The incredible story of the first PC, from 1965

P101 front

Almost 50 years ago, a small team at the Italian company Olivetti managed to do what no one had done before them; they created a computer small enough to fit on a desk, and could be used by regular people. It was the Programma 101, what many consider to be the world’s first personal computer.

To understand just how revolutionary the Programma 101 was when it was unveiled back in 1965, you first have to know what computers looked like at the time. Remember, this was almost 50 years ago. It was the era of huge mainframes, big as fridges, sometimes filling up entire rooms. Only a small elite had access to them.

People were used to seeing this:

IBM mainframes

Above left you have the IBM System/360 Model 30, introduced in 1964, and to the right you have the IBM 1440 Data Processing System from 1962. These massive installations were what people associated with the word “computer”.

So imagine the impact when Olivetti presented this:

Olivetti Programma 101

So how did that incredible paradigm shift happen?

1962, the idea of a personal computer is born

In the early 1960s, computers were not something you could have all to yourself, and most people didn’t even have access to them.

At the time, Olivetti was trying to compete with the American companies, making large mainframe computers. Being from Europe, they were a bit of an outsider, but they had some limited success.

But Roberto Olivetti, then-CEO of the company, wanted to try something truly revolutionary. He wanted Olivetti to make a small computer that was more affordable and could be used by regular people, something that you could place on a regular desk. He wanted to create a personal desktop computer.

He gave the project to Pier Giorgio Perotto, an engineer at Olivetti, who would work with a small team of just four people to try and get around all the technical hurdles and create this revolutionary device.

The P101 team
Above: Pier Giorgio Perotto (bottom left) and his small team (minus one).

The idea was insanely ambitious. Perotto’s team didn’t even know if it would be possible. The technology used in computers back then was way too bulky, so the team had to invent something new for almost every element of the new device, which was meant to be the size of a typewriter. The human-centric approach to designing a computer was also something completely new, and brought its own set of challenges as well.

The project started in 1962.

The challenge of making a PC in the 1960s

The first thing the team had to shrink was the memory. Memory modules in the early 1960s were huge. If Olivetti had used existing technology, the memory alone would have been as large as the entire computer they had envisioned.

So they had to create a memory module from scratch that was small enough for their purposes. Not an easy thing to do, but they came up with a kind of magnetostrictive delay line memory that was just a fraction of the size of other memory modules. It was roughly the size of a small motherboard in one of today’s PCs.

Next came storage. How would they store the programs? It had to be on something small and practical. The team came up with an ingenious solution, a card with two magnetic strips that could be inserted into the machine. Each card could hold one, sometimes two, programs that you could then start by pressing a button.

Programmable magnetic card

They had invented the programmable magnetic card. Thanks to this, anyone could enter a pre-made program into the device and run it in a few seconds. Its legacy lasted decades, including later evolutions of the idea such as the magnetic floppy disk.

Another challenge was ease of use. The way to program the computer had to be easy enough for a regular person, not a computer scientist, to learn and use.

The approach Perotto used was to create a simple programming language, something akin to a simplified assembler consisting of just a small set of instructions. The Programma 101 would later be shipped with a thin book that taught new users how to program the machine.

The actual physical design of the device was also a challenge. It was meant to be an elegant, human-centric and ergonomic device. Olivetti was famed for its attention to design (they made, for example, very beautiful typewriters), and the Programma 101 wouldn’t be an exception. The team involved Mario Bellini to style the device. It was a triumph of industrial design, very innovative for its time. Bellini would go on to become a world-renowned architecture and designer, so they really picked the right person for the job.

What they ended up with was a computer in a chassis that wasn’t much larger than a regular typewriter. It was easy to program, could store and run programs from a magnetic card and could be placed on a regular office desk. It weighed 35 kilograms, a featherweight at that point in time.

The Olivetti Programma 101

The Programma 101 was ready in April of 1964. The first personal computer had been built.

But there was a cloud at the horizon.

It almost didn’t happen

You’d think that the Programma 101 would have been a shoo-in, something Olivetti would have been eager to get behind. Not only that, it had the support of the CEO, Roberto Olivetti, who had spawned the initial idea.

