#computerarchaeology

This old computer is comprised of four big boxes, three of which are ever actually used.

The UNIVAC 1219. This is the brains of the system. It controls the operations of every other device. This is what I'm referring to when I'm not gesturing to the UNIVAC 1219 as a whole.

The UNIVAC 1540. This is the DDR, or Digital Data Recorder. It holds, writes, and reads the magnetic tape operators load into the machine.

The Digital to Analog Converter. The UNIVAC 1219 was the first digital computer on most U.S. Navy ships, most of which had analog weapons systems. This hulking mass of steel translated the digital signals from the computer to the analog signals of the weapon systems and vice versa in regards to the radar.

The UNIVAC 1532. The I/O console managed the...you guessed it, input and output of the UNIVAC 1219. You can load and punch paper tape for programs more bite-sized than would be used for magnetic tape.

In addition, we have two teletype machines. You can think of them like typewriters that don't receive human input (except the one that can if we want), but instead output what the computer tells it to.  We have a Teletype Corporation teletype that is optimized for character compatability and a Kleinschmidt teletype that is optimized for speed. Both rely on the I/O console to send and receive data.

The real ingenuity begins with the floppy drive. Duane, who's career revolved around this system, developed a way for a floppy drive to imitate the I/O console. The computer thinks it is reading and writing to a paper tape, when it is in fact reading and writing to a 5.25in floppy inside an ancient CNC machine floppy drive.

And this, dear reader, is where the magic happens. This framework was originally built for interfacing with the 1219 via BIN files over Serial port and was easily changed to support BIN files over floppy. Duane has been working on an off adapting our purple converter box with a raspi to let the 1219 read and write BIN files over Bluetooth.

Make no mistake, you cannot simply SSH into this machine as tons of setup and channel changes must be performed to ready it to receive and send data. That being said, I don't see any other UNIVAC mainframes with Bluetooth [or any other running UNIVAC 1219s at all :(], so I will take what I can get.

Also, I was wondering if you have any fun stories surrounding it! Strange quirks it has or anything like that.

I'd love to see more photos if you're allowed to post them!

Thanks for the question! These are my favorite part about my blog by far.

Not exactly, the UNIVAC 1219 doesn’t have a nickname. I did realize recently that I should specify the pronunciation (Twelve-Nineteen), but it doesn’t have any nicknames. Apart from ‘the 1219’, it’s also regularly referred to as the CPU or just ‘the computer’.

Fun stories or weird quirks? Boy, I could fill a book with this machine’s weird quirks (or as we say, intermittent issues), but I’ll try to blitz through the most common ones:

Sometimes the computer will stop running and enter a WAIT mode. No reason, it just needs a break. We can’t fix it, it just has to decide to go back into operating mode.

The computer will often start attempting to communicate on IO channel 13. We’re not telling it to talk to anything, it just decides to try to.

One of our teletypes (the Kleinshmidt) stamps ink splotches into the paper rather than characters most of the time. However, this weekend it worked for the first time in 10 months! We didn’t change anything, it just had an extra cup of coffee or something.

The Digital Data Recorder, or the tape drive, has the most gremlins out of any of our units. The top handler works fairly well, but the bottom handler won’t properly read data, write data, move the tape forward, initialize the tape, or any number of other issues.

There’s more but hopefully this satisfies your curiosity.

Fun stories? Well, I can’t name any specific ones, but I can say it’s a very endearing machine. It’s the very last of its kind and being one of three individuals in the world responsible for it makes every issue that more frustrating. There is no real forum for it, the subject matter experts sit next to me and are often just as exasperated as I am.

But the unique nature of this situation make every successful diagnostic test that much sweeter. Every new addition (5.25” floppy drive via serial) that much cooler. I have an IBM PC-XT clone at home, but I thank my lucky stars every day that this big iron is what I get to specialize in.

As for more photos, I have none that are as grandiose as you would probably expect. I do have my working photos though. I took all my photos when I first started working on it and now I am more dedicated to fixes than photo-ops.

This is a photo of our finicky Kleinshmidt teletype. Still has blotches but it actually printed!

This is the back of the bottom handler. Pictured is the vacuum pump in the bottom left (so sudden stops just yank magnetic tape slack rather than ripping tape). The big cylinder in the center is a motor for running the magnetic tape handler itself. The big black ‘hose’ of wires coming out of the steel plate contains all the cables that come right off the handler’s head for reading and writing data!

This is the forward pinch roller of the bottom handler. It was replaced after this photo was taken as you can see the rubber has deteriorated in the 55 years this machine has been operating.

As for being allowed to post photos, that’s not an issue. The last 1219 was decommissioned in 2014 and now you can find all of its documentation online at http://www.bitsavers.org/pdf/univac/military/1219/

The radar(AN/SPG-55B) that this very UNIVAC was connected to at John Hopkins not only survives and was once on public display, but my mentor actually PAINTED back in 2015!!!

Somehow, I have never heard about this until now!

But as I was typing this, I found out that the Museum it belonged to closed. Rest assured, I will try to find it but if anyone has friends previously involved with the National Electronics Museum I would very much so like to get in contact.

