We sometimes forget that the things we think of as trivial today were yesterday’s feats of extreme engineering. Consider the humble pocket calculator, these days so cheap and easy to construct that they’re essentially disposable. But building a simple “four-banger” calculator in 1962 was anything but a simple task, and it’s worth looking at what one of the giants upon whose shoulders we stand today accomplished with practically nothing.
If there’s anything that [Cliff Stoll]’s enthusiasm can’t make interesting, we don’t know what it would be, and he certainly does the job with this teardown and analysis of a vintage electronic calculator. You’ll remember [Cliff] from his book The Cuckoo’s Egg, documenting his mid-80s computer sleuthing that exposed a gang of black-hat hackers working for the KGB. [Cliff] came upon a pair of Friden EC-132 electronic calculators, and with the help of [Bob Ragen], the engineer who designed them in 1962, got one working. With a rack of PC boards, cleverly hinged to save space and stuffed with germanium transistors, a CRT display, and an acoustic delay-line memory, the calculators look ridiculous by today’s standards. But when you take a moment to ponder just how much work went into such a thing, it really makes you wonder how the old timers ever brought a product to market.
As a side note, it’s great to see the [Cliff] is still so energetic after all these years. Watching him jump about with such excitement and passion really gets us charged up.
Thanks to [Mark] and [Jerrad] for the near-simultaneous tips on this one.
“As a side note, it’s great to see the [Cliff] is still so energetic after all these years. Watching him jump about with such excitement and passion really gets us charged up.”
And that’s how one stays so young as we get older. Attitude, and love.
I had one of these until about 10 years ago. It worked. It was fun watching it calculate, the display actually flashed.
Really impressive bit of engineering. Lots of boards, and the contacts would get dirty when left sitting a while. Usually pulling the boards out, and plugging them back in would get rid of odd segment display weirdness.
“[I]t’s great to see the [Cliff] is still so energetic after all these years.”
It’s great to see anyone passionate about knowledge, but he’s not *that* old: 67. He would have been in 7th grade when this calculator was built.
I think it’s the Doc Brown mad scientist hair that makes him come across as having been around forever.
Damn, Doc Brown reincarnated!
Interesting video.
Wait, Doc Brown died?!?!?
Noooooooooooooooooooooooooooooooooooooo!
Yes, he died in the future.
This is astonishing, especially the delay line memory. On conceptual level transistor based memory is so much easier to understand. I have serious difficulties trying to grasp that wire memory! On a side note, same thing goes for analog video. Lighting up pixels in coordinates x and y sounds easy, but all the scanline and timing stuff feels kinda like black magic.
The wire memory is a form of DRAM, it has to by cyclically refreshed. Think of it as a big FIFO buffer.
Oh and btw analog video doesn’t technically have pixels, as there is no discrete positioning, everything is nice and continuous. Yes, the screen has a mask which is fixed (and CCD sensors pushed out tubes from cameras looong ago), but the picture does not have a rigid placement like on digital displays.
Are there start and end bits so you don’t lose track of the order?
Sort of. They’re called the *front porch”, “back porch” and “blanking interval/period” in order they function as (I’m about to send a scan line), (I just finished sending a scanline), and (compare the time now to the last time I went blank so you will know how quickly I am advancing to new scanlines)
Neorpheus, you are correctly describing a raster scan CRT display, but this calculator uses a vector X-Y display. A visible stroke is created when the beam moves from one XY location to another while the beam is on. They are no scan lines.
AKA likens the delay line memory to a DRAM, but the contents are NOT randomly accessible. It is true that the results are volatile and must be refreshed. The FIFO analogy is closer to the truth, but there is no clocking. The delay time is determined by the propagation time of the pulse through the wire and the length of the wire. Also, since this is a single phosphor monochrome display, there is no shadow mask.
http://www.oldcalculatormuseum.com/f130block-l.jpg
Block diagram
Nope. No start nor end bits. When you send bits in one end, they come out the other end, only delayed in time….and in the same order.
