System"), and then APEX, which was sold for use on the
> Apple II by Apparat (better known for NewDOS/80).
Tom Sparks <tom_a_sparks at yahoo.com.au> wrote:
I think I found what you are talking about at
http://bytecollector.com/dg_phideck.htm
Actually not. While some 6502 Group members did use Digital Group
6502 systems, the majority of the FFS work was done with surplus
digital tape drives with bit-banged manchester encoding, not with
Phideck transports or the Digital Group GCR controller.
I've never heard of anyone writing a 6502 or 6800 operating system for
the Digital Group Phideck subsystem. There was Phimon for the 8080,
which of course would also run on the Z80.
I was in junior high at the time, and couldn't afford a computer of my
own of any sort, though I drooled over the Digital Group brochures. In
the late 1990s I purchased a used Digital Group Z80 system with a
Phideck subsystem, and a 6502 Group member gave me his Digital Group
6501 (not 6502!) CPU card. (It was the Digital Group 6800 CPU card
with a MCS6501 installed; I don't know whether the Digital Group
originally sold it that way, or whether my friend replaced the CPU
chip and reprogrammed the 1702 monitor EPROM.) It might be
interesting to try to get Apex to use the Phideck subsystem, but in
general I have better or at least more interesting things to do in my
"copious free time".
12:36 a.m.
TI 99/4A initially presented the "wafertape" as an option
(block-addressable-continuous loop).
And, as mentioned, the stringy floppy, somewhat similar to the later
wafertape and somewhat popular with the TRS-80 Model 1 crowd.
The Coleco ADAM used a block-addressable tape, but I don't know what the
read-after-write capability of the HE format tapes was.
There's no reason that a 7- or 9-track tape could not be used as a
random-access device, provided an adequate IRG was allowed for.
Implementing read/write random-access filesystems on tape presents some
significant issues.
The first is the read-after-write physical block length problem, already
discussed.
The second is that extending files gets to be really interesting, unless
file extensions get appended to the end-of-information point.
The third is that it's pretty much impossible to deal gracefully with
tape block flaws that develop some time after the block has been
written. In particular, the usual 9-track method of erasing over the
bad spot and then writing a new block would create a problem by
overwriting already-written data.
Streamer drives are singularly awkward (although as I mentioned, one of
the QIC vendors implemented a "virtual disk" system for the PC (Archive,
Seagate...I don't recall). Looking at the HP 35480 DDS OEM manual shows
that the drive has to construct a new group from old data as well as
new, write a temporary copy, then back up and attempt to rewrite the old
data with the new. Ugly stuff.
--Chuck
1:25 a.m.
Streamer drives are singularly awkward (although as I
mentioned, one of
the QIC vendors implemented a "virtual disk" system for the PC
(Archive, Seagate...I don't recall
There were a couple of solutions for the PC:
1. Direct Tape Access by Seagate (which was later sold off to VERITAS). If I recall
correctly this was a Windows only solution and worked under win 3.x/Win 95. There never
was an update for Win98 and on.
2. TapeDisk - This worked under DOS (was a DOS device driver) and so continued to work,
with updates, through Win Me. The publishers were perpetually going to offer a Win2K/NT
version which never materialized and the company finally went bankrupt.
3. DATMan - Was limited to DAT drives so never gained any traction (if you could even say
the above SW had traction). This is made by the same guys as XXCopy and I think they are
still around.
I am sure I am missing a few here and there. I used both DTA and TapeDisk. DTA was easier
and more "elegant" but never updated. I switched over to TapeDisk when I got a
Conner/Seagate T-Stor 8000 4GB of "Random Access" Whoohoo! ;)
-Ali
3:18 a.m.
The Sinclair QL which was also sold as the ICL One Per Desk also had a serial tape type
media the ZX MicroDrive?
http://en.wikipedia.org/wiki/ZX_Microdrive
Dave
G4UGM
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Chuck
Guzis
Sent: 09 February 2015 08:37
To: General at classiccmp.org; Discussion at classiccmp.org:On-Topic and Off-
Topic Posts
Subject: Re: Magnetic tape filesystem
TI 99/4A initially presented the "wafertape" as an option (block-addressable-
continuous loop).
And, as mentioned, the stringy floppy, somewhat similar to the later
wafertape and somewhat popular with the TRS-80 Model 1 crowd.
The Coleco ADAM used a block-addressable tape, but I don't know what the
read-after-write capability of the HE format tapes was.
There's no reason that a 7- or 9-track tape could not be used as a random-
access device, provided an adequate IRG was allowed for.
Implementing read/write random-access filesystems on tape presents some
significant issues.