However, in the spring of 1964 the company was in deep financial trouble. A crisis group was set in to restructure Olivetti, and part of their solution was to sell off the company’s electronics division. What did Europeans know about computers anyway? This was a disaster for the Programma 101 team, who were about to be thrown out of the company they loved.

Ultimately, General Electric rushed in and bought Olivetti’s electronics division early in 1965. Perotto’s team didn’t want their hard work to end up in American hands and maybe be cancelled forever, so they had a little trick up their collective sleave.

To avoid being swallowed up by GE, they changed the categorization of the project from “computer” to “calculator.” A tiny change, but it meant they were left under Olivetti’s umbrella. GE got the entire electronics division, minus the Programma 101 team.

Nevertheless, it was an awkward situation. Perotto’s team continued to work at the Italian facility, a building where GE now owned everything except their office. The team even painted the office windows so GE staff wouldn’t be able to see what they were working on.

Sadly, by now Roberto Olivetti had left his position as CEO, and the new management knew little about computers and clearly didn’t see the potential of this new invention. They were convinced there was no market for it.

If it weren’t for the New York World Fair in 1964-65, the Programma 101 would probably never have seen the light of day.

The 1964 New York World Fair

1964 NY world fair

Like many other companies, Olivetti wanted to show off their newest products to the masses at the World Fair. The company was there to demonstrate their new mechanical calculator, the Logos 27. That was their focus.

The Programma 101, on the other hand, was hidden away in a small back room. The new management at Olivetti still didn’t believe in the project, and had it there as a prototype, an interesting oddity.

When the Programma 101 ultimately was unveiled to the masses at the world fair in October of 1965, it was the first time it was viewed by the general public. In an era when people largely regarded computers with suspicion, it had an impact few could have anticipated.

At the unveiling, the presenter from Olivetti explained to the audience that he would now calculate the orbit of a satellite, entered the card with the program, and ran it. It took a couple of seconds, then the computer started printing out the result. To us today this is hardly impressive, but 50 years ago it was en eye-opener.

The reception was overwhelming, the word spread, and soon it became clear that Olivetti had a huge hit on its hands. The Programma 101 was suddenly the front and center of attention at Olivetti’s booth. People were amazed at how something so small could be a fully working computer. Some even suspected it had cables connected to a larger computer hidden somewhere behind the scenes.

And the press loved it. You had articles in the Wall Street Journal and Business Week with titles like “’Desk-top’ Computer is Typewriter Size.”

p101 press clippings

It was too much for the reticent management at Olivetti to ignore. Mass production started soon after, and the Programma 101 went on sale just a couple of months after it had been unveiled at the world fair.

Commercial success

p101-adThe price for a Programma 101 was around $3,200. Adjusted for inflation that is way north of $20,000 today.

That may sound like a lot, but if you compare it with what a mainframe computer cost in the 1960s, it was a bargain. Even just renting access to a mainframe for a month would set you back as much as it cost to own a Programma 101. And actually buying a mainframe, even the cheapest available option, would cost at least $100,000.

Olivetti had created a new market, and was richly rewarded for it, selling around 40,000 units.

Since it was relatively affordable and portable, the Programma 101 gained widespread use. Even though at that price it could hardly be called a computer for everyone, it was still an enormous step in the right direction.

A fun aside is that NASA bought at least 10 Programma 101s and used them for the calculations for the 1969 Apollo 11 moon landing.

Programma 101 UK ad

It wasn’t until HP launched its HP9100 series in 1968 that the Programma 101 got some real competition. Even then, it was a device heavily inspired by the Programma 101, and HP was ordered to pay Olivetti $900,000 in royalties for using such a similar architecture, and the famous magnetic card.

A few tech specs

By today’s standards, it’s not exactly a powerhouse, but remember this was 50 years ago, and the engineers were pushing the limits of what could be fit into such a small device.

