Honeywell 200
Dan Yertzell
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.
Dan
Charlie Gibbs
In article <65g3r9$8...@bgtnsc03.worldnet.att.net> Dan-
Yert...@worldnet.att.net (Dan Yertzell) writes:
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.
Heinz W. Wiggeshoff
"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! ^^^^^^^
The Bakers
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
Dan Yertzell
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
Dan Yertzell
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
The Bakers
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
Charlie Gibbs
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.
Julian Thomas
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.
William E. Boebert
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
Jim Haynes
> 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.
Edward Rice
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).
Edward Rice
> 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.
Dan Yertzell
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
Lisa or Jeff
> When I joined Honeywell in 1966, the company was booming in the 1401
> replacement market.
Wasn't a reason that the H-200 used a more modern electronics during
the time IBM was developing the S/360, and the S/360 was delayed coming
out?
Tom Perrine
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.
Guy Harris
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]
Sarr J. Blumson
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
Jim Haynes
In article <65vtvl$9...@netaxs.com>, Lisa or Jeff <hanc...@bbs.cpcn.com> wrote:
>> When I joined Honeywell in 1966, the company was booming in the 1401
>> replacement market.
>
>the time IBM was developing the S/360, and the S/360 was delayed coming
>out?
I don't think the electronics were more modern in the H-200; but it's
widely believed that the H-200 pushed IBM into announcing the S/360
sooner than they were ready to. I suppose the H-200 electronics were
more modern than the IBM 1401.
The Bakers
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
The Bakers
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
Charles Bertsch
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.
Edward Rice
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
> 1) Assembler language programming (IIRC, it was called "GMAP" -- at
least in
> the GE world)
Any more-specific questions about GMAP?
J. Chris Hausler
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.
Chris
Tom Perrine
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
Carl R. Friend
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 |
|________________________________________________|_____________________|
George Arndt
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.'
The Bakers
(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
===================
Richard Shetron
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.
Jim Macura
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
Bob Moffitt
The Bakers <mba...@monmouth.com> wrote:
>Dan Yertzell wrote:
>Here the next question:
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/
Guy Harris
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 ?
As I remember, the 6xxx EIS version did a lot more than ED or EDMK. (I
never looked at it in detail, and this was over 20 years ago....)
(BTW, the EIS box on the 60xx and 61xx wasn't microcoded; all that stuff
was hardwired....)
>Did it predate the System 360 implementation ?
The 6xxx EIS version didn't.
Mel Wilson
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.
Tom Van Vleck
Guy Harris wrote:
> As I remember, the 6xxx EIS version did a lot more than ED or EDMK. (I
> never looked at it in detail, and this was over 20 years ago....)
>
> (BTW, the EIS box on the 60xx and 61xx wasn't microcoded; all that stuff
> was hardwired....)
EIS added so many instructions to the CPU that another bit was
required for the opcode. Some EIS instructions were "multiword"
and were followed by two or three "descriptor" words (confusing,
not the same as a B5000 address descriptor or a 645 entry in the
descriptor segment). Descriptors contained an address, pointer-
register flag, and indirect modifier.
(7094 had indirect and 7 index register modifiers (ignoring Multiple
Tag Mode). 6xx and descendants generalized this into about 50 addressing
modes, including "calculate an effective address, using as many
indirect words as possible, and THEN add index 7 to the final answer."
But I digress.)
EIS instructions could operate on 4- 6- or 9-bit alphanumeric strings,
4- or 9-bit numeric strings, or bit strings. The EIS processor had
its own set of address registers. 20 opcodes dealt with loading,
storing, and modifying the address registers and the "pointers and
lengths." 7 opcodes with comparison, scanning, and translating.
PL/I index() in one instruction. 4 for moving. 3 for numeric move,
compare, and edit. 5 for bit strings. 2 for binary to decimal and back.
8 for decimal arithmetic, with 2- and 3-operand forms. And the move
with edit instructions used "micro operation codes" to direct the
edit, 17 of these. Maybe I miscounted, thought there were 30 opcodes..