The first is the read-after-write physical block length problem, already
discussed.
The second is that extending files gets to be really interesting, unless file
extensions get appended to the end-of-information point.
The third is that it's pretty much impossible to deal gracefully with tape block
flaws that develop some time after the block has been written. In particular,
the usual 9-track method of erasing over the bad spot and then writing a new
block would create a problem by overwriting already-written data.
Streamer drives are singularly awkward (although as I mentioned, one of the
QIC vendors implemented a "virtual disk" system for the PC (Archive,
Seagate...I don't recall). Looking at the HP 35480 DDS OEM manual shows
that the drive has to construct a new group from old data as well as new,
write a temporary copy, then back up and attempt to rewrite the old data
with the new. Ugly stuff.
--Chuck
8 Feb
8 Feb
10:28 p.m.
On Mon, Feb 09, 2015 at 10:50:32AM +1000, Tom Sparks wrote:
I am wondering if there were any magnetic tape
filesystem?
I have a vague memory of hearing about something (one of the Unices maybe?
although handling tapes well is WAY out of character for them) that would
attempt to use 9-track tapes as block-replaceable media. Which I think
meant using the "erase" command to add 3" of nothing after each block so
that you had at least some chance of rewriting that block w/o trashing what
comes after (this would cause a huge cut in capacity, but maybe I'm
misremembering the whole thing). I've never seen this in person though,
and I'm a little surprised that it would work (maybe only on certain
drives?).
And as others have said, that's on top of intentionally block-replaceable
tapes like DECtapes (which act like disks), and file systems that accept
the tapiness of it all like ANSI or DOS-11 or CAPS-11 or OS-labeled tapes
(where you can write a fresh tail onto the tape from anywhere but you'd
better have a plan). And almost everybody's got something like BRU or
:DUMP or SAVRES or tar which can store/retrieve archives of disk files on
tape -- but that's not the same as accessing tape files through the OS's
usual disk-file API.
John Wilson
D Bit
9 Feb
9 Feb
7:56 a.m.
On 02/09/2015 12:28 AM, John Wilson wrote:
On Mon, Feb 09, 2015 at 10:50:32AM +1000, Tom Sparks
wrote:
Hmmm, that's interesting. At least on a lot of minicomputer
tape controls, I THINK
this would be impossible. There's no way to make it move
tape with the write gate
on without generating bit clocks, which would write
transitions on some channel
of the tape. I'm pretty sure the DEC TM11 and similar units
couldn't do this. (I suppose maybe
there was such a mode in the driver/hardware that I'm
totally unaware of.)
Some other systems may have had specific erase commands or
options that could
alter the length of the tape gap. In fact, I think the
standard interrecord gap
at 800 BPI NRZI WAS 3 inches. At 1900 BPI PE, the standard
gap was reduced
to 0.6 inches to improve capacity. I figured out that I
could write tapes at
800 BPI on my CP/M system with a gap around 0.5" and other
systems had
no problem reading them. On those little 7" reels, you need
to work to retain
what little capacity there was. (No replacement of
individual blocks was ever
attempted on that system.)
Jon
I am wondering if there were any magnetic tape
filesystem?
I have a vague memory of hearing about something (one of the Unices
maybe?
although handling tapes well is WAY out of character for them) that would
attempt to use 9-track tapes as block-replaceable media. Which I think
meant using the "erase" command to add 3" of nothing after each block so
that you had at least some chance of rewriting that block w/o trashing what
comes after 
8 Feb
8 Feb
10:54 p.m.
On 9 Feb 2015, at 11:50 am, Tom Sparks
<tom_a_sparks at yahoo.com.au> wrote:
I am wondering if there were any magnetic tape filesystem?
CAPS-11 comes to mind :-)
I had to use it once and it made RT-11 on 8? floppies seem quick and it was slow compared
to our DECsystem-10 even with 60 users...
Huw Davies | e-mail: Huw.Davies at kerberos.davies.net.au
Melbourne | "If soccer was meant to be played in the
Australia | air, the sky would be painted green"
9 Feb
9 Feb
2:28 a.m.
ANSI?
Johnny
On 2015-02-09 01:50, Tom Sparks wrote:
I am wondering if there were any magnetic tape
filesystem?
I know that some micro you could give a name to your data/program on the
audio cassette and they had a catalogue feature
I also know about Linear Tape File System (LTFS)
Has anyone tried putting cp/m or fat filesystem on audio cassette?
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
7:37 a.m.
On 02/08/2015 06:50 PM, Tom Sparks wrote:
I am wondering if there were any magnetic tape
filesystem?