  • Weight: 35.5 kg (78 lbs)
  • Power: 350 W
  • Display: None. It used a small printer and a roll of paper 9 cm wide.
  • Memory: Approximately 240 bytes.
  • Storage: Magnetic cards. Programs could be split over multiple cards if they didn’t fit on one.
  • CPU: None. This was before microprocessors, and integrated circuits had not really arrived on the scene yet. The logic was all made up of transistors, diodes, resistors and capacitors.

An amazing achievement

P101 advertisement

When you think about it, it’s simply amazing. A small team of just 4-5 people completely rethought the computing experience and ushered in the era of the personal computer.

We’ll leave you with a few snippets from the excellent documentary, Programma 101 – Memory of the Future:

To the Programma 101 team: We salute you!

Data sources: Primarily Wikipedia and the excellent documentary, “Programma 101, Memory of the Future.” There is also a bunch of interesting information at Francesco Bonomi’s Programma 101 site, if you want to really dive into the nitty-gritty of the device. Another great source was the site, dedicated to the memory and achievements of this great engineer.

Image credits: Images courtesy of Francesco Bonomi (magnetic card and the sitting woman ad from his Programma 101 site); press clippings from the Pier Giorgio Perotto site (thanks to his son, Pierpaolo Perotto), (Storia di un’impresa) for the device pics, ads and the photo of the P101 team; World Fair photo from PLCjr (via Wikipedia); IBM Archives for the mainframe pics; the UK P101 ad via


  1. A large part of a generation’s attitude that they are the “first” to get things right, or do cool stuff is ignorance. Those who do not know history repeat it unknowingly. There have been several descriptions of iphone-like phone-calculators going all the way back into the 1960s (the novel “Wired Love” from that era).

  2. My first machine!
    I wrote some engineering programs that pushed the upper limits of one program that could run without resorting to multiple cards, splitting registers, etc.
    I heard they came out with one with an A-D convertor option.

  3. We have an Olivetti Programma 101 in pristine “mint” condition, with all manuals and 30 magnetic cards. Anybody have an idea of it’s current value? We’re interested in selling it. We’ve had this machine displayed in our private collection for about 30 years.

  4. As Italian I am very happy and proud of this article. Roberto Olivetti was as visionary as Steve Jobs, and his engineers were no less visionary than him because they imagined a device that never existed before and – most of all – realised it. Nowadays that’s one of the few things our country can be proud of.

  5. I refuse to give Jobs ANY credit.  While the 101 was made in 1965, I was to go on to be a major player with the DEC PDP-8.  DEC’s work was visionary, clearly.  Just like the 101 absolutely was.  The 101 allowed you to use in an engineering firm a useful amount of work with some effort.  Te PDP-8 gave you a more serious amount of capability that fit into a phone-booth in a lab, and that was set forth earlier by Seymour Cray in 1960 with the CDC 160 and in 1962 by the MIT LINC.  The beginning of the PDP-8 was the1963 PDP-5 which is not quite compatible [and it shows] and the PDP-8 itself also came out in 1965.  These represented milestones towards the notion of a PC on your desk as a practicality as steps away from the “big iron” in raised-floor rooms and batch mentaility and punch-cards and computer “operators” and their bosses who wielded a form of “political” power that “somehow” meant that THEIR jobs [and the jobs of their cronies] got run first.  For the rest of us, come back next Thursday, maybe we’ll have some idea of when [or even if] your job got run, or did your deck of cards just get thrown away “by accident” [or did it actually run and the OUTPUT LISTING was also thrown away!]?
    The 101 and the PDP-8 both helped to do away with that bureaucratic nonsense, and indeed by the late 1970’s, IBM’s big-iron monopoly was broken, largely but the stable of DEC machines from the PDP-8 through the VAX [as well as other competitors; IBM was less than 1/10 of what it was in 1965].  We all “won” because of these true innovations, not anything Apple.
    Anything Jobs did was cheap crap, and the only thing he was “visionary” was to make stuff so poorly it was a false economy but a low net price.  You can fool newbies, but to us “old-timers” please don’t insult us.  Wanna here the “innovation from Jobs back then:  Here is an interesting and true fact:
    1)  He sold a machine with a serial port on it implemented in software at great overhead.  The UART chip was one of the most inexpensive items in all of computer-dom.  DEC as recently as 1972 was selling a UART done “the hard way” but totally funcitonal, using a large bunch of SSI chips to achieve total compatibility with what would become commonplace, including DEC a year or so later. 
    At the time, the few actual UARTs available were expensive, slow and unreliable.  This is why the 1970 DEC VT05 terminal could only go up to 2400 baud; the GI UARTs of the day were not up to the job, and even several years later, the standard part only wend to 9600 baud; hand-picked parts dubbed “special” UARTs and a jumper change got you to 19200.  But these speeds were totally real and the software did conventional serial logic.  The TTL version has been confirmed to work at speeds of at least 76800 baud using non-standard jumpers; when the GI UART was practical enough, they then sold the actual UART replacement part.  [The earlier one is known as the KL8E and the later one the KL8-JA.]
    By the time the [cr]Apple machines came out, it was unconscionably to do anything but an inexpensive UART chip, but Uncle Stevie the “visionary” had a better way:  He instead used a one-bit “bit-banger” interface so that high-overhead software followed the bit changes in real-time thus making the entire machine into the equivalent of a cheap chip.
    Anyone remember there was a company that made money selling replacement plastic covers for Apple machines?  This is because they were so bad they literally could burn up!   I guess planned obsolescence by your machine melting could be considered “visionary”.