All hardwired, as Guy says. As I remember, the H6180 was about
100 logc boards, say 18 inches by 2 feet each. Of these about
33 were the EIS uint. We used to describe the 6180 processor as
a 7094 in unnaturall congress with a 1401, with a lot of Multics
addressing decoration added on top.
Edward Rice
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!
Edward Rice
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.
durga...@gmail.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
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
Scott Lurndal
>o pun
>> intended :]
>>=20
>> Dan Yertzell wrote:
>>=20
>> > is:
>> >
>> > START OF CHECKOUT USING ABAVPA ON 74128 AT 0800 HOURS
>> Here the next question:
>>=20
>> 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 =
>> shown above from the COBOL compile ?
>> Did you have to input the current date and time somehow, perhaps on the c=
>onsole
>> call card ?
>>=20
>> M. Baker
>
>Hi all,
>
>st in 1997?
This isn't Google Groups. It is USENET, which dates from the late
70's.
Google Groups is just a pathetically horrible interface to USENET.
>
>From what I've been reading on the H200/2000 series, there was a "Time of D=
>ay" 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 ent=
>er it each day. I seem to recall something like that being done by the Oper=
>ator of the H2040 I had access to. I did not see any reference to an overfl=
when the system is booted (halt/loaded) from the operator.
Quadibloc
> 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
Quadibloc
> Not sure if I believe the posting dates I'm seeing... did Google Groups exist in 1997?
bought and made Google Groups from.
John Savard
durga...@gmail.com
JimP
regularly... so its not an old newsgroup. But some of us are old
geezers.
--
JimP.
hanc...@bbs.cpcn.com
> It was typical of mainframes in that era to request the current date/time
> when the system is booted (halt/loaded) from the operator.
time automatically from a consistent source, as opposed to the
console operator's wristwatch. Being even just a few minutes off
could make a difference in reconciling reports.
In my COBOL programs, I made it a point to display both the
compilation time (WHEN-COMPILED) and the execution date/time
in the I/O log. It did come in handy in debugging, like making
it easy to quickly to see what program version was being used,
and also the particular test. In addition, on reports, I included
the run time (HH:MM) after the date, knowing that some reports could
be run several times on the same day; this would identify such
multiple runs. Again, that came in handy when questions came up.
In FORTRAN, was there a way for a program to obtain the time
and date? I don't recall it in Fortran IV, but it was added
later or I just don't remember.
durga...@gmail.com
durga...@gmail.com
Lawrence Statton NK1G
when a.f.c. started getting traction.
Quadibloc
> Is there anywhere an archive of newgroup messages? I'm suddenly curious
> when a.f.c. started getting traction.
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
Morten Reistad
-- mrr
JimP
concern.
--
JimP.
Michael Black
so I don't know whether they are still out there.
I'm less certain if they archived control messages for alt newsgroups.
Michael
Jack Myers
> On Wednesday, December 3, 1997 at 2:00:00 AM UTC-6, The Bakers wrote:
> >
> > Did you have to input the current date and time somehow, perhaps on the console
> > call card ?
> >
> > Inquiring minds want to know.....
> > M. Baker
parameter. A large billing operation there have a sign on the wall
saying: Now processing <a date from a couple of days ago>.
The operators needed to know what date to key in for their
desparate attempt to rerun a failed batch job that blocked
the flow of production. Then there was overtime to work
through the backlog.
> Thanks,
> Doug
--
Postfix...a computer term meaning "sendmail is too hard for me"
Lawrence Statton NK1G
>>
>> 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
- 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.
Ahem A Rivet's Shot
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.
--
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/
gareth G4SDW GQRP #3339
news:p1356d-...@sambook.reistad.name...
Peter Flass
tablet and sending them by camel caravan.
--
Pete
jmfbahciv
> 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, durga...@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
>> >> 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?
/BAH
Scott Lurndal
>On the Honeywell, the time-of-day device was an "ordinary" peripheral and s=
>o it's possible you could find the right READ() syntax to retrieve the time=
> (but not date). I'm pretty sure FORTRAN IV could have vendor-specific buil=
>t-in functions/subroutines added, but I don't recall finding documentation =
>about that for the H2040 or OS/2000. But, I probably didn't look for it eit=
>her.
