return to main page

1401 Demo Decks & Planning

Started November 9th, 2006
Updated - April 16, 2008


Goal of this web page:

Present demo software, ideas and plans for 1401 demonstration(s)

Demo Software List Newest near top (here)



Demo Planning activity
-
- Presentation to Senior Docents Feb 3, 2007
- Known Short Term Challenges (our delivery of the demo in February)
- Presentation to Senior Managenment March 21, 2007
- Presentation to Senior Docents May 22, 2007
- David Cortesi's comments November 28, 2007
- 1401 proposed input-output instructions .pdf from David Cortesi - Nov 20, 2007
- 1401 proposed demo script .pdf from David Cortesi - Nov 20
- stacker select, 'X' punch Bill Worthington December 13, 2007 4:01 PM
- Robert Garner complicated - December 13, 2007 6:52 PM
- HzG demo program from Van Snyder December 13, 2007 7:29 PM
- Dec 12, 2007 Meeting Notes Dave Cortesi December 13, 2007 12:49 PM
- Known Challenges, and a dream (complex story, to wide range of folks, in never enough time)
- Project Inputs (and assumptions)
- Who are the Project Customers?,
- Demo Project Goals and History,


- other E-mails with suggestions


"HzG demo program" code

from Van Snyder December 13, 2007 7:29 PM


On Thu, 2007-12-13 at 18:52 -0800, Robert B Garner wrote:
> I think the tapes could be running continuously in the background 
> starting at some point in the demo (perhaps just after the 
> deck is read in and before the 1403 is started, via a sense switch). 
> As soon as the 1403 is finished printing, the tapes could go into 
> their high-speed rewind fling. 
One could maybe snarf some code from the "HzG demo program" code (or add to it). It writes up to four tapes, with record lengths, delays, number of records until rewind, etc selected by a control card, and which tapes to use selected by sense switches (so you can use none of them if they're broken).

It reads a deck, stores it in memory, and prints it as long as a sense switch is on. It shouldn't be too hard to have it read a card, create the box letters, then print them and the stored deck, then loop for another card.

The attached file exercise.hzg.zip is my reverse engineering of the HzG program. It also jiggles the head on the disk drive. The attached file exercise.nodisk.zip is my re-engineering of the HzG program. It doesn't jiggle the head on the disk.

Van


Complicated

from Robert Garner December 13, 2007 6:52 PM
David,

Thanks for continuing to drive the revised demo scenario! I like the updated/general structure for the scenario.

Nevertheless--hopefully not sounding pessimistic -- this scenario may still be a bit too complicated (for us/docents).

In particular, I'm not sure we want to depend on 729 tape drives for the entire demo to work (i.e, be dependent on certificates on a master tape). Why increase the complexity and fallibility by creating a tape "data base" that the demo depends on? What if the 1401 TAU was flaky, or the certificate master tape was damaged or misplaced, etc?

Ed's demo code is all on cards, which I think is safer (with a couple of backup decks, of course. For an additional backup, it might be wise to have a master certificate on a tape, in case the 1402 is having a bad day and can't read thru an entire thick deck, but can read a short stack of visitor names.)

Ed's current demo deck already includes a date input. Also, much of the "certificate" text is an introduction to the restoration project, the 1401 and its technology for at home reading (adopted from my 1401 tutorial paper that's on web site.)

The reality is that people can NOT really tell if certificates are being read from a master tape or not. (However, they may notice the thick demo/certificate card deck being read in.) Unless one is very technical, no-one can really tell what is going on (mater tape, sequence numbers, etc.) I would save this level of vintage authenticity for Ron or David's 1401 hands-on programming class.

Per our Weds meeting discussion, in lieu of Sort7 running (we won't have four tape drives up for a while yet), my suggestion is to augment Ed's demo program to write and read random-length records on whatever number of tape drives are up and running that day. (Or could be pattered accesses for interesting effects to simulate Sort7 running or some other vintage application.) Number of working drives can be entered in sense switches, incl 0.

I think the tapes could be running continuously in the background starting at some point in the demo (perhaps just after the deck is read in and before the 1403 is started, via a sense switch). As soon as the 1403 is finished printing, the tapes could go into their high-speed rewind fling.

It would be good to call another meeting to continue to discuss options.

- Robert


stacker select, 'X' punch

from Bill Worthington December 13, 2007 4:01 PM
Dave, this sounds like a good summary. Thanks for documenting it.

I would make two suggestions for the demo.