If you want to keep the same value, just amplify and clean up the signal and re-launch it back into the same line.
Wonderful!
Very interesting video I enjoyed watching it, very nice to see the machines operate.
The calculations are interesting, I hope he will explain in another video how the CRT was controlled. I expect it to be vector graphics, but how do you do this with as simple as possible, considering transistors were expensive.
Also watched the extra’s, clif explaining RPN, as I don’t own (never did) such a calculator I never understood the usefulness of it compared to modern calculators. But he explained it with such joy and made me realize that this is especially an advantage in early calculators.
Thanks for posting!
It wasn’t quite the first! Beaten by a few months by the British ANITA electronic calculator!
https://en.m.wikipedia.org/wiki/Sumlock_ANITA_calculator
I have an ANITA 1011 from the late 60s! It still had no chips, but about 10 circuit boards filled with DTL logic!
http://anita-calculators.info/html/anita_1010.html
I was careful to not claim these were the first electronic calculators because I knew about the ANITAs. But those used vacuum tubes, while the Friden was all solid state with the exception of the display and memory. Any way you slice it, the engineering of these things was top notch.
You guys are good… so much idea and you can achieve it….
I’ve been meaning to buy a klien bottle off him before he kicks the bucket of infinite surface area.
*off of* my grammar isn’t what it used to be.
Not only your grammar.
These bottles are named after Felix Klein.
I’m sure his widow will have them lined up on the sidewalk a week after his funeral, 25 cents each.
B^)
A mathematician named Klein,
Thought the Möbius band was divine,
He said “If you glue
The edges of two,
You’ll get a weird bottle like mine”
On a much more mundane note I still recall using my Timex Sinclair ZX-80 with 1 KB of RAM when I was in high school (1979 ish) to calculate the 2nd, 3rd and 4th powers of 1 through 5 in BASIC. Hit the enter button and about 1 second later the screen output (black and white TV) reappeared with the results. Actually called my parents in to my room to show them how ‘fast’ the computer could do all these calculations in less than a second. Amazing times.
Along those same lines, I remember about 1984 or so, I had a complex benchmark test that ran on HP workstions (in Rocky Miuntain Basic). It ran in about 45 seconds on the 68000 based 9836/9826/9820. Then HP came out with their 320 (9000 series 300) based upon 68020. Same benchmark ran in 2.3 seconds. I remember saying to another nearby engineer “I don’t care if computers ever get any faster than this!” Same benchmark would now likely run in milliseconds.
It’s relative to what you get used to.
A few years earlier I was bitching to my brother about how slow my Atari 400 was. “It may be slow”, he said, “but I ran three insurance companies on one that wasn’t as powerful when I started back in 70s.” He was serious.
In Romania the communist regime bought some of these, took them apart, rebuilt them using eastern block components and presented them as the greatest achievement of the communist progress.
I have one of these replicas – made with EFT3xx Romanian-made germanium transistors. It looks THE SAME on the outside. I did not know until now that it ts normal for the screen to be filled with numbers.
Maybe I will show it on hackaday.io after I get back from Sahara.
Only a few years after this, HP came out with a similar desktop/CRT calculator, the HP 9100. The HP had a full set of scientific functions.
With so many scientific functions, the algorithms would require nearly 32 K-bits of ROM. But this was back in 1967. 32-kbits was a *LOT* of memory back then.
How to implement 32-kbits of ROM? Using “leading” edge PCBs with 10-mil traces of course!
There were no components, per se, to store the bits. The traces themselves were the bits of this inductive trace memory. A clockwise trace, a 1. Counter-clockwise, a 0. IIRC, about 1000-bits per square inch.
http://www.hp9825.com/assets/images/HP_9100A_ROM_microphotograph.jpg
(btw, the RAM, much smaller by comparison, used ferrite core memory).
I would love to hear more about this calculator and memory. Looks amazing.
This repair video is also hugely entertaining. Main failure mode of these things seems to be the diodes in the 3KV voltage multiplier.
https://www.youtube.com/watch?v=fie-Z-mjnyU