🚀 go-pugleaf

In article <65g3r9$8...@bgtnsc03.worldnet.att.net> Dan-
Yert...@worldnet.att.net (Dan Yertzell) writes:

>Does anyone here know the specs of the Honeywell H-200?  It was the
>first machine I ever worked with in school in the 60's, programmed
>it in Fortran and Assembler, etc.  Anyways, I was just curious how it
>compared to todays machines.  Is it possible to talk about mhz ratings
>and such about such an ancient machine?
>
>I remember it had 4 tape drives, a card reader, a card punch, a line
>printer and a teletype style console.  I think it had 16k of memory.
>We had the usual assortment of IBM 029 and 026 keypunch machines, and
>I remember being wowed when we got our first 129.
>
>I only bring this up because I just came across a box full of 30 year
>old program listings and punch cards in my attic.  Quite a nostalgic
>charge.

I dug through my collection of obsolete computer manuals and found
some stuff on the Honeywell 200 series.  (I got to play with a 120,
myself.)  Here are some excerpts from the "Series 200 Summary
Description", second edition, first printing, February 1966,
file number 143.0005.0000.0-012.

------------------------------------------------------------------------

The six processors of Series 200 are the Models 120, 200, 1200,
2200, 4200, and 8200.  The 8200, the most powerful member of the
series, is completely compatible not only with other Series 200
models but also with Honeywell's 800 and 1800 systems.  Because
of its dual processing ability it is described in a separate
publication entitled 'Honeywell Series 200 / Model 8200 Summary
Description', File No. 143.0011.0000.0-191.

Further description of Series 200 within this publication refers
only to the Models 120, 200, 1200, 2200, and 4200.


Processing Dimension

- Memory speeds ranging from 3 microseconds to 188 nanoseconds
  per character

- Memory capacities ranging from 2,048 to 524,288 characters,
  in modular increments

- Up to 30 index registers; flexible nanosecond control memories

- A universal set of powerful instructions affording program
  compatibility between processors

- Instruction and data compatibility with 1401, 1410, 1460, and
  7010 systems

- Advanced programming and memory addressing methods, plus editing
  and multiply/divide operations

- Powerful floating-point capability


Input/Output Dimension

- Up to 16 peripheral operations performed simultaneously with
  computing

- Up to 48 peripheral control units connected to a processor; each
  accomodates one or serval peripheral devices and is equipped with
  an automatic program interrupt facility

- A wide variety of peripheral equipment available in a range of
  performance capabilities, including communication devices, card
  equipment, magnetic tape and paper tape units, mass storage units,
  high-speed printers, banking equipment, and memory-to-memory
  adapter units

- Broad-scale real-time capability that includes an efficient
  interrupt facility, single- and multi-channel communication
  controls (the latter accepting data from up to 63 lines
  simultaneously), multi-level code handling, and a wide range
  of remote terminal facilities


Software Dimension

- Basic Programming System, which provides flexible software modules
  employing self-loading, unit-record techniques for use in small,
  card-oriented installations

- Operating system - Mod 1, which provides semi-centralized, automatic
  control for medium-scale tape- or mass-storage installations

- Operating System - Mod 2, which provides completely centralized,
  highly automated computer management, with a minimum hardware
  overhead requirement, for medium- and large-scale installations
  of all types

- Application Systems, which assist directly in the performance
  of functions which are part of a user organization's operations

------------------------------------------------------------------------

[Moving on to section 2, "Processors", here's a small table giving
execution times of a few instructions.  Detailed execution times
of all instructions are given later in the book for those who are
really interested.  This stuff was nice to have if you were trying
to optimize CPU usage.]

                                   Execution Time, Microseconds
Operation                       Model 120   Model 1200   Model 4200
---------                       ---------   ----------   ----------
Decimal Add                         69          35           10
Compare                             57          29            9
Branch if Character Equal           36          18            7
Move Characters to Word Mark        54          27            8
Floating Multiply                  n/a          56           18

Three-character addresses are used to refer to five-character
operand fields.  Instruction access times are included in the
times shown.  The times for floating multiply refer to operations
using a 36-bit mantissa and a 12-bit exponent.

------------------------------------------------------------------------

SUMMARY OF PROCESSOR CHARACTERISTICS

           Main    Memory       Number  Number of
           Memory  Capacity     of      I/O operations
           Speed   (thousands   Input/  Simultaneous
Processor  (cycle  of           Output  with
Model      time)   characters)  Trunks  Computing
-------------------------------------------------------
120          3        2-32        *         2-3
200          2        4-65       8-16       3-4
1200         1.5     16-131      16         4
2200         1       16-262      16-32      4-8
4200         0.188   65-524      32-48      8-16

* Basic print, card read, card punch controls.
  Tape control and II [sic] I/O trunks optional.

------------------------------------------------------------------------

In order to appreciate the full power of the read/write channel
concept, consider the following statistics.  In one minute, a
Series 200 system equipped with a Model 200 processor having four
read-write channels can:

    read 800 cards;

    punch 10 columns of data into 400 cards;

    print 1300 lines of 120 characters each;

    read or write 4360 tape records of 500 characters;

(or perform any combination of four I/O operations) and in the
same minute, execute 1.25 million instructions.

[That's right, 1.25 million instructions per _minute_, not second.
It's amazing how much work you can get done when you're not
spending all your CPU's power pushing pixels around.]

------------------------------------------------------------------------

The manual goes on to describe the usual assortment of card readers
and punches, printers, drums, and magnetic and paper tape units.
And, further to another thread that's cropped up here from time
to time, Honeywell even had their equivalent of the infamous IBM
2321 data cell.  The description of Honeywell's Mass Memory File
is detailed enough to leave no room for doubt that it's the same
sort of device.  Types 251, 252, and 253 offered storage capacities
of 15, 63, and 317 million characters respectively.

If I had a scanner, I could put up photos of the various pieces of
hardware, complete with smartly-dressed women posing among them.

--
cgi...@sky.bus.com (Charlie Gibbs)
Remove the first period after the "at" sign to reply.

"Charlie Gibbs" (cgi...@sky.bus.com) writes:

[massive spec sheet snip]
> 
> [That's right, 1.25 million instructions per _minute_, not second.
> It's amazing how much work you can get done when you're not
> spending all your CPU's power pushing pixels around.]

  Ahmen, brother.  OS/2 Warp 4 ruining on my 486-DX4 75 MHz machine
  can't even do voice type!    ^^^^^^^

Glad to see some postings about the H-200.  I also played with one in the
very early 1970s.

Some quick reminiscing about this particular system (perhaps this will
spark more discussion):

* Had an assembler (called Easycoder?), FORTRAN compiler, COBOL compiler,
and I think it also had an RPG compiler.

* Had 4 sense switches on the console.....lit push buttons, which you
pushed once to set the sense switch on, and pushed again to turn it off.

* The console was indeed built by Teletype.  Had a "printbox" sort of
mechanism which printed on roll paper.  Could act as a full featured I/O
device from your program (COBOL DISPLAY/ACCEPT, etc.) but also had a
handful of keys which were active when the machine was stopped & would be
used to enter bootstrap commands, display & change memory (in octal),
etc.  Still remember booting from tape as

    B 40 000000    where B meant Bootstrap, 40 was device address for tape
drive, and 000000 was octal address to load into.

