Not Quite My Era

The First Digit Computer

Here is an interesting look back at the first digital computers we used for computation. It is a lovely BBC video about a restored system from days of old.

The 'Witch'
The ‘Witch’ – the first digital computer runs again after rebuild


Programming was limited as old systems had 90 bytes of memory, not 900, not 9000 not 90,000, not 900,000 or 9 meg, etc. – just 90 bytes !! A real challenge for the programmers of that era.

Burroughs Advertising, circa 1967
Burroughs Advertising Video, circa 1967


Pay particular attention to the punched paper tape descriptions in that BBC video. This was the “I/O” of the day. Fortunately for me, my then employer Burroughs Federal Systems Division of McLean, Virginia, (mid-1970’s) had the latest kit from our factory in Paoli, Pennsylvannia. Our division of Burroughs sold kit to the US government federal departments, like US Army, US Navy, US Air Force, Dept of The Treasury, Dept. of Labor, Architect of The US Capitol and the government of the District of Columbia. We always had the latest kit as a foundation for our demos to many US agencies and departments.

Control Panel for Burroughs B-2500/B-3500 Computer
Control Panel for Burroughs B-2500/B-3500 Computer

Kit and Coding

We had a Burroughs B-3500 with a cobol compiler and it could generate machine code for the striped-ledger accounting machines that Burroughs sold to many banks. The L-5000, L-6500 and L-8000 systems were programmed via punched-paper tape. To make it easier to program the L-series, our B-3500 system had a version of cobol known as L-Cobol, presumably for ‘Little’-Cobol and we also had a paper-tape punch machine to output machine code from the L-Cobol compiler. Paper tape was fragile, so for more robust storage we might use paper-tape composed of a more fiberous version known as Mylar. The photo right shows a sample. Note there are 5 channels or tracks of holes, may be wrong on this but the encoding was the baudot format rather than later ascii format used when 7 channel tape became available.5 Channel Punch-Paper Tape


The work flow was : 1) write cobol statements on coding sheets of paper, 2) use 80 column card punch to turn code from your coding sheets into a deck of several thousand 80 colum cards, 3) go to B-3500 4) place deck into 80 column card reader, 4) go to B-3500 control control – the ‘SPO’ – for system print-out, 5) put in a command to the MCP – the Master Control Program, 6) wait for the L-Cobol compiler to run thus producing a compiler listing on our 132 column printers, and if no errors, the generated machine code was punched out on the paper-tape punch, 7) carefully transport paper-tape to your target L-series system, 8) power-on and reboot that system, 9) invoke load command so  L-system would read it’s ‘new’ program from the paper-tape, 10) test, 11) a bug ? Drat !!, do steps 1 thru 11 until success

While my earliest experiences were with ibm 1401 and ibm 1620 computers, these products had moved on to the point where we could use 80 column punched cards rather than plug-boards as a programming tool. We still needed to program the IBM 704 punch-card systems with telephone-exchange-style plug-boards buts that another story for another day. 🙂

One thought on “Not Quite My Era

  1. Just loved your Burroughs nostalgia entries! I was an MCP specialist from 1972 to 1977 for the Chicago district. Still looking for a copy of a medium system MCP listing. Greatest O/S ever (at least for its time.)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s