Burroughs had extensions in all the supported languages to obtain
the current time/date in various formats (e.g. julian). They generated a
Branch Communicate (BCT) instruction (system call to the MCP).
Kerr Mudd-John
wrote:
>
>
>
> http://www.laughingpoliceman.com/court_amateurs.htm
>
[]
take it to the funfair that is ukra
--
Bah, and indeed, Humbug
Quadibloc
> 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.
> the people using that computer.
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
> I don't even remember now, but I do recall chipping messages into a stone
> tablet and sending them by camel caravan.
chance?
John Savard
Charlie Gibbs
> It was typical of mainframes in that era to request the current date/time
> when the system is booted (halt/loaded) from the operator.
about the rapid miniaturization of computers, suggesting that one day
we might have a 3090 in a wristwatch.
"Great," he quipped, "you'll be able to boot MVS on your watch
and the first thing it'll do is ask you the time."
--
/~\ cgi...@kltpzyxm.invalid (Charlie Gibbs)
\ / I'm really at ac.dekanfrus if you read it the right way.
X Top-posted messages will probably be ignored. See RFC1855.
/ \ HTML will DEFINITELY be ignored. Join the ASCII ribbon campaign!
Ahem A Rivet's Shot
> On 2016-07-20, Scott Lurndal <sc...@slp53.sl.home> wrote:
>
> > It was typical of mainframes in that era to request the current
> > date/time when the system is booted (halt/loaded) from the operator.
>
> Which led to a rec.humor.funny posting from someone who saw an article
> about the rapid miniaturization of computers, suggesting that one day
> we might have a 3090 in a wristwatch.
>
> "Great," he quipped, "you'll be able to boot MVS on your watch
> and the first thing it'll do is ask you the time."
do just that.
Kerr Mudd-John
Peter Flass
--
Pete
David Wade
>> M. Baker
>
> Hi all,
> Not sure if I believe the posting dates I'm seeing... did Google Groups exist in 1997?
>
>
mains frequency. I remember the engineer showing it to me one day..
> 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
>
Dave
G4UGM
Charles Richmond
news:op.ykyss...@dell3100.workgroup...
http://www.aquaporin4.com/lispstone/
--
numerist at aquaporin4 dot com
Charles Richmond
news:nmri9u$54r$1...@news.albasani.net...
>>> 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.
>>
>
>
> From what I remember it was a "normal" clock that was synced to the mains
> frequency. I remember the engineer showing it to me one day..
>
will *not* be all that accurate...
Lawrence Statton NK1G
> The "mains" frequency can drift quite a bit. A clock based on that
> fequency will *not* be all that accurate...
For the last decade, there are 5.184 million +/- 4 cycles of 60 Hz per
day. (with an extra sixty for those days when we have leap seconds
... saw the report recently that we're due for another).
The maximum error can be 60 cycles (leading or lagging) before
correction must be applied to bring the total for the day back into line.
So -- a mechanically accurate synchronous clock will be off by no more
than one second over the course of a day, and no more than two seconds
per month in the long term.
Morten Reistad
for the UK and Nordpool (Scandinavia ++) grids in Europe. But if
you go to China, India, Africa this is not even close to true anymore.
Grid-synced clocks may drift several minutes per day.
-- mrr
Kerr Mudd-John
jmfbahciv
/BAH
Quadibloc
> There's no legacy code running in a deep dark hole?
Of course, I can't speak to what might be going on behind the Schwarzschild
radius.
John Savard
William Pechter
I remember building Bnews and Cnews and running them both out of my house
with rn and trn and tin running on various boxes from a Xenix86 box
on and AT&T 6300 and later on a couple of MicroXelos (UniPlus SysV)
(Concurrent Computer XF200 and Perkin-Elmer 7300) boxes and finally
moving to FreeBSD/NetBSD/Linux on a 2nd hand 386sx...
We supplied a number of news groups for free around a three county
area using select forward to avoid having to pay phone charges. For
$5.00/month NJBell would let me forward to a friends modem line and
he/she would be either receiving the news or forwarding to the next
person on the chain by routing my modem line to the next number in
the UUCP chain.