    B 41 000317    for loading object decks produced by COBOL from the
card reader (device address 41) -- don't remember reason for offset
loading address.

    B 44 000000    for booting up the disk-based OS (44 was the address of
the disk drive(s))

    In fact, the device addresses on this system were:      00/40  - tape
write/read, 01/41 - card punch/card reader, 03 - line printer, 04/44 -
disk write/read, 07/47 - console typewriter/keyboard.  It too had 4 tape
drives (200 bpi, 7 track IIRC) and also 1 disk drive (multiple platters,
loaded pack onto spindle from the top after it spun down and you lifted
the drive's lid like a toploading washer)

* There was a tape-based operating system, and a disk-based operating
system too.

* Bigger processors had 4 word addressing (IIRC), but ours was at the
small end of the line and only had 2 and 3 word addressing [there were
lights on the console for each, but I never remember the 4 word addressing
light coming on]

* You always pressed STOP and INITIALIZE before booting.  Then the run
button after the B command from the console had done its work.

* The tape drives could be mapped (physical drive # to logical drive #
with a set of thumbwheels on the front of the tape controller in the CPU
frame).  Likewise for the disk drives, although we only had one so it
didn't make a lot of sense to change its thumbwheels.

* I/O was done at machine language level with the PDT (Peripheral Data
Transfer) instruction.  There was also a PCB (Peripheral Control &
Branch?) instruction for I/O control and status checking.  Don't remember
any interrupt driven I/O although that's not to say there wasn't any on
that CPU.

* You could write nifty one card programs in machine language to do 80/80
listings, 80/80 card dupes, make a copy of a source deck (cols. 1-72) and
punch sequence # in 73-80, etc....just bootstrap the card in, and go.

Nifty machine for a 14 year old to explore, and in retrospect, probably
very little damage I could have done as long as I didn't physically mount
any tapes or disk packs which were used for "production".

Any other recollections out there ?

M. Baker
mba...@monmouth.com

On 27 Nov 97 11:13:44 -0800, "Charlie Gibbs" <cgi...@sky.bus.com>
wrote:

>I dug through my collection of obsolete computer manuals and found
>some stuff on the Honeywell 200 series.  (I got to play with a 120,
>myself.)  Here are some excerpts from the "Series 200 Summary
>Description", second edition, first printing, February 1966,
>file number 143.0005.0000.0-012.
>

      --   much deleted  --

>
>If I had a scanner, I could put up photos of the various pieces of
>hardware, complete with smartly-dressed women posing among them.

Wow.  More info than I thought anyone would (or should!) have.  Thanks
for the info.  I'd love to see the photos.  

Dan

On Fri, 28 Nov 1997 22:51:40 -0500, The Bakers <mba...@monmouth.com>
wrote:

>Glad to see some postings about the H-200.  I also played with one in the
>very early 1970s.
>
>* Had 4 sense switches on the console.....lit push buttons, which you
>pushed once to set the sense switch on, and pushed again to turn it off.

OK, you made me do it.  I dug out my hand written notes on the system.
The only sense switch we ever used was 1, to bypass bad spots on
tapes.
>
>* The console was indeed built by Teletype.  Had a "printbox" sort of
>mechanism which printed on roll paper.  Could act as a full featured I/O
>device from your program (COBOL DISPLAY/ACCEPT, etc.) but also had a

Our display console was address 07, accept console was 47.

>handful of keys which were active when the machine was stopped & would be
>used to enter bootstrap commands, display & change memory (in octal),
>etc.  Still remember booting from tape as
>
>    B 40 000000    where B meant Bootstrap, 40 was device address for tape
>drive, and 000000 was octal address to load into.
>
>    B 41 000317    for loading object decks produced by COBOL from the
>card reader (device address 41) -- don't remember reason for offset
>loading address.

How about B 41 001620 for Easycoder Assembler?

>
>    B 44 000000    for booting up the disk-based OS (44 was the address of
>the disk drive(s))

We had no disk drives!!

>* There was a tape-based operating system, and a disk-based operating
>system too.
>
>* You always pressed STOP and INITIALIZE before booting.  Then the run
>button after the B command from the console had done its work.

COBOL compiles were 40 - B, 40 - B, Run - Run - Run!

>
>* The tape drives could be mapped (physical drive # to logical drive #
>with a set of thumbwheels on the front of the tape controller in the CPU
>frame).  Likewise for the disk drives, although we only had one so it
>didn't make a lot of sense to change its thumbwheels.

Yeah I remember those too.  Our instructor used to threaten us with
severe bodily harm if he ever saw us set the SYSTEM drive with PERMIT
status.

I see from one of my many printouts that the COBOL compiler was
version 12.0.  SOURCE-COMPUTER is "H-200 SPECIAL".

>* You could write nifty one card programs in machine language to do 80/80
>listings, 80/80 card dupes, make a copy of a source deck (cols. 1-72) and
>punch sequence # in 73-80, etc....just bootstrap the card in, and go.

Yeah we had all those too.

>
>Nifty machine for a 14 year old to explore, and in retrospect, probably
>very little damage I could have done as long as I didn't physically mount
>any tapes or disk packs which were used for "production".

I was 17 & 18 when I used it.  I've even got a gen-yew-wine Honeywell
flowcharting template (remember flowcharts?).

One problem we had was the character set differed somewhat from that
used by the IBM keypunch machines, so we had a chart taped to the wall
behind each keypunch machine with stuff like this:

                    IBM    =     Honeywell
                     @     =         :
                     %     =         (
                     <      =         )

etc.  That kept things interesting.

We even had an 024 keypunch, the kind that didn't even print readable
text above each column, so you had to be *really* good to read a card
from it!!  

According to my notes the H200 had 16k of memory.  I don't know if
this is bytes or words.

Dan

Dan Yertzell wrote:

> How about B 41 001620 for Easycoder Assembler?

Vaguely remember that one......do you think the "1620" was just a coincidence
???  :-)


> COBOL compiles were 40 - B, 40 - B, Run - Run - Run!

Yes, now I remember having to enter the bootstrap command twice and the three
RUN button pushes.  Wasn't the first B40 to effectively "skip" over the tape's
header record with the actual first record of the boot loader being the second
record on the tape ?  As for the reason behind three hits on the RUN button, I
am a little bit rusty,  Was it pausing at various points in the load to check
sense switches for options ???

I *do* remember that the console typewriter would display "COMPILING PROGNAME"
when the COBOL compiler began it's work.  PROGNAME was your program's name as
entered on a control card at the beginning of the deck [separate from the COBOL
Program-ID in the Identification Division".

Also I seem to recall a program (overlay?) called something like ABAVPA010.
Was this the name for the executable on tape which began the COBOL compile ?
Seems like this was the name that you told the tape operating system to load &
go when you went to do a COBOL compile.

> Yeah I remember those too.  Our instructor used to threaten us with
> severe bodily harm if he ever saw us set the SYSTEM drive with PERMIT
> status.

Ah yes......PERMIT & PROTECT.  Wasn't there a big black rotary switch knob on
each tape drive to set this (along with the write protect ring, of course) ?


> One problem we had was the character set differed somewhat from that
> used by the IBM keypunch machines, so we had a chart taped to the wall
> behind each keypunch machine with stuff like this:
>
>                     IBM    =     Honeywell
>                      @     =         :
>                      %     =         (
>                      <      =         )
>
> etc.  That kept things interesting.