Use the stacker select for the "detail" cards so that the attendee information is separated from the program data. We could then include the detail card as part of the souvenir of the demo. Stan had some great cards with the Museum's logo on the back which could be used for this purpose which do not have to be four color since the world was black and white (or green bar) in those days.

Don't forget the "X" punch in column 80 to indicate that it is the date card. Maybe this should be selected into the third stacker for card input data. This was the standard for all customers who migrated from tab machines since that was the way a date card was identified. (OK! Three items!!) I wonder if the sign-up sheet should be filled in before the tour starts. This would mean that the attendees would not be distracted trying to fill out the sheet while they are being shown the unit record machines in the first aisle of Visible Storage. Hence the volunteer who keypunches the detail card would also have enough time to get the cards punched before the tour arrives in the restoration room.

Did anything more transpire beside your discussion with Ron on the demo after I had to leave? I'm sure there was more commentary on how the time should be allocated. Was there more?

Regards, Bill


Dec 12, 2007 Meeting Notes

from Dave Cortesi December 13, 2007 12:49 PM
1401 software persons,

After yesterday's meeting on 1401 Demo plans, Ron & I kicked around ideas for the software to run during the demo.

Here is the context: It was determined in the meeting that the first 1/3 of the demo takes place in Visual Storage with talk on unit-record methodology.

Prior to and during this portion, audience will be encouraged to print (!) their names on a sign-up sheet, and a volunteer will take this away and keypunch one card per name while the talk is going on.

Then the demo moves to the ramp of the 1401 machine room to see "what followed on and replaced" unit record methods. (At this time, the keypunch volunteer would probably still be punching cards or will finish up the last card so the audience can see and hear the 029 in operation.)

At this point the 1401 would be put in operation running the demo application. The objectives of this software would be:

0. Simulate the general look/feel of a master-detail update job:

a. "detail" input from cards
b. "old master" input from tape
c. "new master" output to tape if 2nd drive available

1. Exercise & demonstrate the 1402 reader

2. Exercise & demonstrate the 1403 printer

a. producing output the viewers can take away as souvenir

3. Exercise & demonstrate 729 tapes,

a. minimum of 1 tape
b. preferably 2 tapes if available
c. use sense-switch input to select which tape(s) to use

5. Run off at least 3/4" (radial) of a tape reel so as to trigger high speed rewind

What we propose is that the output (2 above) would be a one-sheet "Certificate of Attendance" document stating that JOE BLOW (name from punched "detail" card) has attended this demo, blah-de-blah -- suitably amusing a/o informative certificate text TBS.

Each viewer's name would be printed in big block letters (using Ed's code) in the middle of the "Certificate" and there might be other bits of ascii art as flourishes in the corners, decorative borders, whatever.

Now in order to take care of objective (3) we would have the boiler-plate text of the "Certificate" on the "old master" tape, one record per print line (thus, 40-60 records per sheet of output). There would be many, many, MANY copies of this boilerplate text on the tape.

In order to simulate master-update, when the 2nd drive is available, the program would copy each input record to the output tape.

In order to use up lots of tape (objective 5) the program would skip a random number of certificates between each input "detail" card. This also simulates detail/master updating.

A possible enhancement is that the printed "certificates" are sequentially numbered, with the number incrementing over time so that they reflect the actual count of people who have EVER seen the demo. Thus, "JOE BLOW was viewer #423 to see this demo". This would require getting the starting number from somewhere, updating it for each viewer, and recording the new starting number at the end of the run. That could be done either punching a summary card (demonstrates the 1402 punch if available) or by rewinding and overwriting record 1 of the "new master" (or the "old master" if using only one tape).

Another enhancement would be having today's date in the "certificate" text. This would require punching a date card and reading it in at the start of the run. (Same card could have the starting certificate number, if on card.) (Or could take the date from a designated field of the first detail card.)

So that is the "functional spec" so far. Comments and improvements are welcome!

Even more welcome? Somebody volunteering to write & test this puppy!

Thanks,

Dave Cortesi


Demo Project Goals and History

I/we would like a visit to the restored IBM 1401 system to be *the* high light of a tour to the Computer History Museum.

a) see a working machine read cards, print, spin tape.
b) key punch a card on an old 026 key punch, see it go through a restored IBM 083 sorter (just for fun)
then read into the 1402 card reader, and have the printout of their name to take home.