Sometimes we had up to 8 forwarding points me->fwd1_modem_line->
fwd2->modem_line... all the way down to the 8th number in the calling
chain.
We pretty much eliminated any connection charges beyond our flat-rate
phone line charges. Thanks to one of my late wife's roommates who
tipped us to this available service. They also had 17 phone lines
into the house for an early modem chat system, two BBS's and our
own modem line.
Worked great and was blowing the minds of local BBS users who all wanted
to put it into Fido or Citadel...
Bill
--
--
Digital had it then. Don't you wish you could buy it now!
pechter-at-gmail.com http://xkcd.com/705/
Bob Eager
> Heinlein's books and talking to other people that enjoyed them, I wanted
> to like the Barsoom books.
>
> I didn't particularly care for the several that I read, although they
> were better than the Tarzan books I tried. I kept thinking, "When is it
> going to get good, like I've heard people say it is?"
>
> Heinlein also talked up EE Smith. I tried some of the Lensman books and
> Skylark books and they're not my cup of tea either.
>
> I'll go for a trifecta. One of my friends tried to get me interested in
> the Conan books. They put me to sleep.
>
> I'm sure there is someone out there who will denounce me for not knowing
> great fiction when I see it, but I'll just have to hang my head in shame
> and live with it.
I had exactly the same experience as you. You are not alone.
--
Using UNIX since v6 (1975)...
Use the BIG mirror service in the UK:
http://www.mirrorservice.org
Quadibloc
> On Fri, 22 Jul 2016 14:22:33 -0400, Bud Frede wrote:
> > After reading a number of references to ER Burroughs in Robert
> > Heinlein's books and talking to other people that enjoyed them, I wanted
> > to like the Barsoom books.
> > I didn't particularly care for the several that I read, although they
> > were better than the Tarzan books I tried. I kept thinking, "When is it
> > going to get good, like I've heard people say it is?"
meat to it, something that deals with ideas - you will be disappointed in Edgar
Rice Burroughs' Barsoom novels.
All they are is melodramatic adventures that give the reader a rollicking good
time - like, say, the James Bond novels of Ian Fleming. And not everyone enjoys
those either; if you _are_ looking for what someone called "cotton candy for
the mind", you will still also want to choose a particular genre that is
gratifying to you.
Since the Burroughs oeuvre consists of deathless literature that is still not
great literature, not... finding it useful... is nothing to be ashamed of.
John Savard
ma...@mail.com
> Peter Flass <peter...@yahoo.com> writes:
>
> Heinlein's books and talking to other people that enjoyed them, I wanted
> to like the Barsoom books.
>
> I didn't particularly care for the several that I read, although they
> were better than the Tarzan books I tried. I kept thinking, "When is it
> going to get good, like I've heard people say it is?"
>
> Skylark books and they're not my cup of tea either.
>
> I'll go for a trifecta. One of my friends tried to get me interested in
> the Conan books. They put me to sleep.
>
> I'm sure there is someone out there who will denounce me for not knowing
> great fiction when I see it, but I'll just have to hang my head in shame
> and live with it.
a good writer, but a psyc* would spend weeks dredging his deepest needs.
Then psych* aare full of shi* anyway. I must reread Clarke's
"The City and the Stars" again.
--
greymaus.ireland.ie
.
.
...
Quadibloc
> Then psych* aare full of shi* anyway.
that sentiment... but then, they thought he was practicing quackery in their
field.
John Savard
Charles Richmond
news:PM0005383...@aca41e1e.ipt.aol.com...
conversion program... at least by now. But I'm sure that converted
AUTOCODER is still running somewhere as we speak.
hanc...@bbs.cpcn.com
> Almost all fantasy-fiction books, say 1880-1950 were racist. Heinlein was
> a good writer, but a psyc* would spend weeks dredging his deepest needs.
> Then psych* aare full of shi* anyway. I must reread Clarke's
> "The City and the Stars" again.
Mark Twain in public school due to its racism and liberal use of
the n-word, as in "nobody was killed except some n-----s".