Sounds like the machine used the "old" character set mappings (as were usually
found on 026 keypunches) while you had the newer 029 keypunches with the newer
EBCDIC mappings.  Just a guess.


M. Baker
mba...@monmouth.com

In article <34818B2D...@monmouth.com> mba...@monmouth.com
(The Bakers) writes:

>Also I seem to recall a program (overlay?) called something like
>ABAVPA010.  Was this the name for the executable on tape which began
>the COBOL compile?  Seems like this was the name that you told the
>tape operating system to load & go when you went to do a COBOL compile.

I played around with the FORTRAN compiler (I hadn't learned COBOL
yet.)  The value that sticks in my head was ACADRV010.  It was
referred to as a "console call card".  Kind of odd, I thought,
considering that that H-120 didn't even have a console...

Other random reminiscences:  One of the tape drives was a bit flaky
in that if the temperature went over about 70 degrees F, it wouldn't
sense the beginning-of-tape marker while rewinding - the end of the
tape would run off the end of the take-up reel and drop to the bottom
of the vacuum column with a loud THWOP, and you'd have to re-thread
the tape before you could use it again.

--
cgi...@sky.bus.com (Charlie Gibbs)
Remove the first period after the "at" sign to reply.

> somebody wrote:
>
>>If I had a scanner, I could put up photos of the various pieces of
>>hardware, complete with smartly-dressed women posing among them.
>

The Honeywell 200 won an award from Industrial Design magazine for its
uniquely neat design.  For those who may have never seen one, I'll try to
describe.  The cabinets were all about 3 feet high and had Formica tops.
This was because the designer realized that in a computer room one thing
you never have enough of is counter space where you can put things.
The cabinets were raised off the floor on legs, so it was possible to
sweep under them.

Another award-winning design was the G.E. 400 family, which had vertical
cabinets fastened together at one edge so that as viewed from above it
could be a 3-pointed or 4-pointed star configuration.

In article <7277.270T2...@sky.bus.com>,
"Charlie Gibbs" <cgi...@sky.bus.com> wrote:

 > The six processors of Series 200 are the Models 120, 200, 1200,
 > 2200, 4200, and 8200.  The 8200, the most powerful member of the
 > series, is completely compatible not only with other Series 200
 > models but also with Honeywell's 800 and 1800 systems.  Because
 > of its dual processing ability it is described in a separate
 > publication entitled 'Honeywell Series 200 / Model 8200 Summary
 > Description', File No. 143.0011.0000.0-191.

The 8200 was actually a really cool machine.  It was a processor from the
Honeywell 800 line acting as the operating system and executive manager for
a system on which much of the work was handled by units from the Honeywell
200 family.  It was, apparently, a /superb/ machine.  I never worked on it
but I did work for Honeywell when the 8200 was still being sold, and knew a
number of people who had used it intensively.  Apparently, when the
Worldwide Military Command and Control System (WWMCCS, the original WWMCCS)
was up for bids, Honeywell technical staff wanted to bid 8200's for the
project because at the time, the 8200 was a more powerful and
better-equipped system than the Honeywell 6000-series, which was just a
cleaned-up GE-600 machine.  The 8200, it was felt, would handle the WWMCCS
benchmark much more easily than the newer 6000, which had fairly unreliable
software at the time.  Due to some quirk in the RFP (real-time clock,
guarnateed maintenance of FORTRAN compiler, something really pretty
trivial), the 8200 was edged out and the 6000-series made it to the
winner's circle.  That was probably a really good thing because the 8200
wasn't a machine that had a future, it was already about 20% bigger than
maxed-out, while the 6000 product line still had a lot of room for growth
and improvement and handled the WW requirements sort-of tolerably well. 
(It depends on who you talk to.)

The 8200 was of the same flavor as the Honeywell 1648 timesharing system,
cobbled together from available processors that individually couldn't get
much done but able to do very well as long as they flew in the same
direction.  I'm not sure how many other manufacturers did that kind of
thing with their systems -- I can't recall Univac doing it, and CDC didn't
(although they did dual-process a little on their 6000's).

In article <7277.270T2...@sky.bus.com>,
"Charlie Gibbs" <cgi...@sky.bus.com> wrote:

 > And, further to another thread that's cropped up here from time
 > to time, Honeywell even had their equivalent of the infamous IBM
 > 2321 data cell.  The description of Honeywell's Mass Memory File
 > is detailed enough to leave no room for doubt that it's the same
 > sort of device.  Types 251, 252, and 253 offered storage capacities
 > of 15, 63, and 317 million characters respectively.

Sort of.  Honeywell bught their worst devices from CDC and re-badged them. 
The data-cell device was in that category.  Later, they co-sponsored a
factory in Oklahoma City to build disk and tape peripherals, and actually
brought out some decent devices.  Honeywell had a good sense, like IBM, of
what the customer wanted to see, and CDC knew engineering.  The reverse was
not true -- Honeywell couldn't (generally) engineer their way out of a
paper bag, and CDC's idea of user-friendly was to slap more push-buttons in
less obvious places for the users to look for.

The Honeywell 200 was originally designed as an input/output machine for 
the H800 (a truly wierd computer, with process multiplexing in hardware, 
among other things). When the machine was being designed there was a 
major dispute about what the instruction set should be. Somebody tossed a 
1401 manual on the table and said "why not this?" Thus, by accident, was
born the first clone (ca 1963).

When I joined Honeywell in 1966, the company was booming in the 1401 
replacement market. This was due to three factors: the cycle-stealing I/O 
invented by my old boss, Lou Oliari (a character if there ever was one), 
a program called "Liberator" done by Dick Little, which converted 1401
assembly programs to 200 Easycoder, and, most importantly, the read 
backward cascade polyphase sort, whose author's name escapes me. Sorting 
was critical to performance in the batch tape days, and we would beat 
IBM's brains out in benchmarks (I was in the benchmarking/demo group part 
time).  The 360 came out, running 1401 programs under emulation, and we'd 
cream them even worse. It was so bad that people ran service bureaus that 
did nothing but sort tapes; you brought your daily work in at 5:00 PM and 
picked up the sorted tape in the morning. Even big dedicated IBM shops 
had 200's sitting in the background chugging away sorting tape.  The 
funnies story in that regard came from GM; when the IBM salesman would 
show prospects the acre of dark grey boxes, there would be this bright 
blue thing with Honeywell written on it back in the corner. If the 
prospect asked what it was, the IBM salesman replied that it ran the air 
conditioning (Honeywell being best known for building controls).

Earl

In <34818B2D...@monmouth.com>, on 11/30/97 
   at 10:50 AM, The Bakers <mba...@monmouth.com> said:

::Sounds like the machine used the "old" character set mappings (as were
::usually found on 026 keypunches) while you had the newer 029 keypunches
::with the newer EBCDIC mappings.  Just a guess.

Not quite.  The commercial character set was distinct from the scientific
set in the 026 era, the code points were the same but what was printed
differed.  I don't remember all of them now, but the sci ) was mapped to
the same code point as the square blob in commercial.  When 360 came out
with ebcdic, the dual mapping was done away with - at least for these
characters.

I don't remember much about the H200 - worked on the D1000, the H800, and
the H400.  Wasn't the 200 originally designed for process control?
 
-- 
 Julian & Mary Jane Thomas
 j...@epix.net  http://www.epix.net/~jt
 In the beautiful Finger Lakes Wine Country of New York State!
 --------------------------------------------------
 One man's Windows are another man's walls.