In the September 13th 2006 1401 Project meeting, Robert Garner stated:

The Museum's goal is to start 1401 public demos in February. Several things must fall into place for that to happen: We need to have a demo scenario worked out and practiced, a schedule and volunteer scheduling system needs to be in place, and we need to confirm that the 1401 will run without failures for several months. (btw, only one of the 729 drives needs to work.)

and the following seemed agreeable to all - viewable in large print at the above web site -

  • The rest of the meeting we collected ideas for possible demo scenarios. We agreed that music playing on the 1403 would be too stressful on the chain, so that will not happen.**

  • We concurred that the criteria for a good demo is one that is interesting/fun, involves the visitors, and typifies how the 1401 was used in its time. For example, we noted that a payroll program itself my not meet these criteria, a game on the panel switches is not prototypical, and almost anything that shows off the peripheral devices will be fun and engaging.

  • We agreed that it would be ideal if the demo visitors (or a subset) could punch say their name + whatever on a punched card. We could then sort the newly-punched cards on say the 083 or 077 collator and then print them in large block font on the 1403 together with some other items. These other items could include an imitation check written out to the person ;-), job accounting charge (in today's dollars), some information about the 1401 (i.e., speed, cost of lease and memory in today's dollars, etc.), and some vintage printer art.

  • Several folks suggested that each phase of the demo program could be controlled by 1401 sense switches. That way a particular phase could always be skipped (if a peripheral were down) or repeated.

  • We agreed there should be some artifacts that the demo person shows/passes around to the visitors when describing the 1401: plug board, core memory, SMS card, punch cards, mag tape, etc.

  • We agreed we should make a video of a representative live demo that can be viewed on-line or in lieu of the demo itself.

  • It was suggested that the machines, such as the 1402 card reader and 729 tape drive, would look more impressive if one or more of their steel skins were replaced with plexiglas.

  • We contemplated whether two 026 punch units would be sufficient. We may need a third? On the other hand, some visitors might enjoy watching others (younger kids?) punch a card, passing up themselves.

Jeff Stutzman agreed to lead the effort -

then in October, tragedy struck - Jeff got a job :-((

So - what to do? In a show of real desperation, November 8th, Robert Garner asked Ed Thelen (me) to take over from Jeff.

I kid you not!
Well, time is short, with a long list of serious holidays just around the corner - and a short fuse.

Fortunately a number of people e-mailed suggestions to Jeff,

and there is a large pool of hardware and software talent to beg, borrow, steal, ... from.


Who are the Project Customers?

OK, who is the customer?
Personally I don't care about museum connoisseurs, with check lists of the current museum fads.
I would like "just plane folks" to have an enjoyable, memorable, maybe even educational experience.
And I bet the connoisseurs and "heavy hitters" would enjoy themselves also -

And who are "just plane folks" - let's take a Norman Rockwell Thanksgiving painting - and bring these folks to see "our" 1401. These "just plane folks" include

- some teen aged kids - (we left the younger kids with the neighbors)
lets say the hormones are on hold for the day, - have never "worked on" a car or cracked open a cell phone
- a couple of college age "kids" of varied interests and majors
a political science major and a techie - never saw a transistor, vacuum tube, IBM card
- parents who remember
radios with tubes, black & white TV, "don't fold, spindle, or mutilate this bill" (returnable IBM card),
- gramps and grandma -
gramps spent many hours on big monster computers, and grandma talks of keypunching cards

Or one can say the above is a typical collection of people who have just met on a tour of Computer History Museum.

A widely diverse group, a typical group?
How to get our 1401 demo to appeal to all of the above?
How get their enthusiastic endorsement? How to get them to tell their neighbors of this real cool place?
A serious challenge!!

Project Inputs (and assumptions)

  1. A place with 50Hz power, security, roof, parking, cleaning and trash hauling ... - Computer History Museum is ideal.

  2. (nice to have) An old fashoned computer room, with raised floor, glass wall, IBM clock on wall, tab equipment about -
    As Steve Jobs might say, the current 1401 restoration lab is "INSANELY GREAT" ;-))

  3. Working equipment, including key punches, 1401, 1402, 1403, at least one workng 729 tape unit

  4. A staff of knowledgeable, (OJT?) trained, folks to keep the above going. With schematics, scopes, spares, cabinets, ...

  5. Knowledgeable docents to demonstrate, help with fumble fingered amateur key punchers, crowd control, ...

  6. A plan, script, story board, ... to keep focus and sequence.

  7. Software to support the story board ;-))

  8. Props, signs, posters, handouts, expendables including IBM cards, printer paper, snippets of mag tape, ...


Known Challenges, and a dream - (complex story, to wide range of folks, in never enough time)

Dear Friends -

Do have an impossible mission? - "Mission Impossible?" - We would like to tell our diverse group of folks many hours of history, facts and folklore. Most of us would love to make a curriculum of maybe one hour a day for a college quarter to give the folks the flavor of the times, the technology - with past and present progress, the story of IBM and how tabulating cards revolutionized scientific calculating and business accounting which hadn't changed much since "Arabic" numerals replaced Roman numerals many centuries before. The book keeper and calculator with the green eyeshade was a fact - and the electric light only made his/her hours longer.