Oh my goodness, my first question! Well, in a sentence, the UNIVAC 1219 was used to point and fire missiles at enemy aircraft. Now, we can break this down:

The UNIVAC 1219 is a militarized version of the UNIVAC 418. This thing has better resistance to salty air and other things you’d want a seafaring computer going into battle to have, but at their hearts they’re the same. There were three different missile systems the UNIVAC was a part of: Talos, Terrier, and Tartar. These three systems were among the first sea-to-air missile systems fielded by the U.S. Navy.

The UNIVAC 1219 would receive data from the AN/SPG-55 radar and analyze the data it received. Running one of dozens of programs lost to time (or on a single last-of-its-kind magnetic tape we are worried about damaging), it will pass on the results of its calculations to the missile system itself. The missile will go where it is told, and hit an enemy plane.

But our UNIVAC 1219 is unique! The specific machines here are from the John Hopkins Applied Physics Laboratory. At John Hopkins, they developed the fire control programs ran on Navy ships! They had a radar and everything, and I only wish we had a radar and missile battery too ;D.

As for what we do for demonstrations, we have assortments of cool programs we run, and I encourage you to swing by and see for yourself. Our oldest party trick is printing ASCII art of Spock and George Washington, but maybe if this post performs well I will divulge some other cool things we do ;).

With the UNIVAC 1219 being 62 years old now, you would be forgiven for thinking we spend our days figuring our how to get old programs to run and what they did.

Oh no. We make our own programs. We make A LOT of our own programs!

Duane, the person whose career revolved around the 1219 pumps out a new or updated program about every two months for us to try on the UNIVAC 1219.

See, while our tapes for diagnosing issues may date back to when the machine still served, we often find ourselves dissatisfied with them. We are realistically limited to about 100,000 addresses in the machine. The first 537 are reserved, then there are random gaps up to 70k where our utility programs end. Sometimes though, old programs don't play nice with OTHER old programs or programs that Duane has written in the past few years. Bill relays this to Duane and Duane gets to work rewriting these conflicting programs.

Duane has developed an entire 1219 EMULATOR to make sure that these things work properly. I can't give you the modern version, but a 2013 version can be found here under 1219files.zip. Duane also owns a tape punch and punches his own paper tape and mails it to Bill. Bill and I then set off to the museum and make sure it works as expected.

Thanks for the question!

I have a bit of a silly answer. Punch letters is my favorite program to run! After loading it into memory and running it, the IO console’s paper punch spools up and you get a prompt on the IOC Teletype asking you to enter a string. Once you submit it, the punch will print out your string in ASCII text on a piece of Mylar-core paper tape! It’s simply and not too flashy, but easily my favorite.

That being said, I have some other favorites. Our grand Spock and George Washington Art have become a staple Vintage Computer Festival here at InfoAge, so of course they hold a special place in the history of this machine. Recently, now that our tape drive is more functional, our flashy tape drive diagnostic program has become popular too. It performs rapid reads and writes, forwards and backwards, and puts on a great show. You can see this in action in Fran Blanche’s video from VCF East this year at

Other programs of note are Matrix letters, which is punch letters but using the Teletype paper rather than paper tape.

Maybe some more cop-out answers would be the paper tape load program and PGM249A. The paper tape program allows us to load any data from mag tape or paper tape, so of course it’s a valued and indispensable part of this machine (and something I have to manually hand key once every working day it seems) while PGM249A is our diagnostics program. Most issues we have can be found with this program that analyzes all facets of the machine from IO channels to memory to timing, but as we have solved more commonplace issues I feel as though we luckily need to use it less and less.

We have a few program lists scattered throughout our manuals. I’ll see what I can dig up!

Welcome back, please accept +1 frequent flier milers as thanks!

To answer your question...well, not really. Intercontinental Ballistic Missiles, or ICBMs, saw deployment beginning in 1959. This would become the de-facto way to launch nuclear missiles, and the Soviet Union never had a similar program to the United States with round-the-clock patrols of nuclear-ready bombers (Operation Chrome Dome).

In theory, with the Soviet TU-95 Bear's (and TU-160's in the 70s) cruising altitude being 56,000ft and 45,000ft respectively, that puts them in range of all three missile systems (Terrier, Talos, and Tartar) that the UNIVAC 1219 operated. That being said, I would be skeptical that these systems could score a kill on a bomber. They only ever killed four unidentified MiGs and a MiG-17 (early fighter with a flight ceiling of 54,500ft and 95mph faster cruising speed than a TU-95).

So...I guess it is possible, but extremely unlikely. The AN/SPG-55 radar the UNIVAC 1219 was hooked into had 150nmi range and wouldn't have an issue finding a bomber if they knew where to look, but its narrow field-of-view (as an 'illumination' radar, they're called) would mean that the AN/SPS-48 with a range of 100,000ft would have to find it first. Using Pythagorean theorem and the TU-95's cruising altitude, we can determine that the plane would have to be within 15 miles of the ship. In the Pacific Ocean, that's nothing.