In article <65ulpg$26d$1...@sloth.swcp.com>,
William E. Boebert <boe...@swcp.com> wrote:
>The Honeywell 200 was originally designed as an input/output machine for 
>the H800 (a truly wierd computer, with process multiplexing in hardware, 
>among other things). When the machine was being designed there was a 
>major dispute about what the instruction set should be. Somebody tossed a 
>1401 manual on the table and said "why not this?" Thus, by accident, was
>born the first clone (ca 1963).

Hmmm, another clone was born at GE a bit later, I know that the GE600
series was a an IBM 7090 with the opcode and address fields swapped
(GE600 had the address in the UPPER half of the word, and the OPcode
in the lower half).  Of course the GE 600/6000 series became the
Honeywell 600/6000 series in about 1972 or so.  I was at the Explorer
Post at Honeywell Phoenix at about that time...

>
>When I joined Honeywell in 1966, the company was booming in the 1401 
>replacement market. This was due to three factors: the cycle-stealing I/O 
>invented by my old boss, Lou Oliari (a character if there ever was one), 
>a program called "Liberator" done by Dick Little, which converted 1401

I still remember seeing the back cover ads for the Liberator in the
trade rags.  It included a lion made out of electronic components.  It
"liberated" all the IBM shops from having to keep their 1401s running.
I seem to recall that the Libertator emulator ran at about 2.5+ times
of IBMs own 1401 emulator running on IBM hardware (360, maybe)?

>assembly programs to 200 Easycoder, and, most importantly, the read 
>backward cascade polyphase sort, whose author's name escapes me. Sorting 

Perhaps John Wertz was the author of this sort?

>was critical to performance in the batch tape days, and we would beat 
>IBM's brains out in benchmarks (I was in the benchmarking/demo group part 
>time).  The 360 came out, running 1401 programs under emulation, and we'd 
>cream them even worse. It was so bad that people ran service bureaus that 
>did nothing but sort tapes; you brought your daily work in at 5:00 PM and 
>picked up the sorted tape in the morning. Even big dedicated IBM shops 
>had 200's sitting in the background chugging away sorting tape.  The 
>funnies story in that regard came from GM; when the IBM salesman would 
>show prospects the acre of dark grey boxes, there would be this bright 
>blue thing with Honeywell written on it back in the corner. If the 
>prospect asked what it was, the IBM salesman replied that it ran the air 
>conditioning (Honeywell being best known for building controls).
>
>Earl
>

--tep (formerly Perrine.{G8ARCH,Multics,Scouting})



-- 
Tom E. Perrine (t...@SDSC.EDU) | San Diego Supercomputer Center 
http://www.sdsc.edu/~tep/     | Voice: +1.619.534.5000
I miss my 36-bit friends: Multics, TOPS-10, and TOPS-20.

Tom Perrine <t...@galt.sdsc.edu> wrote:
>Hmmm, another clone was born at GE a bit later, I know that the GE600
>series was a an IBM 7090 with the opcode and address fields swapped

...and a significantly different instruction set (e.g., I don't *think*
the 709x had as many addressing modes as the GE6xx).

I vaguely remember somebody, perhaps here, saying why the GE6xx had an
instruction to convert Gray code to binary; anybody know why?
-- 
Reply, or follow up, but don't do both, please.

postmaster@localhost
postmaster@[127.0.0.1]

On Sun, 30 Nov 1997 10:50:05 -0500, The Bakers <mba...@monmouth.com>
wrote:


>Also I seem to recall a program (overlay?) called something like ABAVPA010.
>Was this the name for the executable on tape which began the COBOL compile ?
>Seems like this was the name that you told the tape operating system to load &
>go when you went to do a COBOL compile.
>

Wow you have a good memory!   The first line of the COBOL printouts
is:

START OF CHECKOUT USING ABAVPA ON 74128 AT 0800 HOURS

Of course, the last line is:

OBJECT PROGRAM RUN TAPE IS ON LOGICAL TAPE UNIT 2

I'd still like to find some pictures of the old beast!!

Dan

In article <662aej$7...@tooting.netapp.com>, Guy Harris <g...@netapp.com> wrote:
>Tom Perrine <t...@galt.sdsc.edu> wrote:
>>Hmmm, another clone was born at GE a bit later, I know that the GE600
>>series was a an IBM 7090 with the opcode and address fields swapped
>
>...and a significantly different instruction set (e.g., I don't *think*
>the 709x had as many addressing modes as the GE6xx).

No where near as many.  On the other hand, the product spec for the 6x5
actually did say something close to "it should feel familiar to a
programmer experienced with the 7090."  This isn't folklore, I actually
saw the GE specs, although it was a long time ago.

>I vaguely remember somebody, perhaps here, saying why the GE6xx had an
>instruction to convert Gray code to binary; anybody know why?

Because it was also had a military version: the A605 (airborne) M605 (truck
ship born).  Gray code is used to encode the position of rotating shafts
(like old style radar antennas) because it has the nice property that only
one bit changes at each transition.

A cute generation in hardware design.  The M605 had discrete component
boards (this was 1965), but _every_ transistor was in its own little
shock absorbing socket.

-- 
--------
Sarr Blumson                    sa...@umich.edu
voice: +1 313 764 0253          home: +1 313 665 9591
ITD, University of Michigan     http://www-personal.umich.edu/~sarr/
535 W William, Ann Arbor, MI 48103-4943

While we are on the subject of the GE6xx/H6yyy machines, any recollections (fond
or otherwise) about ---

    1) Assembler language programming (IIRC, it was called "GMAP" -- at least in
the GE world)

    2) DATANET communications controllers

M. Baker
mba...@monmouth.com

This has been a great discussion so far!  Brings back lots of memories [no pun
intended :]

Dan Yertzell wrote:

> The first line of the COBOL printouts
> is:
>
> START OF CHECKOUT USING ABAVPA ON 74128 AT 0800 HOURS

Here the next question:

Did the H-200 have a built-in clock/calendar hardware (I don't *think* it did, at
least the one I used) ?  Then where did the date & time come from on the header
shown above from the COBOL compile ?

Did you have to input the current date and time somehow, perhaps on the console
call card ?

Inquiring minds want to know.....
M. Baker
mba...@monmouth.com

In article <34860CF2...@monmouth.com>, The Bakers
<mba...@monmouth.com> wrote:

> While we are on the subject of the GE6xx/H6yyy machines, any
recollections (fond
> or otherwise) about ---
>  1) Assembler language programming (IIRC, it was called "GMAP" -- at least in
> the GE world)
> 
With General Electric, the assembler was call "GEMAP", Generalized
something something.  When Honeywell bought the operation, they changed
the name to GMAP, as well as producing GLOAD from GELOAD.  The OS had been
named GECOS, Generalized Comprehensive ...  I really think that Honeywell
should have called it "HOCUS" rather than GCOS.

In article <bertschc-051...@ip-23-174.phx.primenet.com>,
Charles Bertsch <bert...@acm.org> wrote:
...
>With General Electric, the assembler was call "GEMAP", Generalized
>something something.  When Honeywell bought the operation, they changed
>the name to GMAP, as well as producing GLOAD from GELOAD.  The OS had been
>named GECOS, Generalized Comprehensive ...  I really think that Honeywell
>should have called it "HOCUS" rather than GCOS.