    While we in the real-world were writing complex real-time software with such things as full-dupkex, special terminal handling characteristics for all sorts of affordable terminals [not all from DEC] Mr. Visionary made his already underpowered toy effectively even slower by high-overhead software just as a dedicated part of any program that actually had to deal with a modem.  This is “visionary”?  No way, it is merely extremely false economy.  And the sofrware was all primitive as a result, but few noticed it because there were either a) haves with real computers, or b) users of toys with delusions of grandeur.  [Some may not remember the lunatic articles discussing the likes of “time-sharing 8080 chips among several users” which is beyond preposterous other than for some demo of multiple terminals outputting simple text strings.  No time left to actually PROCESS anything TO output, etc.
    Sorry, if that’s visionary, I got a bridge in Brooklyn to sell you.
    [Oh, and I forgot using the CPU to simulate the chip that is used in most keyboards as well at the same time!]
    The 101 was gutsy.  It tried to do something no one ever did before.  And my discipline as a programmer was born on the 101 and prepared me to work on a computer that was “only” 4096 words and develop highly effective software for that class of machine, etc.  Jobs’ crap was major steps backwards unless your view of computing was not grounded on then-current technology.
    We can excuse the 101 for doing what it did at the time, and we can excuse the PDP-8 for doing what more it did at a whole lot higher price.  But the engineering spirit in both shows many parallels.
    Doing something already done before, but doing it on the cheap is NOT visionary at all.  In fact, it borders on fraud.

    cjl [I was there at the 101 beginning; I know people who were “there” years earlier]

  6. I agree that the Olivetti Programma 101 was a game changer from 1965 and an amazing machine. In fact, I have rare unit that I’m willing to sell that’s in excellent condition. It’s been in my personal collection for 30 years and I’m slowly starting to liquidate. I sold a rare HP 9100A last year, the machine that copied the Programma 101 and caused HP to pay Olivetti for patent violations.

  7. I am trying to bring back to life my Progamma 101 and I am looking for manuals dealing with the electronic section (if existing).
    for Talisp1, in your manuals is there such a manual for electronic section?

  8. I am trying to bring back to life my Progamma 101 and I am looking for manuals dealing with the electronic section (if existing).
    for Talisp1, in your manuals is there such a manual for electronic section?

  9. patbin   Hi Patrick,

    We have all of the programming and users manuals, but nothing for troubleshooting the electronics. Sorry. We want to sell this unit but I don’t think that will help you revive yours.


