The year was 1974. My grandmother just returned to Hungary from a visit to Canada. She brought the most amazing gift an 11-year old geeklet could have had: an electronic pocket calculator!
One year later, my calculator expired. This was a result of my incessant attempts to operate it, in the absence of a suitable wall wart, from all sorts of external power sources. The poor calculator proved remarkably resilient, but 16V AC from a model train transformer eventually proved sufficient to kill it. Surprisingly, I found a service center in Budapest that repaired the unit for a modest fee. At this service center, I saw the marvel of marvels: a Texas Instruments SR-52 programmable calculator. In those days, most ordinary mortals never saw a computer except on television; here, I was holding a device that actually fit the palm of my hand, yet did everything a "real" computer was supposed to do! I was hooked. I returned to this service center many times, pleading that they let me play with this beast, until they eventually told me to never come back. In the years that followed, I often visited those few Budapest stores that sold privately imported devices. Unfortunately, owning one was out of the question: programmable calculators sold for more than a new Lada 1200 in the mid-1970s! Despite my pleading, my parents flatly refused to sell the family car and satisfy my obsession.
In the summer of 1979, things changed. The Hungarian electronics company HÃradástechnika began manufacturing an OEM version of Commodore's PR-100 programmable calculator and was selling it at the relatively moderate price of 6,500 Hungarian Forints. While still quite a bundle, more than the average monthly salary at the time, the amount was no longer unattainable. I was able to save up the money from a summer job and come September, I returned to school with my very own PTK-1072 pocket calculator.
Unlike the SR-52 that I fell in love with earlier, this calculator offered no permanent storage; when you switched it off, your programs were erased. Which is why I recorded my programs using a low-tech but reliable method: pencil and paper.
On of my best friends in class also had a calculator. A genuine Commodore unit, purchased for many hundreds of dollars during a visit to the States, his machine wasn't programmable, but it did everything and more a calculator was ever expected to do. In addition to complex arithmetic and advanced statistical functions, it had yet another built-in function that I've never seen: the Gamma function. I couldn't rest until I found a way to replicate this function on my programmable unit.
Since then, I used many other programmable calculators, including the Cadillac of the day, the TI-59, and a really strange Texas Instruments contraption, the SR-60A desktop calculator. But as personal computers became commonplace, my interest in electronic calculators waned, until they became just another tool I carried in my briefcase, to be used when needed, forgotten otherwise.
Sometime around 1990, however, I received an HP-97 desktop calculator as a gift. This machine was in a sad shape, battery pack missing, exterior all dirty, but it still had its charger, and it was still functional. Moreover, Hewlett-Packard still provided support for this unit, and I was thus able to obtain a replacement battery pack, magnetic cards, and manual. A bit of a cleaning effort made the machine look, well, not quite like new, but at least like a used, but well-respected tool. It continued to work flawlessly for several years and in fact, with its built-in printer it was a great tool at year's end when I was doing my taxes. Alas, one day its card reader began to act up, until eventually it completely broke down, leaving only a yucky, gummy deposit on cards.
Although I no longer used this calculator regularly, I was still quite fond of it. Its broken card reader kept bugging me. I took the calculator apart, removed the completely yuckified (for lack of a better term) rubber pinch wheel from inside the card reader, and kept scratching my head as to how I could possibly repair the unit.
Finally, in early 1999, success: I managed to fashion a suitable replacement wheel from a small rubber ball we originally bought as a toy for our cat. The reader has been working flawlessly for months now. My success also helped renew my interest in old programmable calculators, and I began to acquire old units wherever I could. Having gotten my hands on an old TI-59, I was amazed to discover that some nearly 20-year old magnetic cards I had were still readable by this machine's card reader! My simulation of the take-off of TU-154 passenger jets ran again after almost two decades... So I guess I am now a calculator collector. Hey, others collect stamps... at least calculators are useful devices!
During my "day job", I work with computers that are literally many orders of magnitude faster and bigger than these vintage calculators. Calculations that took hours, even days, to execute on these relics are done in less than a blink of an eye on my dual P-II workstation. Still, there is something uniquely magical about watching the blinking LEDs, about being able to fit a complex algorithm into a few dozen bytes of calculator memory. Perhaps part of the magic is that these calculators are relics of the Space Age: a seemingly bygone era when nothing seemed impossible, be it building a hand-held computer at a time when a radio with 12 transistors was considered high tech... or be it landing a man on the Moon in a vehicle that had less computing power than a modern washing machine.
There are many resources on the Web dedicated to calculators, calculator collecting, and restoration. Since my interest was always in programmable units, I decided to add something different: a set of pages devoted to early programmables. True to my roots, I used the factorial and the Gamma function as programming examples wherever possible, when demonstrating the programming model and techniques on these various units.