The generations of key punches, from human power through the modern 026 printing keypunch :-))

The generations of sorters, collators, accounting machines - all made obsolete by the 1401 system.

The glory and speed of the vacuum tubes as switches, replacing gears and levers.

The compactness and low power and long life of the transistor.


Fortunately, the docents taking our group of folks through the rest of the museum are charged with the above awesome challenge.

We can specialize in the era of the 1401, the transition from sorters, collators, accounting machines to the 1401 business computer.

So - how can we size our response to this more limited challenge. How many minutes and facts will our typical audience tolerate before eyes glaze over and they are lost. What items in our bag of tricks can keep "them" with us for how long. It the current moment I think we have a maximum of ? 10 minutes ? for the usual non-data processing specialist visitors. (I'm open to longer or shorter - the 10 minutes is just an initial guess.)

What should our ? 10 minute ? curriculum be? in what sequence?
At the end, many would like our visitors to leave with:

  1. an IBM card they punched, using an 026 key punch, with their name on it :-))
  2. seen their card (with many others) run through a 083 or other sorter
  3. - seeing the operator/docent read their card (with many other cards as in 083?) into the 1402 card reader
  4. - a sheet(s) of output from the 1403 printer with their name
    - say in large "patch separator" font
    - and a fake check with their name as the payee from some humorously fake institution?
    - a comparison of the 1960s 1401 and their likely 2005 home computer
    in speed, cost capability, maintenance, size, storage, power consumption, ...
    (is there information about the average cell phone? or something intermediate?)
    ?Names and e-mail addresses of active team members and URL of this web site?
  5. - have heard an explanation of how batch data processing worked?
  6. - heard a recording of "music" played on the 1403 chain printer
  7. - watched one or more tape units in action.
    - seen/handled our artifacts such as core memory, SMS cards, IBM cards, a snippet of mag tape
  8. viewed some IBM machinery through plastic skins.

I think the above is a lovely, possible dream - much easier than placing a man on the moon ;-))


or a backup video if not a scheduled time or operational (human or machine) problem.


Known Short Term Challenges - (our delivery of the demo in February)

There are a number of unrealized items dreamed of in the above curriculum. In the order above, they are:
  • The 026 key punch is normally operated by a relatively skilled operator, or a student with some familiarization lesson. The key punch is easily jammable. Each of our novice guest/operators should have a docent tutoring him/her in its use. I think that a person likely could spend up to five minutes getting their card ready to go (the first time). There for several key punches and docents are needed per group?

    Task? to be volunteered for? A quick key punch tutorial?

    - There is currently one key punch working at CHM, restored to functionality and printing with considerable pain and administrative struggling, over a period of 6 months.

    Robert Garner and Grant Saviers are working hard to obtain two more working printing 026s for the 1401 demo.

  • We need some ideas about the 083 sorter usage - should guest cards be shuffled into a deck with say a "1" in column 80, and separate out (most) guest cards in a pass? Using the 083 could be fun, especially with some fun ideas.

    Task? to be volunteered for? A sorter story board?

  • Show that the 1402 can select and pocket cards as can the sorter, under program control.

    Task? to be volunteered for? A 1402 reader/pocket story board and program?

  • Program to take guest data, and print it on the 1403
    one page with large block letters as in a batch separator
    another page with a fake check or bill

    Task? to be volunteered for? A 1401 programming task

  • A story board, as a hand out, how say how data a bill or pay check flows through the tab room and computer.

    Task? to be volunteered for?

  • An audio system with speakers to play existing "1403 printer music".

    Task? to be volunteered for?

  • A tape exercise/demo program with switch controls on which tapes to exercize - to provide for outages.

    Task? to be volunteered for?

  • Jeff has asked device leaders to determine what could/should be made visible to the guests.
    Where to use plastic skins, where to open doors or panels?

    Task - assigned to device leaders,

    Ron Williams (1401 processor), ?view end and core memory through plastic?
    Bill Flora (1402 card reader/punch), ?view card paths?
    Frank King (1403 printer), ?current windows are enough?
    Frank King (083 sorter), ?top to view card path?
    Allen Palmer (729 tape drive(s), ?sides to view clutches?
    Bob Erickson (077 collator), ?front and back gears?