I read a 1910 engineering book which explicitly recommended
against utilizing "colored labor", despite the cheaper price.
Even Eleanor Roosevelt's memoirs--going by today's standards--
would not pass muster with political correctness, and she was a
fighter for civil rights.
hanc...@bbs.cpcn.com
> > 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.
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.)
JimP
nowhere, Texas. If you've ever seen the cartoon 'Courage the Cowardly
Dog', it wasn't much different than the area I lived in way back when.
Maybe a few more houses, and a few small towns. But mostly nothing.
Harlan Ellison scared me, I was under 10 years of age when I read his
books for the first time.
--
JimP.
J. Clarke
hanc...@bbs.cpcn.com says...
> On Friday, July 22, 2016 at 3:34:06 PM UTC-4, ma...@mail.com wrote:
>
> > Almost all fantasy-fiction books, say 1880-1950 were racist. Heinlein was
> > a good writer, but a psyc* would spend weeks dredging his deepest needs.
> > Then psych* aare full of shi* anyway. I must reread Clarke's
> > "The City and the Stars" again.
>
> Lots of old books were racist. I was surprised they had us read
> Mark Twain in public school due to its racism and liberal use of
> the n-word, as in "nobody was killed except some n-----s".
racist is reaching.
ma...@mail.com
> ma...@mail.com writes:
>
>
> I can make some allowances for racism, at least if I feel there are
> other redeeming features to the book.
>
> I read "The City and the Stars" again last year and enjoyed it as much
> as ever. He wrote some really great stories, although I'm not as fond of
> some of the things he collaborated on later in life.
>
Like pterry. I suppose that at a certain time, one feels the need to get
as mucch cash together to provide for who is left.
Ahem A Rivet's Shot
Bud Frede <fr...@mouse-potato.com> wrote:
> I read "The City and the Stars" again last year and enjoyed it as much
> as ever. He wrote some really great stories, although I'm not as fond of
> some of the things he collaborated on later in life.
happy with some of them too (IIRC it was Garden of Rama he made a
disparaging remark about). As for the Baxter collaborations they're pretty
much pure Baxter AFAICS - Baxter never has worked out how to finish a story,
which Clarke was *very* good at.
Peter Flass
my mind at that time that equated to 1401. Maybe it was close enough that
programs can be run on an emulated 1401?
--
Pete
Peter Flass
> Heinlein's books and talking to other people that enjoyed them, I wanted
> to like the Barsoom books.
>
> I didn't particularly care for the several that I read, although they
> were better than the Tarzan books I tried. I kept thinking, "When is it
> going to get good, like I've heard people say it is?"
have gotten excited by them, and usually whatever you like at age 10
remains a sentimental favorite. Remember this was the time Lowell was
seeing "canals" on Mars, so thoughts of martians and lost civilizations
were in the air.
I read the Barsoom Series when I was a teenager, and I enjoyed them, but
thought they weren't in the same class as RAH and ACC.
>
> Heinlein also talked up EE Smith. I tried some of the Lensman books and
> Skylark books and they're not my cup of tea either.
think most early SF is dated, with wooden characters, clunky plots, and
science that didn't hold up. I recently read one of the Lensman books and
was startled to see references to atomic energy, but I now understand that
Smith went back and revised some of his stuff.
>
> I'll go for a trifecta. One of my friends tried to get me interested in
> the Conan books. They put me to sleep.
>
> I'm sure there is someone out there who will denounce me for not knowing
> great fiction when I see it, but I'll just have to hang my head in shame
> and live with it.
>
Pete
Peter Flass
--
Pete
Peter Flass
>
>> On Friday, July 22, 2016 at 12:51:09 PM UTC-6, Bob Eager wrote:
>>> On Fri, 22 Jul 2016 14:22:33 -0400, Bud Frede wrote:
>>
>>>> After reading a number of references to ER Burroughs in Robert
>>>> Heinlein's books and talking to other people that enjoyed them, I wanted
>>>> to like the Barsoom books.