I was at the Honeywell facilities in Phoenix just after Honeywell
bought the GE computer divisions.  They Explorer Post there got all
its manuals as second-hand; as each new SW release came out, we would
get the old ones.  Until recently, I still had Multics manuals in my
garage, they were disposed of via alt.os.multics, which I recommend if
you really want to know more about the GE600/6000 series.

Any way, for a while I had both GE and Honeywell manuals for some
things.

GEMAP was General Electric Macro Assembly Processer
GMAP was  General Macro Assembly Processer
GECOS was General Electric Comprehensive Operating System
GCOS was General Comprehensive Operating System

At least, that's what was in the early manuals.  Later I saw
Generalized in place of General in some manuals.

GECOS was my first OS, GMAP was my first assembly language (after
learning GCOS Basic, FORTRAN, COBOL and Multics PL1).

I was also extremely fortunate to serve a long internship under John
Wertz, was one of the primary designers of the architecture of the
instructions to be provided by "EIS box" (pronounced ICE box), which
was the primary difference between the (original) 600 and 6000 series.

As I recall, the EIS box was an option on later 600 models, the 6000
had the functionality of the EIS box incorporated into the overall
design of it CPU.

The EIS was the Extended Instruction Set, which had all kinds of
complex instructions for supporting things that COBOL and FORTRAN
liked to do.  I seem to remember one insttuction (moved under edit
with translation), which would move a value from one place to another,
convert formats (say from floating point or BCD to ASCII characters,
with a COBOL-like editting field to put in $-signs, commas, and
decomal points).  One seriously studly instruction :-), but it made
the COBOL benchmarks *scream*, much to the chagrin of CDC and
Burroughs sales people :-)

--tep



-- 
Tom E. Perrine (t...@SDSC.EDU) | San Diego Supercomputer Center 
http://www.sdsc.edu/~tep/     | Voice: +1.619.534.5000
I miss my 36-bit friends: Multics, TOPS-10, and TOPS-20.

In article <34860CF2...@monmouth.com>,
The Bakers <mba...@monmouth.com> wrote:

 >     1) Assembler language programming (IIRC, it was called "GMAP" -- at
least in
 > the GE world)

Any more-specific questions about GMAP?

In article <662aej$7...@tooting.netapp.com>,
g...@netapp.com (Guy Harris) wrote:

 > ...and a significantly different instruction set (e.g., I don't *think*
 > the 709x had as many addressing modes as the GE6xx).

NOBODY had as many addressing modes as the GE600 and follow-on machines. 
Not even the French postal service.  Nobody.

 > I vaguely remember somebody, perhaps here, saying why the GE6xx had an
 > instruction to convert Gray code to binary; anybody know why?

Because it was overengineered by engineers, for other engineers, is my
guess.  GE built it because they were one of IBM's major customers, and as
the 7094 aged they felt the need for a largely compatible (scratch the
System/360) machine for their own use, which costs they could defray by
selling them to other people as well.

I have heard that GTB was unused except for a quickie "encryption" scheme
that made us of it, but that story's a little too slick to be true.  Any
idea what Jack Whitney's or Charlie Fisher's e-mail address is, they'd
probably know.

Going through the plant that GE built in Phoenix, Arizona, to build
computers was a humbling experience.  (Tom knows this, of course, but for
the other readers...)  Every damn thing in the place had a GE label on it. 
And I don't mean a property identification label, I mean a manufacturer's
label.  From the water coolers to the ceiling girders to the electrical
panels to the computers (until Honeywell stepped in), that plant was
absolutely full of things manufactured by the General Electric Company.  It
was just amazing.

(The major exception that I recall was one area, and later a second one,
that was filled with Gardner-Denver wire-wrap robots -- the best
entertainment that side of the Mississippi.)

Edward Rice <ehr...@his.com> writes:
 
>Going through the plant that GE built in Phoenix, Arizona, to build
>computers was a humbling experience.  (Tom knows this, of course, but for
>the other readers...)  Every damn thing in the place had a GE label on it. 
>And I don't mean a property identification label, I mean a manufacturer's
>label.  From the water coolers to the ceiling girders to the electrical
>panels to the computers (until Honeywell stepped in), that plant was
>absolutely full of things manufactured by the General Electric Company.  It
>was just amazing.
 
Hmmm, might explain the "General" in the name :-)
Chris

Edward Rice, in article nr.<B0AE4EE1...@0.0.0.0>, wrote:
> 
> In article <662aej$7...@tooting.netapp.com>,
> g...@netapp.com (Guy Harris) wrote:
> 
>  > I vaguely remember somebody, perhaps here, saying why the GE6xx
>  > had an instruction to convert Gray code to binary; anybody know
>  > why?
> 
> Because it was overengineered by engineers, for other engineers, is
> my guess.

   The GE machines weren't alone in their posession of Gray code
conversions instructions. The Packard Bell 250 from 1961 also had
instructions to convert from Gray to binary and back again implemented
in hardware.

   Cheers.

-- 
 ______________________________________________________________________
|                                                |                     |
| Carl Richard Friend (UNIX Sysadmin)            | West Boylston       |
| Minicomputer Collector / Enthusiast            | Massachusetts, USA  |
| mailto:carl....@stoneweb.com                |                     |
| http://www.ultranet.com/~engelbrt/carl/museum  | ICBM: N42:22 W71:47 |
|________________________________________________|_____________________|

In a previous article, t...@galt.sdsc.edu (Tom Perrine) says:

>liked to do.  I seem to remember one insttuction (moved under edit
>with translation), which would move a value from one place to another,
>convert formats (say from floating point or BCD to ASCII characters,
>with a COBOL-like editting field to put in $-signs, commas, and
>decimal points).  One seriously studly instruction :-), but it made
>the COBOL benchmarks *scream*, much to the chagrin of CDC and
>Burroughs sales people :-)

An interesting instruction that was the invention of Jack Merner for the
Q-machines, and later lifted, right down to the numeric value of the
instruction code, and denied by its plagiarizers. They were confronted 
subsequently with their shoplifting and vigorously denied it, but there they
stood with egg all over their faces. 

There were many instructions in the Burroughs machines that were 
tailored to the language that was to be compiled and executed, something 
that was duly noted by the machine designers at Honeywell, and even 
perhaps (uh) `emulated.'

> >--tep > 
>-- >Tom E. Perrine (t...@SDSC.EDU) | San Diego Supercomputer Center 
>http://www.sdsc.edu/~tep/     | Voice: +1.619.534.5000
>I miss my 36-bit friends: Multics, TOPS-10, and TOPS-20.
>

(Previous discussion snipped)

How does this sort of instruction (albeit in Burroughs or GE or Honeywell, etc.)
differ from the IBM System 360's ED (Edit) and EDMK (Edit & Mark) machine
instructions ?  Did it predate the System 360 implementation ?  Did NCR, for
example, have an equivalent instruction in hardware (or hardware/firmware)
eventually too ?

M. Baker
mba...@monmouth.com

===================

In article <66cb1d$6...@rosebud.sdsc.edu>,
Tom Perrine <t...@galt.sdsc.edu> wrote:
[nsip]

>I was also extremely fortunate to serve a long internship under John
>Wertz, was one of the primary designers of the architecture of the
>instructions to be provided by "EIS box" (pronounced ICE box), which
>was the primary difference between the (original) 600 and 6000 series.
>
>As I recall, the EIS box was an option on later 600 models, the 6000
>had the functionality of the EIS box incorporated into the overall
>design of it CPU.