We can probably use the non 1311 part of the "HzG demo program" program for the 1401 from HzG which Van Snyder mentions. Maybe this could be "played" while folks are struggling with the 026 key punches?

In any case - there is a large amount of work that can be performed in parallel by relatively isolated people, and funneled into the project for integration, hopefully in early January - because "Murphy" still lives :-)) and prospers :-(( Volunteers for any of the above? And what did I miss in my haste? :-|

Please e-mail responses to ed@ed-thelen.org.
Here is a 1401 development environment from Ron Mak.

David Cortesi's comments - November 28, 2007

I would like to make it explicit what I tried to convey by implication in the demo "script" I wrote: [Nov 20]

In my opinion, the HISTORICAL value of a 1401 demo is not that we have these really cool machines. The historical and teaching value is that we can use those machines to recreate the look, the sounds, the feel -- and the methods -- of a commercial computer center of fifty years ago!

The value of the demo cannot only be, "hey look at these cool machines." First, machines in general are not especially cool to many people. And the coolness of these old machines is only evident to nerds like us.

I think the real value of the demo for a general audience should be in showing three things:

  1. "Here is how real people lived and worked before (most of) you were born."
      a. what a data processing center looked and sounded like b. what kinds of jobs people spent their hours doing: data entry, sorting, mounting tapes
  2. "The whole idea of computer use was different then"
      a. Today, "computer" means "screen, keyboard, disk drive" Then, "computer" meant a lot of big boxes in a locked room -- not personal and no screens b. Today, "output" means "info on a screen" and "pictures" and "music" and "proportional fonts on paper" Then, "output" meant "invoices, bills, and checks on a printer" and "permanent records on tape"
  3. "Here's how they got work done"
    - without a keyboard, how does data get into the system?
    - without an OS, how do you load a program?
    - without a disk, how do you store and update a database?

The blinking lights, the twitching tapes, the sounds and look of cards zipping through, this is all sensory candy & wonderful, but by itself it has no real historical value.

There will be SOME people in any audience who would be really interested and intrigued by the technology -- the discrete transistor logic, the numbers on lines or cards per minute and so on, and a few who will be really interested in the restoration process -- but they are a minority who can be served AFTER the main demo for the general audience.

Dave Cortesi

Notes for operating BIGPRINT & Tape - from Robert Garner, May 8, 2008
Here's a summary of the power-on to demo steps from my notes, 
probably with errors/omissions (esp in 1403 alignment steps and 729...) 
Frank, Bill, Allen -->  feedback? 
(perhaps you could mark-up after Saturday, then we'll put on-line.) 

1.  System power on procedure: 
        a.  Turn on the wide, 1401-labeled circuit breaker in the AC panel in the corner near the entrance. 
        b.  Turn on the wide circuit breaker at the bottom of the Pacific Power 60->50 Hz Converter. 
        c.  Turn on the Pacific Power Converter by depressing its green "Power On" push button at the top. 
             The 1401 now has system power.  Note that the 3 phase voltages should each read near 220 V. 

2.  1401 CPU initialization: 
        a.  Set "Mode" rotary switch to "Run" position 
        b.  Set "Tape Unit Select" rotary switch to "N" position 
        c.   Momentarily depress "Start Reset" push button 

3.  1402 Card Unit initialization and card setup: 
        a.  Confirm reader gate is closed and momentarily depress "Run Out" push button on front panel 
        b.  Lower reader gate and place cards in face down, 9/bottom edge first/towards reader 
        c.  Place card weight over cards and close reader gate 
        d.  Momentarily depress "Check Reset" push button on front panel 

4.  1403 Printer initialization and setup for aligning cartridge to paper for BigPrint demo: 
        a.  Move clutch dial on right into "Aus" position to disengage clutch 
        b.  Open/unlatch train assembly and rotate paper roller handle so top of paper is line 33 (?) 
        c.  Close/latch swinging train assembly 
        d.  Momentarily depress "Carriage Restore" button 
        e.  Move clutch dial into "Ein" position to engage clutch 

5.  729 Tape Unit initialization: 
        a.  Verify supply tape has an enable write ring and mount on left 
        b.  Depress supply tape unlock push button located ... 
        c.  Feed tape over tape heads and then clockwise into take-up reel on right 
        d.  Hand rotate take-up reel until the reflective tape mark is past the read head 
        e.  Close door and close sliding glass front cover upward         
        f.   Momentarily depress "Reset," then "Rewind," then "Start" buttons. 

