Pinball machines were the video games of their day. Back when they were king, there were no microcontrollers — everything was electromechanical. We know from experience that fixing these was difficult but we imagine that designing complex play behavior with a bunch of motors, relays, clutches, contacts, and more would have been excruciatingly difficult. [Technology Connections] has several videos about an old Aztec machine and he promises more to come. You can watch the first two below.
To give you an idea of what’s involved, imagine a very simple pinball machine that supports a single player and a handful of targets. When the ball hits a target, that could trigger a micro-switch. The switch closure could trigger a relay that closes a contact for a short period of time. That contact energizes a solenoid that advances the score wheels. So now, when a ball hits a target, the score wheel will spin enough to award ten points. To make sure there is enough time for the score to advance, the relay uses something like a mechanical flip flop.
Sound complicated? That’s nothing. Don’t forget, the machine also has to reset the score at the start of the game, count the ball in play, and end the game when the last ball returns. Then consider a real game. There will be multiple players and fancy sequences (e.g., hit the red target three times to award double scores for other targets).
While we knew a fair bit about the design of pinball machines already, we did learn a lot about their history and where the idea came from. The video also explains why it is called pinball since modern machines don’t really have pins — these were like relay-based computers with strange electromagnetic I/O devices.
While pinball machines were the best example of this sort of thing, there were also things like bowling machines and ladder-logic industrial control systems. We’ve even seen an electromechanical phone answering machine.
Mechanical computer is still a computer though
I don’t think this fits definition of a computer. It does not compute, only counts. It sequences a set of movements and follows hard-wired rules. It’s more like a FSM.
Still I find it quite impressive, what one can accomplish with some clever electromechanical design. There is also a quite interesting material on a jukebox with mechanical memory…
I think the only difference is that the logic is fixed rather than (easily) programmable. After all, at a low level, what more does a computer do except count?
I have a similar pinball machine waiting to be restored. The schematic is six feet long and two feet wide.
“After all, at a low level, what more does a computer do except count?”
A discreet logic circuit would change its outputs depending on the inputs. AND, NAND (Not And), XOR etc.
There’s no actual math in the sense of an ALU being done, I believe.
Those circuits can still be useful for microcontroller applications, though.
A code lock, for example, can be built using relay logic.
A series of relays can become self-holding if the right buttons on a number pad were being pressed.
However, if the wrong button is pressed (trap) , the power is cut to the chain of relays, causing a reset.
Such logic doesn’t really count in a common sense, yet it’s still useful logic.
Most EM pinball machines do more than count. Most have sequences that involve several and/or processes.
It IS a digital computer (that interacts with an analog ball), but not a general purpose computer (requires instructions in memory, some memory/registers and a CPU). It computes results by adding scores, subtracting balls, reading inputs and checking conditions. It also has sequences (Finite State Machine). It is programmable to a limited extend by a few configuration wires/switches. But it uses a lot of shortcuts. To make it a general purpose relay computer would require a lot more relays and a way to load a program.
“A finite-state machine (FSM) or finite-state automaton (FSA, plural: automata), finite automaton, or simply a state machine, is a mathematical model of computation. It is an abstract machine that can be in exactly one of a finite number of states at any given time. The FSM can change from one state to another in response to some inputs; the change from one state to another is called a transition. An FSM is defined by a list of its states, its initial state, and the inputs that trigger each transition. Finite-state machines are of two types—deterministic finite-state machines and non-deterministic finite-state machines. For any non-deterministic finite-state machine, an equivalent deterministic one can be constructed.”[1]
1. Finite-state machine
https://en.wikipedia.org/wiki/Finite-state_machine
Pinball is a pathetic and hopeless form of fighting gravity.
But on the other hand, what a deployment of technology, it leaves one speechless.
I once saw a pinball machine being maintained, it was very impressive. The fact that the technician seemed to know his way around was also impressive.
But it remains a futile machine !
“[…]pathetic and hopeless form of fighting gravity.”
Well, that certainly is a way to describe it.
Absent the gravity factor, characterization of pinball would be reduced to just a single word: “boring”. :-)
It’s only “boring” if you don’t understand how to play the game. Today’s pinball machines require skills on par with many video games. Sequences that must be fulfilled in specific orders, bonuses for various shots that require timing and skill. Those who are only knowledgeable in mashing a keyboard have no idea the level of complexity and skill involved in these games. Perhaps spending some time away from the screen would be enlightening.