I think the later GE 6xx models may have actually been renamed to the
HIS 6xxx series.  The 655 (?) was the first 6xx series to use integrated
circuits, the 645 and earlier were discrete component or very small scale
integration.  I would not be surprised to find that the HIS 6180 was the
GE 655 with a new model number and name.  EIS was an option on the 6xxx
as far as I know, but was required for Multics, optional for GCOS.

RADC had 2 6180's, one without cache ran GCOS and the one with cache
ran Multics.  The Multics 6180 CPU failed and they tried to switch
to the GCOS 6180 and found the EIS hardware needed so much repair it
was quicker to repair the Multics 6180 and fix the GCOS 6180 later.
The maxload went from 36 without cache to 42 with cache.

>The EIS was the Extended Instruction Set, which had all kinds of
>complex instructions for supporting things that COBOL and FORTRAN
>liked to do.  I seem to remember one insttuction (moved under edit
>with translation), which would move a value from one place to another,
>convert formats (say from floating point or BCD to ASCII characters,
>with a COBOL-like editting field to put in $-signs, commas, and
>decomal points).  One seriously studly instruction :-), but it made
>the COBOL benchmarks *scream*, much to the chagrin of CDC and
>Burroughs sales people :-)
>Tom E. Perrine (t...@SDSC.EDU) | San Diego Supercomputer Center 

The EIS was real wierd.  Several instructions had both 2 and 3 operand
modes:  a=a <op> b and c = a <op> b.  the weird part was a, b, and c
could all be different data types and the appropriate conversions would
be done as needed.  The EIS string instructions oculd operation on up
to a segment (just under 1MB) of data at a time.

Charles Bertsch wrote:

> With General Electric, the assembler was call "GEMAP", Generalized
> something something.  When Honeywell bought the operation, they
> changed
> the name to GMAP, as well as producing GLOAD from GELOAD.  The OS had
> been
> named GECOS, Generalized Comprehensive ...  I really think that
> Honeywell
> should have called it "HOCUS" rather than GCOS.

   Yes, it was GCOS on the 6000, and the Level 66 and later the DPS
range of machines.
GCOS stood for General Comprehensive Operating System (but we preferred
to call it God's Chosen Operating System).  I cut my teeth on these
machines and have very fond memories of them.

It seems everything started with a G or a GE, a hang over from the
General Electric origins.  Remember the MME's?  MME GEBORT, MME GELOAD,
etc....

GCOS was a really neat OS, and way ahead of it's time.  True
multitasking on multi CPUs, timesharing support, a powerful file sysetm
(FMS), virtual memory support and lots of other features.

Those were the days......

Jim Macura

In article <B0AE4EE1...@0.0.0.0>, ehr...@his.com (Edward Rice) wrote:
>In article <662aej$7...@tooting.netapp.com>,
>g...@netapp.com (Guy Harris) wrote:
> > I vaguely remember somebody, perhaps here, saying why the GE6xx had an
> > instruction to convert Gray code to binary; anybody know why?
> [ ... ]
>I have heard that GTB was unused except for a quickie "encryption" scheme
>that made us of it, but that story's a little too slick to be true.  Any
>idea what Jack Whitney's or Charlie Fisher's e-mail address is, they'd
>probably know.

   It could have been used for instrumentation, I guess, but the only
use I ever saw was in the remote-access software -- .MROUT maybe?  --
where it, with a big whoopee comment, did even-parity generation on a
data byte.

        Regards.        Mel.

The Bakers <mba...@monmouth.com> wrote:

>This has been a great discussion so far!  Brings back lots of memories [no pun
>intended :]

>Dan Yertzell wrote:

>> The first line of the COBOL printouts
>> is:
>>
>> START OF CHECKOUT USING ABAVPA ON 74128 AT 0800 HOURS

>Here the next question:

>Did the H-200 have a built-in clock/calendar hardware (I don't *think* it did, at
>least the one I used) ?  Then where did the date & time come from on the header
>shown above from the COBOL compile ?

>Did you have to input the current date and time somehow, perhaps on the console
>call card ?

>Inquiring minds want to know.....
>M. Baker
>mba...@monmouth.com
It was an Optional peripheral called the Time of Day Clock, three
square feet of transistor logic.  Otherwise the operator entered the
date/time at boot time, or not if he was lazy. 
I only ever saw one (I was a H200 field engineer from 1970) . This one
developed an intermittent fault, but only at around midnight or later,
for about five minutes. It took me three goes to fix it, and it was
always refered to as the Time of Night Clock ever afterwards!

Bob Moffitt            			  	Recycle: Once is never enough
b...@skirrid.demon.co.uk(no-spam-please)		Remove braces to reply.         
http://www.skirrid.demon.co.uk/

In article <348E03E4...@monmouth.com>,
The Bakers <mba...@monmouth.com> wrote:

 > How does this sort of instruction (albeit in Burroughs or GE or
Honeywell, etc.)
 > differ from the IBM System 360's ED (Edit) and EDMK (Edit & Mark)
machine
 > instructions ?  Did it predate the System 360 implementation ?  Did NCR,
for
 > example, have an equivalent instruction in hardware (or
hardware/firmware)
 > eventually too ?

Honeywell's EIS instruction set was more comprehensive and more useful than
IBM's, but where both systems had a single instruction of the SS-type to do
a certain thing, then in general, the function was similar.  Honeywell had
VASTLY better addressing schemes than IBM did.

In article <348E84DF...@nospam.ace.co.uk>,
Jim Macura <j...@nospam.ace.co.uk> wrote:

 > GCOS was a really neat OS, and way ahead of it's time.  True
 > multitasking on multi CPUs, timesharing support, a powerful file sysetm
 > (FMS), virtual memory support and lots of other features.

Yes -- one note, though -- the system was really an almost-symmetric
multiprocesing system, from day 1.  It also had built-in system security
for both what was running and the file system.  There were holes, but it
wasn't a tacked-on bandaid solution such as IBM tried to foist off on their
customer base.

 > Those were the days......

"Honk if you love GECOS," as the bumper stickers used to say!

On Wednesday, December 3, 1997 at 2:00:00 AM UTC-6, The Bakers wrote:
> This has been a great discussion so far!  Brings back lots of memories [no pun
> intended :]
> 
> Dan Yertzell wrote:
> 
> > The first line of the COBOL printouts
> > is:
> >
> > START OF CHECKOUT USING ABAVPA ON 74128 AT 0800 HOURS
> 
> Here the next question:
> 
> Did the H-200 have a built-in clock/calendar hardware (I don't *think* it did, at
> least the one I used) ?  Then where did the date & time come from on the header
> shown above from the COBOL compile ?
> 
> Did you have to input the current date and time somehow, perhaps on the console
> call card ?
> 
> Inquiring minds want to know.....
> M. Baker

Hi all,

Not sure if I believe the posting dates I'm seeing... did Google Groups exist in 1997?

>From what I've been reading on the H200/2000 series, there was a "Time of Day" clock (peripheral model 213-4) that would supply the time. Not sure how date was supplied - perhaps it was part of the Operator's procedure to enter it each day. I seem to recall something like that being done by the Operator of the H2040 I had access to. I did not see any reference to an overflow interrupt/indicator for the time-of-day clock.