6.  Read cards in and start demo: 
        a.  Momentarily depress "Start" button on 1401 or 1403. 

- Robert 

	


Printout V3 for BIGPRINT by Robert Garner

Note: BIGPRINT uses a data deck using pairs of data cards of 65 characters per card, making each printed line 130 characters wide.
Here's V.3 of the 1401 print-out blurb, incorporating feedback from Dick, David, and Shel and a few changes of my own....

ANNOUNCED OCTOBER 5, 1959 BOTH WORLDWIDE AND IN THE U.S. OVER CLOSED-CIRCUIT TV, THE IBM 1401 DATA PROCESSING SYSTEM WAS ONE OF IBM'S MOST SUCCESSFUL PRODUCTS AND THE WORLD'S FIRST HIGH-VOLUME STORED-PROGRAM RANDOM-ACCESS COMPUTER. IN ITS FIRST 5 WEEKS ALONE, IBM BOOKED ORDERS FOR 5,000 SYSTEMS. A TOTAL OF 15,000 WERE PRODUCED WORLDWIDE BY THE CLOSE OF THE 1960'S, FAR EXCEEDING INITIAL EXPECTATIONS. BY 1962, RENTAL INCOME SURPASSED THAT FROM UNIT-RECORD ACCOUNTING MACHINES, DEMONSTRATING THAT GENERAL-PURPOSE COMPUTERS HAD ENTERED AN ERA OF HIGHER ACCEPTANCE. BY 1964, 40% OF ALL COMPUTERS IN THE WORLD WERE 1401-FAMILY MACHINES. ITS RUNAWAY SUCCESS INSPIRED TWO APPLICATION-COMPATIBLES: THE HONEYWELL H-200 COMPETITOR, MARKETED AS THE "LIBERATOR"; AND THE IBM 360 MODEL 30, IBM'S 1965 LOW-END 360 FAMILY MEMBER THAT COULD DIRECTLY EXECUTE 1401 APPLICATIONS. THE 1401 AND RELATED FAMILY SUCCESSORS, INCLUDING THE 1410, 1440 & 1460, WERE OFFERED BY IBM UNTIL 1971.

THE DESIGN GOALS OF THE 1401 BUSINESS SYSTEM, WITH THE SPUTNIK-INSPIRED CODE-NAME "SPACE" FOR "STORED-PROGRAM ACCOUNTING AND CALCULATING EQUIPMENT," WAS TO BE 3X FASTER YET CHEAPER THAN EXISTING PLUG-BOARD UNIT-RECORD ACCOUNTING AND CALCULATING MACHINES (E.G., IBM 402, 407 AND 604) AND TO OUTPERFORM THE FRENCH "MACHINES BULL GAMMA III" OUTSELLING IBM ACCOUNTING MACHINES IN EUROPE. ITS LEAD DESIGNER, MR. FRANCIS O. UNDERWOOD, INSISTED ON A STORED-PROGRAM ARCHITECTURE OVER THE TRADITIONAL PLUG-BOARD CONTROL AND INCORPORATED INNOVATIVE DEVELOPMENT EFFORTS IN IBM ENDICOTT, INCLUDING A NEW HAMMER AND CHAIN-BASED PRINTER AND "STANDARD MODULAR SYSTEM" (SMS) PACKAGING WITH COST-EFFECTIVE GERMANIUM TRANSISTOR CIRCUIT BOARDS. THE STORED-PROGRAM METHOD OF GETTING DATA INTO CORE MEMORY AND MANIPULATING IT THERE VIA SOFTWARE, ALONG WITH THE USE OF MAGNETIC TAPE OR RANDOM-ACCESS (RAMAC) DISK TO REPLACE BULKY CARD DECKS FOR MASTER AND SCRATCH DATA, MADE FILE UPDATING AND REPORT PRINTING MUCH FASTER AND CHEAPER THAN THE PREVAILING PUNCHED-CARD-BASED ACCOUNTING SYSTEMS.

EACH 1401 SYSTEM COULD BE CONFIGURED WITH 1,400 TO 4,000 MEMORY POSITIONS IN ITS CPU MAINFRAME WITH UP TO 12,000 MORE IN AN OPTIONAL EXTENDED-MEMORY UNIT FOR A MAXIMUM OF 16,000. THE MEMORY TECHNOLOGY WAS MAGNETIC CORE: FINE WIRES STRUNG THOUGH TINY FERRITE DONUTS INTO 2-D MESH ARRAYS THAT WERE STACKED. EACH DONUT WAS INDIVIDUALLY MAGNETIZED EITHER CLOCKWISE OR COUNTERCLOCKWISE TO REPRESENT 0 OR 1. HAND-MADE CORE MEMORY COST IBM AS MUCH AS $5 PER MEMORY POSITION -- SEVERAL HUNDRED MILLION TIMES THE COST OF AN EQUIVALENT MEMORY "BYTE" TODAY.