I know that some micro you could give a name to your data/program on the
audio cassette and they had a catalogue feature
I also know about Linear Tape File System (LTFS)
Has anyone tried putting cp/m or fat filesystem on audio cassette?
I had 9-track 800 BPI mag tape on my CP/M system, almost
entirely software-driven.
I used it for backup, and it made ANSI-D compatible tapes
that could be read on
the VAX at work.
Jon
7:49 a.m.
On 2/8/2015 7:50 PM, Tom Sparks wrote:
I am wondering if there were any magnetic tape
filesystem?
I know that some micro you could give a name to your data/program on the
audio cassette and they had a catalogue feature
I also know about Linear Tape File System (LTFS)
Has anyone tried putting cp/m or fat filesystem on audio cassette?
*This question looks to be overanswered
by now, but the HP 9144 1/4" tape drive
with the HP 200 or 300 series... works
well as a boot and mass storage device
under HP Basic.
- J.
*
10:40 a.m.
All this talk about magtapes brought some memories to the fore.
While a single tape drive was the rule for personal computers, it
certainly wasn't that way for mainframes. In particular, one very
important aspect was sorting (and, by extension, merging). The wonder
of a polyphase (or better yet, oscillating) sort running on a bank of 8
300 ips drives was mesmerizing.
The book by Ivan Flores on sorting was pretty much a bible until Don
Knuth came out with his "Searching and Sorting" book, complete with
fold-out charts. That's the one out of the whole "Art" series that
sticks in my mind.
Now, I suspect that the average programmer doesn't worry (or perhaps
even know about) external sorting what with vast RAM and disk drives,
but back in the day, it was fascinating.
Undoubtedly, the folks before me who spent their time shuttling trays of
punched cards around an IBM 082 have a similar reaction to the
"easiness" of tape sorts as compared to card sorts...
--Chuck
11 a.m.
On Mon, Feb 09, 2015 at 10:40:06AM -0800, Chuck Guzis wrote:
[...]
Now, I suspect that the average programmer doesn't
worry (or perhaps even
know about) external sorting what with vast RAM and disk drives,
One still needs to know about external sorts once the data is larger than RAM,
but these days it's often called "Big Data", a term which usually means
"I want
to use these sledgehammers on this nut so I can add more buzzwords to my CV".
but back in the day, it was fascinating.
JOOI, why was there such a need to sort large datasets back in the mainframe
era?
11:41 a.m.
On 02/09/2015 11:00 AM, Peter Corlett wrote:
JOOI, why was there such a need to sort large datasets
back in the mainframe
era?
Knuth's book's title gives a hint. He called it "Searching and
Sorting". The two go hand in hand.
I think the simple answer is that it's easier to search sorted data,
particularly when said data is offline--i.e. hanging in a slot in a tape
rack, or in a drawer in a card cabinet. Nowadays, when everything can
be kept online, it's probably less important than it used to be.
--Chuck
2:34 p.m.
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Chuck
Guzis
Sent: 09 February 2015 19:42
To: General at classiccmp.org; Discussion at classiccmp.org:On-Topic and Off-
Topic Posts
Subject: Re: Magnetic tape filesystem
On 02/09/2015 11:00 AM, Peter Corlett wrote:
Performance? Where I worked our master file was small by modern standards, a
mere 4million record, but it took up 14 reels of tape. The update program
had many tricks to speed things up...
JOOI, why was there such a need to sort large
datasets back in the
mainframe era?
Knuth's book's title gives a hint. He called
it "Searching and Sorting".
The two
go hand in hand.
I think the simple answer is that it's easier to search sorted data,
particularly
when said data is offline--i.e. hanging in a slot in a
tape rack, or in a
drawer in
a card cabinet. Nowadays, when everything can be kept
online, it's
probably
less important than it used to be.
--Chuck
3:12 p.m.
On 02/09/2015 02:34 PM, Dave G4UGM wrote:
Performance? Where I worked our master file was small
by modern standards, a
mere 4million record, but it took up 14 reels of tape. The update program
had many tricks to speed things up...
Now, consider the situation if you were the IRS in 1970.
--Chuck
1:52 p.m.
On Mon, 9 Feb 2015, Peter Corlett wrote:
JOOI, why was there such a need to sort large datasets
back in the mainframe
era?
Consider, for example, a national telephone directory.
But, what constitutes "large" has changed over the years. At one time,
mainframes did not have much memory.
I tried to teach some of the basic concepts in a Data Structures And
Algorithms class. Several students (from UC Berkeley) insisted that
the ONLY appropriate way to sort a dataset that was larger than RAM
was to increase the RAM of the computer.