I've been working on a virtual H2000-series machine (http://honeywell2000.durgadas.com/) and am trying to get more information about peripherals - and also any software that might be out there. Even information about how software was operated could be useful.

Is anyone out there?

Thanks,
Doug

On Thursday, November 27, 1997 at 1:00:00 AM UTC-7, Charlie Gibbs wrote:

> And, further to another thread that's cropped up here from time
> to time, Honeywell even had their equivalent of the infamous IBM
> 2321 data cell.  The description of Honeywell's Mass Memory File
> is detailed enough to leave no room for doubt that it's the same
> sort of device.  Types 251, 252, and 253 offered storage capacities
> of 15, 63, and 317 million characters respectively.

When I think of the IBM 2321 data cell, it's usually in the terms "IBM even had 
their equivalent of the NCR CRAM device". CRAM, though, wasn't infamous as far 
as I know, it may have been reliable.

There was, at the time, a crying need for some sort of bulk storage that was 
cheaper than disk but fully automated, unlike tape. That meant there was a 
market for various unlikely attempts to meet that need.

So, many years later, IBM tried again with the 3850 mass storage system - the one with tape cylinders in honeycomb cells.

John Savard

On Wednesday, July 20, 2016 at 1:18:31 PM UTC-6, durg...@gmail.com wrote:

> Not sure if I believe the posting dates I'm seeing... did Google Groups exist in 1997?

USENET existed in 1997. And, in fact, so did Deja News, which was what Google 
bought and made Google Groups from.

John Savard

On Wednesday, July 20, 2016 at 2:56:00 PM UTC-5, Quadibloc wrote:
> On Wednesday, July 20, 2016 at 1:18:31 PM UTC-6, durg...@gmail.com wrote:
> 
> > Not sure if I believe the posting dates I'm seeing... did Google Groups exist in 1997?
> 
> USENET existed in 1997. And, in fact, so did Deja News, which was what Google 
> bought and made Google Groups from.
> 
> John Savard

Yeah, I realized after I replied that this was an old newsgroup.

On Wed, 20 Jul 2016 13:05:03 -0700 (PDT), durga...@gmail.com wrote:

>On Wednesday, July 20, 2016 at 2:56:00 PM UTC-5, Quadibloc wrote:
>> On Wednesday, July 20, 2016 at 1:18:31 PM UTC-6, durg...@gmail.com wrote:
>> 
>> > Not sure if I believe the posting dates I'm seeing... did Google Groups exist in 1997?
>> 
>> USENET existed in 1997. And, in fact, so did Deja News, which was what Google 
>> bought and made Google Groups from.
>> 
>> John Savard
>
>Yeah, I realized after I replied that this was an old newsgroup.

Well, you replied to an old post, but we still post here fairly
regularly... so its not an old newsgroup. But some of us are old
geezers.

-- 
JimP.

On Wednesday, July 20, 2016 at 3:46:01 PM UTC-5, JimP wrote:
> On Wed, 20 Jul 2016 13:05:03 -0700 (PDT), durga...@gmail.com wrote:
> 
> >On Wednesday, July 20, 2016 at 2:56:00 PM UTC-5, Quadibloc wrote:
> >> On Wednesday, July 20, 2016 at 1:18:31 PM UTC-6, durg...@gmail.com wrote:
> >> 
> >> > Not sure if I believe the posting dates I'm seeing... did Google Groups exist in 1997?
> >> 
> >> USENET existed in 1997. And, in fact, so did Deja News, which was what Google 
> >> bought and made Google Groups from.
> >> 
> >> John Savard
> >
> >Yeah, I realized after I replied that this was an old newsgroup.
> 
> Well, you replied to an old post, but we still post here fairly
> regularly... so its not an old newsgroup. But some of us are old
> geezers.
> 
> -- 
> JimP.

I dare say, most of us are! Who else would care about a Honeywell computer?

JimP <solo...@gmail.com> writes:

> On Wed, 20 Jul 2016 13:05:03 -0700 (PDT), durga...@gmail.com wrote:
>
>>On Wednesday, July 20, 2016 at 2:56:00 PM UTC-5, Quadibloc wrote:
>>> On Wednesday, July 20, 2016 at 1:18:31 PM UTC-6, durg...@gmail.com wrote:
>>> 
>>> > Not sure if I believe the posting dates I'm seeing... did Google Groups exist in 1997?
>>> 
>>> USENET existed in 1997. And, in fact, so did Deja News, which was what Google 
>>> bought and made Google Groups from.
>>> 
>>> John Savard
>>
>>Yeah, I realized after I replied that this was an old newsgroup.
>
> Well, you replied to an old post, but we still post here fairly
> regularly... so its not an old newsgroup. But some of us are old
> geezers.

Is there anywhere an archive of newgroup messages?  I'm suddenly curious
when a.f.c. started getting traction.    

On Wednesday, July 20, 2016 at 3:22:31 PM UTC-6, Lawrence Statton NK1G wrote:

> Is there anywhere an archive of newgroup messages?  I'm suddenly curious
> when a.f.c. started getting traction.

Why, yes, there is: and, in fact, it's that Google Groups thing we were talking about.

There used to be one called Deja News, which Google bought.

Of course, there are also "new" discussion groups that are Google Groups 
discussion groups, not part of USENET. You can tell which ones they are because 
they do have spaces in their names, and they don't have periods in their names 
- which is what tripped up the earlier poster.

John Savard

On Wed, 20 Jul 2016, Lawrence Statton NK1G wrote:

> JimP <solo...@gmail.com> writes:
>
>> On Wed, 20 Jul 2016 13:05:03 -0700 (PDT), durga...@gmail.com wrote:
>>
>>> On Wednesday, July 20, 2016 at 2:56:00 PM UTC-5, Quadibloc wrote:
>>>> On Wednesday, July 20, 2016 at 1:18:31 PM UTC-6, durg...@gmail.com wrote:
>>>>
>>>>> Not sure if I believe the posting dates I'm seeing... did Google Groups exist in 1997?
>>>>
>>>> USENET existed in 1997. And, in fact, so did Deja News, which was what Google
>>>> bought and made Google Groups from.
>>>>
>>>> John Savard
>>>
>>> Yeah, I realized after I replied that this was an old newsgroup.
>>
>> Well, you replied to an old post, but we still post here fairly
>> regularly... so its not an old newsgroup. But some of us are old
>> geezers.
>
> Is there anywhere an archive of newgroup messages?  I'm suddenly curious
> when a.f.c. started getting traction.
>
>
There were archives of control messages, but I've lost track of the URLs, 
so I don't know whether they are still out there.

I'm less certain if they archived control messages for alt newsgroups.

   Michael

Michael Black <et...@ncf.ca> writes:
>>
>> Is there anywhere an archive of newgroup messages?  I'm suddenly curious
>> when a.f.c. started getting traction.
>>
>>
> There were archives of control messages, but I've lost track of the
> URLs, so I don't know whether they are still out there.
>
> I'm less certain if they archived control messages for alt newsgroups.
>
>   Michael

ftp.isc.org has what appears to be a reasonably complete set.