THE 1401 WAS DESIGNED TO EFFICIENTLY HANDLE BUSINESS DATA-PROCESSING APPLICATIONS AND TO REPLACE MOST PUNCHED-CARD UNIT-RECORD ACCOUNTING MACHINES. SINCE BUSINESS APPLICATIONS PROCESS CHARACTER STRINGS (NAMES, ADDRESSES) AND DECIMAL NUMBERS (SALARIES, HOURS), THE 1401 WAS OPTIMIZED AROUND THESE VARIABLE-LENGTH DATA TYPES. THE 1401 CPU ORCHESTRATED THE SERIAL TRANSFER OF CHARACTERS AND NUMBERS BETWEEN ITS CORE MEMORY AND PERIPHERALS SUCH AS CARD, TAPE OR DISK UNITS. APPLICATION PROGRAMS EMPLOYED SIMPLE INSTRUCTIONS THAT OPERATED ON VARIABLE-LENGTH CHARACTER STRINGS OR DECIMAL NUMBERS IN MEMORY. EACH MEMORY POSITION COMPRISES 6 DATA BITS (ALLOWING FOR 64 CHARACTERS), A PARITY BIT, AND A WORD-MARK BIT USED TO DETERMINE THE BEGINNING OR END OF AN INSTRUCTION, CHARACTER STRING, OR DECIMAL NUMBER. THERE ARE 34 INSTRUCTION TYPES.

CONTRIBUTING TO THE SUCCESS OF THE 1401 SYSTEM WAS ITS FAMILY OF HIGHLY RELIABLE AND CAPABLE PERIPHERALS. BOTH THEN AND AS RESTORED TODAY, THE 1402 CARD UNIT READS 800 CARDS/MINUTE AND PUNCHES 250 CARDS/MINUTE. THE 1403 CHAIN PRINTER PRINTS UP TO 600 LINES/MINUTE. AS MANY AS 6 HIGH-PERFORMANCE 729 REEL-TO-REEL TAPE DRIVES CAN BE ATTACHED. THE MODEL 1311 DISK STORAGE UNIT BECAME AVAILABLE IN 1962 WITH REMOVABLE DISK PACKS HOLDING UP TO 2 MILLION CHARACTERS. AS THERE WAS NO 1401 OPERATING SYSTEM(!), PROGRAMS ACCESS FIXED MEMORY LOCATIONS TO INTERFACE WITH THE PERIPHERALS AND WOULD ACCESS DISK DATA BY PHYSICAL HEAD & TRACK NUMBER.

WHILE THE 1401 WAS KNOWN AS THE MODEL-T OF COMPUTERS, IT WAS BY NO MEANS SIMPLE: A LARGE 1401 SYSTEM COMPRISED ABOUT 100,000 MECHANICAL PARTS AND ABOUT 50,000 ELECTRICAL COMPONENTS, INCLUDING 10,000 TRANSISTORS AND 14,000 DIODES ON ABOUT 3,000 SMS CARDS. EACH SYSTEM CAME WITH A ROLLING CART OF A DOZEN LARGE BINDERS OF LOGIC DIAGRAMS AND MAINTENANCE INSTRUCTIONS.

THE IBM BUSINESS MODEL, DEVELOPED AROUND THE LARGELY MECHANICAL UNIT-RECORD BUSINESS, WAS TO BUILD TO A HIGH STANDARD AND RENT, NOT SELL, THE HARDWARE. A TYPICAL 4,000 MEMORY-POSITION SYSTEM RENTED FOR ABOUT $7,000 PER MONTH, WHICH IS ABOUT $50,000/MONTH IN TODAY'S DOLLARS, ADJUSTED FOR INFLATION. THE RENTAL FEE INCLUDED MAINTENANCE FROM IBM CUSTOMER ENGINEERS ON CALL 24/7. THE PURCHASE PRICE OF OUR 1401 SYSTEM IN TODAY'S DOLLARS WOULD HAVE BEEN ABOUT 2.5 MILLION! IF WE, PERHAPS UNFAIRLY, COMPARE THE COMPUTATIONAL POWER OF A MODERN PC TO OUR 1401, TODAY'S PC IS ABOUT 10 BILLION TIMES FASTER PER DOLLAR!