'course there were also people who thought that the only appropriate
way to insert an additional record into a large file was to append it,
and then sort the file. And then got upset when I explained that if they
were going to do that particular sort, then a properly designed bubble
sort would be superior to Shell/Metzner.
--
Grumpy Ol' Fred cisin at xenosoft.com
11:03 a.m.
All this talk about magtapes brought some memories to
the fore.
While a single tape drive was the rule for personal computers, it
certainly wasn't that way for mainframes. In particular, one very
important aspect was sorting (and, by extension, merging). The wonder of
a polyphase (or better yet, oscillating) sort running on a bank of 8 300
ips drives was mesmerizing.
Undoubtedly, the folks before me who spent their time shuttling trays of
punched cards around an IBM 082 have a similar reaction to the "easiness"
of tape sorts as compared to card sorts...
I remember a regular job (every week....) in my servicebureau where I had to
sort about 40.000 200-byte records on a DOS machine. DOS did have a SORT
command, but it was horribly insufficient.
I therefore had to get back to my experience as IBM 82 operator (!), and do
it more or less that way. The first stop was to create x files, each
containing about 1000 key fields. When the input was read, I would sort one
keyfile at a time internally and write it to a new file. After the sorting,
I merged two indexfiles and write a new file, etc. When all indexfiles were
merged into one, I found the respective full records, and wrote the output
file.
It took about 15 minutes or so for the whole job, but the customer was
happy.
/Nico
11:31 a.m.
----- Original Message -----
From: "Chuck Guzis" <cclist at sydex.com>
To: <General at classiccmp.org>;
"Discussion at classiccmp.org:On-Topic and Off-Topic
Posts" <cctalk at classiccmp.org>
Sent: Monday, February 09, 2015 1:40 PM
Subject: Re: Magnetic tape filesystem
All this talk about magtapes brought some
memories to the fore.
While a single tape drive was the rule for
personal computers, it certainly wasn't that way
for mainframes. In particular, one very
important aspect was sorting (and, by extension,
merging). The wonder of a polyphase (or better
yet, oscillating) sort running on a bank of 8
300 ips drives was mesmerizing.
...
Undoubtedly, the folks before me who spent their
time shuttling trays of punched cards around an
IBM 082 have a similar reaction to the
"easiness" of tape sorts as compared to card
sorts...
--Chuck
----- Reply -----
... especially when you tripped and dropped a
stack of sorted cards on the way to the collator
or 403...
Was just leafing through documentation on using
cassettes in Burroughs L systems:
" It is possible to sort data by using 3 or 4
magnetic tape stations ...
_However, this is a slow operation and it is not
advised that magnetic tape cassettes are used in
this way._"
Sorting 900 records of 256 bytes on a 512 word
base memory machine with three drives took 1 hour,
41 minutes. Increasing memory up to 1536 words and
adding a fourth drive brought that down to a mere
55 minutes!
But these were different times where life moved
more slowly. I actually designed several systems
using four cassette drives and magnetic cards, and
one or two days every week you just set aside an
hour or two for sorting and housekeeping.
And on the other hand,:
"A cassette is easily transportable from one site
to another compared to a disk. Drop a disk on the
floor and the chances are that it can't be used
anymore."
Now, sorting using paper tape, that oughta be
fun...
m
11:45 a.m.
On 02/09/2015 11:31 AM, Mike Stein wrote:
But these were different times where life moved
more slowly. I actually designed several systems
using four cassette drives and magnetic cards, and
one or two days every week you just set aside an
hour or two for sorting and housekeeping.
Magnetic card storage was left out of this discussion. NCR 353 CRAM,
anyone? Did anyone ever write an external sort using that beast?
--Chuck
3603
days inactive
3604
days old
40 comments
22 participants
participants (22)
-
abuse＠cabal.org.uk -
ard＠p850ug1.demon.co.uk -
bqt＠update.uu.se -
cclist＠sydex.com -
cctalk＠fahimi.net -
cisin＠xenosoft.com -
dave.g4ugm＠gmail.com -
db＠db.net -
elson＠pico-systems.com -
huw.davies＠kerberos.davies.net.au -
js＠cimmeri.com -
jules.richardson99＠gmail.com -
jws＠jwsss.com -
leec2124＠gmail.com -
lproven＠gmail.com -
mhs.stein＠gmail.com -
nf6x＠nf6x.net -
nico＠farumdata.dk -
snhirsch＠gmail.com -
spacewar＠gmail.com -
tom_a_sparks＠yahoo.com.au -
wilson＠dbit.com