  10. Hi Tanya,
    Thanks for you prompt answer.
    I think you will sell your unit with no problems. Sometimes there are units on Ebay, I think price is between 2000 and 3000$.
    The unit I have has a sentimental value because it is the one I used at technical school in 1969 just when it arrived. I found some infos and diagrams for electronic rack but it is in italian, I am trying to translate in french!
    I have the technical manual for mechanical parts in french and in english, and the part manual in french.
    I have also a second electronic chassis comming from a machine I used at work a long time ago. Unfortunatly this machine was destroyed and I only saved the chassis with boards and memory.
    May be you can scan the manuals and sell copies, to my knowledge, only the general programming manual is free on the Net.

  11. patbin  I haven’t seen any on ebay and have been looking for about 1 year because we don’t know what it’s worth. I think it’s worth more than you do though because we sold an HP 9100 for $8,000 last year and the Programma 101 seems to be more rare. I do understand the “sentimental value” aspect. My partner found this machine back in the early 80s when he was working at HP, about 30 years ago, and has been holding it in his collection. He loves it and raves about how beautiful it is! What an engineering marvel! Good luck with getting your unit restored!


  12. talsip1 patbin  

    Hi Tanya

    Last year 2 machines passed on Ebay, but Ebay Europe, for one I bid 1600€ and gone 1601 to a german guy!
    May be in the US the value is higher and you can try to fix your price at the level you vish.
    Last year I bought a P203 (electronic same as P101) for 200€ in France (but I had 250€ high way toll and gas to get it, 650 km from home and more than 100 kg!)
    the HP 9100 is a beautiful machine, don’t think I will have one but hope!
    I bought a lot of 12 HP9825 and one HP85 for almost nothing and will try to restaure also after the P101.

  13. talsip1 patbin  

    Hi Tanya

    Last year 2 machines passed on Ebay, but Ebay Europe, for one I bid 1600€ and gone 1601 to a german guy!
    May be in the US the value is higher and you can try to fix your price at the level you vish.
    Last year I bought a P203 (electronic same as P101) for 200€ in France (but I had 250€ high way toll and gas to get it, 650 km from home and more than 100 kg!)
    the HP 9100 is a beautiful machine, don’t think I will have one but hope!
    I bought a lot of 12 HP9825 and one HP85 for almost nothing and will try to restaure also after the P101.

  14. Wilst the 101 was amazing the 203 was stunning . Why this development of the 101 seems to get little mention is a mystery to me? It had a larger number of instruction steps and had full control of the integrated Tecne electric typewriter. Properly laid out and easy to read reports could thus be generated. With Olivettis co-operation we developed an interface to the 101 and 203 for our scientific instruments and used them for automativ data processing. The 203 was more popular with our customers as it was more powerful and the laid out typed reports were infinitely more preferable than the “bog-roll” presentation of the 101.

    We also srarted to use PDP8’s around 1968 These were as I recall considerably more expensive than even the 203. We sold the PDP to Research institute etc. type customers whereas the 203 was adequate for many routine process control applications.

    I wrote a multivariate least squares regression programme for the 203 wheras a simple least squares regression was all I could fit on a 101 (both with previous entry error recovery!)

  15. Fenman I started on the 101 and then went into a school situation with IBM 360 and punch-cards.  The “politics” was horrible, but then i discovered the PDP-8, the machine i am most known for being associated with.  [I personally created alt.sys.pdp8 on usenet].


  16. While reading your article I was taken back to an earlier time in my life. My first job after discharge from the United States Navy as an Electronics Technician serving on a Polaris submarine, was as a factory technician repairing the Programma 101 boards that were DOA as they came off of the assembly line at the Harrisburg, Pennsylvania assembly plant. After 18 months as a technician, I became a line foreman, first supervising the final assembly, test and shipping groups and than supervising the micro-module assembly. Olivetti built this plant in 1966-67 to build the Programma and their latest electric typewriter. At our peak we built 22 machines a day. The case and memory were shipped to us from Italy and we built the micro-modules and assembled and tested the boards. In final assembly we assembled, tested and shipped the machines. This assembly line was also one of the first to use the flow soldering technique. This was a wonderful time of my life and It would have been nice to see a reference, in your article, to the plant in Harrisburg and it’s history. Thanks for the memories. Steve

Leave a Reply

Comments are moderated and not published in real time. All comments that are not related to the post will be removed.