>>
>>>> I didn't particularly care for the several that I read, although they
>>>> were better than the Tarzan books I tried. I kept thinking, "When is it
>>>> going to get good, like I've heard people say it is?"
>>
>>> I had exactly the same experience as you. You are not alone.
>>
>> Well, if you're looking for something to take seriously - something that has
>> meat to it, something that deals with ideas - you will be disappointed in Edgar
>> Rice Burroughs' Barsoom novels.
>
> anything very deep.
>>
>> All they are is melodramatic adventures that give the reader a rollicking good
>> time - like, say, the James Bond novels of Ian Fleming. And not everyone enjoys
>> those either; if you _are_ looking for what someone called "cotton candy for
>> the mind", you will still also want to choose a particular genre that is
>> gratifying to you.
>
> first read them as a teenager. I also really like Chandler and
> Hammett. I've even read a Mickey Spillane novel and it killed some
> time. :) I got stuck overnight at an airport and there wasn't anything
> really worth reading, and beggars can't be choosers...
I'm not sure I ever read Hammett, but Chandler seems to have held up quite
well. One thing I like are his descriptions of an L.A. that's long gone.
>
>>
>> Since the Burroughs oeuvre consists of deathless literature that is still not
>> great literature, not... finding it useful... is nothing to be ashamed of.
>
> you can't even use a book for warmth or to wipe your posterior. (The
> Sears website just would not have done the trick when most houses had an outhouse.
>
>
--
Pete
Peter Flass
hasn't changed much.
>
> Harlan Ellison scared me, I was under 10 years of age when I read his
> books for the first time.
>
--
Peter Flass
the "more" is so inferior to the original.
--
Pete
Peter Flass
> On Friday, July 22, 2016 at 3:34:06 PM UTC-4, ma...@mail.com wrote:
>
>> Almost all fantasy-fiction books, say 1880-1950 were racist. Heinlein was
>> a good writer, but a psyc* would spend weeks dredging his deepest needs.
>> Then psych* aare full of shi* anyway. I must reread Clarke's
>> "The City and the Stars" again.
>
> Lots of old books were racist. I was surprised they had us read
> Mark Twain in public school due to its racism and liberal use of
> the n-word, as in "nobody was killed except some n-----s".
groundbreaking for the time. Jim was shown as a human being and not a
watermellon eating jive-monkey.
>
> I read a 1910 engineering book which explicitly recommended
> against utilizing "colored labor", despite the cheaper price.
>
> Even Eleanor Roosevelt's memoirs--going by today's standards--
> would not pass muster with political correctness, and she was a
> fighter for civil rights.
>
--
Pete
Quadibloc
> In article <798342bb-e74b-4f0a...@googlegroups.com>,
> hanc...@bbs.cpcn.com says...
> > On Friday, July 22, 2016 at 3:34:06 PM UTC-4, ma...@mail.com wrote:
> > > Almost all fantasy-fiction books, say 1880-1950 were racist. Heinlein was
> > > a good writer, but a psyc* would spend weeks dredging his deepest needs.
> > > Then psych* aare full of shi* anyway. I must reread Clarke's
> > > "The City and the Stars" again.
> > Lots of old books were racist. I was surprised they had us read
> > Mark Twain in public school due to its racism and liberal use of
> > the n-word, as in "nobody was killed except some n-----s".
> For certain values. Twain was an abolitionist. Painting him as a
> racist is reaching.
his innocence and naivete, having difficulties understanding why it is wrong,
as the adult world claims, to help a runaway slave to escape.
For Twain to attempt to shield his readers from words that were used every day
in his time would then have been quite useless, only making him subject to
ridicule. When slavery was the law of the land, the battle for what is itself
ridiculed by calling it political correctness - originally a Soviet term - lay
in the future.
John Savard
jmfbahciv
a surprise about some machine running. ;-)
/BAH
Ahem A Rivet's Shot
give them too much.
durga...@gmail.com
I recall reading about a program/product called "Liberator" to help attract IBM 1401 customers. What little I read on the 1401 made me think it was decimal-only (even the address bus/memory decoders were decimal), while the Honeywell 200/2000 series include binary operations (and actually used binary electronics to address memory, etc, internally). I'm not sure of the details, but have seen mention of features in the 200/2000 docs that might help, like the "trap on Item Mark" feature that could allow for emulating 1401 instructions on the fly.
durga...@gmail.com
J. Clarke
yaho...@news.eternal-september.org>, peter...@yahoo.com says...