 - From st...@gandalf.UMCS.Maine.EDU Sun Apr 26 10:52:56 1992
 - Path: rpi!usenet.coe.montana.edu!decwrl!bu.edu!dartvax!mars.caps.maine.edu!gandalf!steve
 - From: st...@gandalf.UMCS.Maine.EDU (Steve E. Goldsmith)
 - Newsgroups: alt.folklore.computers.ctl
 - Subject: newgroup alt.folklore.computers
 - Message-ID: <15...@gandalf.UMCS.Maine.EDU>
 - Date: 25 Apr 92 23:11:12 GMT
 - Control: newgroup alt.folklore.computers
 - Distribution: umcs
 - Organization: University of Maine Department of Computer Science
 - Lines: 0
 - Approved: st...@gandalf.UMCS.Maine.EDU

That seems later than I remember.

On Wed, 20 Jul 2016 15:45:05 -0500
JimP <solo...@gmail.com> wrote:

> Well, you replied to an old post, but we still post here fairly
> regularly... so its not an old newsgroup. But some of us are old
> geezers.

	We are indeed the Auld Farts of Computing.

-- 
Steve O'Hara-Smith                          |   Directable Mirror Arrays
C:>WIN                                      | A better way to focus the sun
The computer obeys and wins.                |    licences available see
You lose and Bill collects.                 |    http://www.sohara.org/

durga...@gmail.com wrote:
> On Wednesday, July 20, 2016 at 3:46:01 PM UTC-5, JimP wrote:
>> On Wed, 20 Jul 2016 13:05:03 -0700 (PDT), durga...@gmail.com wrote:
>>
>> >On Wednesday, July 20, 2016 at 2:56:00 PM UTC-5, Quadibloc wrote:
>> >> On Wednesday, July 20, 2016 at 1:18:31 PM UTC-6, durg...@gmail.com
wrote:
>> >>
>> >> > Not sure if I believe the posting dates I'm seeing... did Google
Groups exist in 1997?
>> >>
>> >> USENET existed in 1997. And, in fact, so did Deja News, which was what
Google
>> >> bought and made Google Groups from.
>> >>
>> >> John Savard
>> >
>> >Yeah, I realized after I replied that this was an old newsgroup.
>>
>> Well, you replied to an old post, but we still post here fairly
>> regularly... so its not an old newsgroup. But some of us are old
>> geezers.
>>
>> --
>> JimP.
>
> I dare say, most of us are! Who else would care about a Honeywell computer?

the people using that computer.

/BAH

On Thursday, July 21, 2016 at 6:21:39 AM UTC-6, jmfbahciv wrote:
> durga...@gmail.com wrote:
> > On Wednesday, July 20, 2016 at 3:46:01 PM UTC-5, JimP wrote:

> >> Well, you replied to an old post, but we still post here fairly
> >> regularly... so its not an old newsgroup. But some of us are old
> >> geezers.

> > I dare say, most of us are! Who else would care about a Honeywell computer?

> the people using that computer.

Since the Honeywell 200 computer was developed as a competitive alternative to 
migration to the System/360 for 1401 users, while many people used that 
computer at the time, the set of "people currently using a Honeywell 200 for 
production use" is very likely to have no members.

Of course, somebody in a computer museum somewhere may be turning one on 
occasionally to demonstrate it.

John Savard

Quadibloc wrote:
> On Thursday, July 21, 2016 at 6:21:39 AM UTC-6, jmfbahciv wrote:
>> durga...@gmail.com wrote:
>> > On Wednesday, July 20, 2016 at 3:46:01 PM UTC-5, JimP wrote:
>
>> >> Well, you replied to an old post, but we still post here fairly
>> >> regularly... so its not an old newsgroup. But some of us are old
>> >> geezers.
>
>> > I dare say, most of us are! Who else would care about a Honeywell
computer?
>
>> the people using that computer.
>
> Since the Honeywell 200 computer was developed as a competitive alternative
to
> migration to the System/360 for 1401 users, while many people used that
> computer at the time, the set of "people currently using a Honeywell 200 for
> production use" is very likely to have no members.
>
> Of course, somebody in a computer museum somewhere may be turning one on
> occasionally to demonstrate it.

There's no legacy code running in a deep dark hole?

/BAH

On Friday, July 22, 2016 at 5:58:30 AM UTC-6, jmfbahciv wrote:
 
> There's no legacy code running in a deep dark hole?

The Honeywell 200 is just a bit too old *now* for that to _still_ be true. Although in emulation on a modern PC, I suppose _that_ is remotely possible.

Of course, I can't speak to what might be going on behind the Schwarzschild 
radius.

John Savard

"jmfbahciv" <See....@aol.com> wrote in message 
news:PM000538382BD47765@aca41e1e.ipt.aol.com...
> Quadibloc wrote:
>> On Thursday, July 21, 2016 at 6:21:39 AM UTC-6, jmfbahciv wrote:
>>> durga...@gmail.com wrote:
>>> > On Wednesday, July 20, 2016 at 3:46:01 PM UTC-5, JimP wrote:
>>
>>> >> Well, you replied to an old post, but we still post here fairly
>>> >> regularly... so its not an old newsgroup. But some of us are old
>>> >> geezers.
>>
>>> > I dare say, most of us are! Who else would care about a Honeywell
> computer?
>>
>>> the people using that computer.
>>
>> Since the Honeywell 200 computer was developed as a competitive 
>> alternative
> to
>> migration to the System/360 for 1401 users, while many people used that
>> computer at the time, the set of "people currently using a Honeywell 200 
>> for
>> production use" is very likely to have no members.
>>
>> Of course, somebody in a computer museum somewhere may be turning one on
>> occasionally to demonstrate it.
>
> There's no legacy code running in a deep dark hole?
>

ISTM that most have converted their AUTOCODER programs to COBOL via some 
conversion program... at least by now.  But I'm sure that converted 
AUTOCODER is still running somewhere as we speak.

-- 

numerist at aquaporin4 dot com

On Friday, July 22, 2016 at 4:10:59 PM UTC-4, Charles Richmond wrote:

> > There's no legacy code running in a deep dark hole?

> ISTM that most have converted their AUTOCODER programs to COBOL via some 
> conversion program... at least by now.  But I'm sure that converted 
> AUTOCODER is still running somewhere as we speak.

I guess it's now a solid 20 years since we converted our Autocoder
emulation stuff.  The application was rewritten.

I suspect any Autocoder stuff left had to be rewritten in COBOL
to handle Y2K.

I do wonder if a user did want to run Autocoder if IBM still can
supply a simulator for it for the Z series.  I see no mention of
it on the IBM website.  But since it was available for machines
in the 1990s, there's probably one still out there if need be.

(Presumably, at some point the machines got so fast that they no
longer needed to bother with firmware emulation, and could run
1401 stuff strictly by software simulation.)

However, with S/360 technology now over 50 years ago, there's
probably a ton of applications written long ago still in service.
Stuff I wrote 35 years ago still runs, and there are applications
40 years old still running (much modified, of course, but the
basic core is still the same.)

Quadibloc wrote:
> On Friday, July 22, 2016 at 5:58:30 AM UTC-6, jmfbahciv wrote:
>
>> There's no legacy code running in a deep dark hole?
>
> The Honeywell 200 is just a bit too old *now* for that to _still_ be true.
Although in emulation on a modern PC, I suppose _that_ is remotely possible.
>
> Of course, I can't speak to what might be going on behind the Schwarzschild
> radius.

I made the comment because, every once in a while, this newsgroup gets
a surprise about some machine running. ;-)

/BAH