COMPUTER HISTORY MUSEUM VOLUNTEERS, INCLUDING ABOUT A DOZEN RETIRED IBM ENGINEERS, HAVE WORKED FOR OVER 15,000 HOURS AND 3 YEARS TO FULLY RESTORE THIS 1401 SYSTEM (FROM GERMANY) AND ARE NOW BEGINNING TO RESTORE A SECOND SYSTEM (FROM CONNECTICUT). WE'RE LOOKING FOR NEW VOLUNTEERS.


E-mails with suggestions

 

Tore Sinding Bekkedal - 
Though I due to some geographical concerns can't attend the meetings,
here I humbly submit a possibly obvious suggestion of mine - how about a
program that would let you punch a card with your name, and have it
print it out on the 1403 in 'figlets' - large, ASCII-art letters?
Possibly for a small fee? Possibly double-printed so it's readable from
a distance, for posters or something? Possibly with an ascii-art CHM
logo at the bottom, and/or with the names of the team members present,
possibly with a pen signature? Maybe even batch them up on a magtape
first, so that the audience can see how this worked? Maybe even generate
the figs before writing it to tape, and then dumping the tape to
printer. This would allow for use of the peripherals, interactivity,
and, very importantly, spinning tapes. :)


Peter Bassett
I used to give 1401 demos every year (starting in 1961) to high students 
who came to our company during "Business Week" for hands-on observations 
of typical office work. One of the favorites went something like this ....
 

Put the CPU in a never ending loop so that every light possible blinks and blinks and blinks. They especially liked the register lights because of their size. I would then ask someone to turn on sense switch B which caused a BSS B,loop2. Loop2 fired up another closed loop moving the access arms of the three 1311 disk drives in what appeared to be a rhythm pattern. Then someone would turn on sense switch C so that a BSS C, loop3 could be executed. Loop3 executed a closed loop routine spinning four tape drives. Sometimes in unison, and sometimes one after the other in varying sequences. The tape finally would be a high-speed rewind.

Sense Switch D would cause the card punch to punch out cards into multiple pockets which were later interpreted along the top edge. The final Sense Switch caused a printing demo on the 1403. The twist was that I had internally stored information on each of the students (obtained from the guidance counselor). The information would print out for each person with a page skip between each person. One excited girl said "quick - quick ... ask it when I'm getting married!" The class roared in laughter. The sense switch technique gives you ample and variable time to explain what's going on at the moment and to answer questions. Van Snyder I have the HzG demo program for the 1401 from HzG. It reads a deck and stores it in core, then repeatedly prints it (if SS C is on) while spinning the tape drives 3-6 (if SS D-G respectively are on) and jiggling the 1311 disk head (if SS B is on). There's a parameter card that controls the timing. To avoid trying to jiggle the head on the 1311 disk drive that the CHM 1401 doesn't have, don't turn on SS B. It would also be easy to patch the code to NOP the SS B test to jiggle the disk head (change column 16 of card 11 of the HzG deck, or column 25 of card 10 of my deck, from B to N). The deck from HzG prints a "Madonna and Child" picture followed by a greeting from HzG, but that could easily be changed, for example, to print a CHM logo. Whatever is printed, it would be easy to incorporate visitor's cards after (or before) the "standard" image deck. I've reverse engineered the HzG deck, and re-written it. The original and its reverse-engineered source are attached as exercise.hzg.zip. My re-writing of it is included as exercise.new.zip. Just for grins, I modified the exercise.new code not to use the disk. It's attached as exercise_nodisk.zip. Ron Mak I believe there are two types of demos: (1) programs that make the machine do something interesting for a visitor to watch and hear, and (2) programs that do something personalized for a visitor to take home. (1) "Interesting" demos include: Make the hardware perform (read/punch cards, blink console lights, spin tapes, print, etc.) Calculate and print something impressive, such as Ed Thelen's 5000 digits of pi (2) "Personalized" demos include: - Visitor submits a punched card with name and birthdate, gets a printout of that information plus a biorhythm chart - Visitor submits a punched card with name, get a printout of name in large block letters Speaking of Ed's pi program, on my 2.39 GHz Pentium 4 workstation, the simh 1401 simulator takes about 6 seconds to run it. Would anyone care to venture a guess as to how long this program would run on the actual hardware (cycle time 11.5 usec)? I've attached the program I ran on simh; Ed may have a more recent version.


return to main page