> Bud Frede <fr...@mouse-potato.com> wrote:
> > Peter Flass <peter...@yahoo.com> writes:
> >
> >> Quadibloc <jsa...@ecn.ab.ca> wrote:
> >>> On Thursday, July 21, 2016 at 6:02:07 AM UTC-6, Peter Flass wrote:
> >>>
> >>>> I don't even remember now, but I do recall chipping messages into a stone
> >>>> tablet and sending them by camel caravan.
> >>>
> >>> Have you visited Mars - after an encounter with hostile Apache Indians - by any
> >>> chance?
> >>>
> >>
> >> Yes, I left my thoat double-parked when I came back.
> >
> > After reading a number of references to ER Burroughs in Robert
> > Heinlein's books and talking to other people that enjoyed them, I wanted
> > to like the Barsoom books.
> >
> > I didn't particularly care for the several that I read, although they
> > were better than the Tarzan books I tried. I kept thinking, "When is it
> > going to get good, like I've heard people say it is?"
>
> I think if you were 10 at the time the books came out (1920s?) you would
> have gotten excited by them, and usually whatever you like at age 10
> remains a sentimental favorite. Remember this was the time Lowell was
> seeing "canals" on Mars, so thoughts of martians and lost civilizations
> were in the air.
> I read the Barsoom Series when I was a teenager, and I enjoyed them, but
> thought they weren't in the same class as RAH and ACC.
>
> >
> > Heinlein also talked up EE Smith. I tried some of the Lensman books and
> > Skylark books and they're not my cup of tea either.
>
> Same scenario. I read one or two many years later. Unfortunately I now
> think most early SF is dated, with wooden characters, clunky plots, and
> science that didn't hold up. I recently read one of the Lensman books and
> was startled to see references to atomic energy, but I now understand that
> Smith went back and revised some of his stuff.
you? The first use of the term "atomic bomb" in a story was by HG Wells
in 1914. Heinlein wrote two stories involving nuclear energy before the
Manhattan Project was started--one pretty much predicted the Cold War
and the other involved a power plant accident.
As for "science that didn't hold up", one large subgenre of science
fiction involves fictional science. Why should fictional science "hold
up"? Smith was a PhD engineer--when he was using fictional science he
knew he was using it and did it by intent.
ma...@mail.com
> On Sat, 23 Jul 2016 08:24:35 -0400
> Peter Flass <peter...@yahoo.com> wrote:
>
>> <ma...@mail.com> wrote:
>> > Like pterry. I suppose that at a certain time, one feels the need to get
>> > as mucch cash together to provide for who is left.
>>
>> Something like Rama leaves the reader wanting more, but it's a shame when
>> the "more" is so inferior to the original.
>
> It is always better to leave your audience wanting more than to
> give them too much.
>
J. Clarke
ste...@eircom.net says...
> On Sat, 23 Jul 2016 08:24:35 -0400
> Peter Flass <peter...@yahoo.com> wrote:
>
> > <ma...@mail.com> wrote:
> > > On 2016-07-23, Bud Frede <fr...@mouse-potato.com> wrote:
> > >> ma...@mail.com writes:
> > >>
> > >>
> > >> I can make some allowances for racism, at least if I feel there are
> > >> other redeeming features to the book.
> > >>
> > >> I read "The City and the Stars" again last year and enjoyed it as much
> > >> as ever. He wrote some really great stories, although I'm not as fond
> > >> of some of the things he collaborated on later in life.
> > >>
> > >
> > > Like pterry. I suppose that at a certain time, one feels the need to get
> > > as mucch cash together to provide for who is left.
> >
> > Something like Rama leaves the reader wanting more, but it's a shame when
> > the "more" is so inferior to the original.
>
> It is always better to leave your audience wanting more than to
> give them too much.