He said it would be boring without gravity, which I think is undeniably true.
There are also variations of the game (Flipper automata, named after the flipper elements used in newer pinball machines; pachinko, a mixture of a slot machine and an arcade cab)..
Anyway, relay logic is impressive for sure.
The underlying principle is also used in early elevator logic, I suppose.
These all were early applications for 1-Bit and 4-Bit microcontrollers, too, I suppose.
I watched these videos last week and as soon as I saw the schematic I thought of ladder logic. That’s a lot of similarity between these mechanical pinball machines and the relay logic used in industrial controllers (which were mostly replaced with PLCs, many of which are programed with visual tools that draw ladder logic)
more than a lot, actually. I am a PLC programmer, and my retirement project is to buy an EM machine and fully automate it…in other words, have it play itself. The biggest challenge will be a digital camera that can tell where the ball is, in real time..as well as modeling the velocity and position of the flippers.
Just this morning I was contemplating gravity, and how much effort I make every moment of the day fighting it. Basically not even moments, the fight with gravity is a continuous process where the moments are created by me, and are basically nothing more than short moments where I created an equilibrium. I basically spend all day creating gravitational equilibriums so that I can have short moments of rest. :)
I got to think about it after I dropped something for some millionth time in my life and contemplated how life without gravity would be.
My conclusion was: completely boring. Apparently we need the challenge to properly function as an intelligent being.
My final conclusion was that the probability for any lifeform emerging in a gravity-less environment to ever develop intelligence is extremely close to 0. Of course life will always find a way. But if there are no continuous stresses like gravity, there is no pressure to develop intelligence. If intelligence would develop in these creatures, it would possibly take millions of times longer than it took us.
I also should state that one of my beliefs is that the development of intelligence is a response to challenges posed by the environment. But I have no evidence whatsoever that my belief is actually true. Which is a long way of saying that I don’t have a clue what I’m talking about, and that morningcontemplations like these are not reliable at all. They are amusement for myself and maybe a few other people. :D :P
That picture at the top of the article reminds me very much of the insides of theatre organs (wurlitzer, Compton), straight from the 1920s/30s and insanely complicated, addressing multiple instruments with brass mechanisms that’d wouldn’t look out of place in a telephone exchange.
It just so happens that Alec also has a couple of videos on that topic as well: https://youtu.be/NmGaXEmfTIo?si=fGq3JcOJRhIO-xAu
I see the wurlitzer jukebox but didn’t see a reference to wurlitzer or compton theatre organs, like this: https://www.youtube.com/watch?v=5ageV2dVD5I
My sister has a Tommy “Pinball Wizard” one. No circuits. Whenever she calls someone to fix it, the repair guys say “yeah, not sure what to do, I don’t want to touch it”.
This one. https://www.youtube.com/watch?v=6oovOnpQM8g Just needs the Elton John one to go with it.
Wow, I didn’t know that one existed. I only knew the Data East “Tommy Pinball Wizard” one, from 1994.
https://www.youtube.com/watch?v=n8wffa6qXMM
To be honest, the Data East one is one of the greatest pinball machines I have ever played!
One of my former managers used to restore pinball machines and he’d always have at least one in the office for us to play on. The Star Trek the Next Generation was my favorite.
I’ve got (and fixed) one of these… They are electro-mechanical nightmares… but very cool
I really prefer the sturdy relays of my Bally “The Wiggler” from 1967. zipper flippers, multi ball, wiggle alley. just beautiful. but i think in those days they did not use paint, just pigment with soap or something. the paint just fell off right from the factory. mine needs a new coating as well. I got it for free in 1984 and finally in 1995 i took the time to fix it properly. up till then the counters were not reset at the start of the game. of course a dirty relay contact somewhere. mine had a broken head plate but it turns out there is a shop here in The Netherlands who specializes in recreating those semi transparent glass plates. I only had to provide the artwork. after a evening retouching a photo of my plate it was sufficient for a new one.
the interesting thing is that there electromagnetic (EM) pinballs can multitask. upping the counters while also opening gates or zipping the flippers. 1st generation computer games worked also like this. build from simple TTL chips, there was no central processor unit involved, just quasi synchronized circuits rambling along. I think that the first generation solid state pinballs already had a cpu with roms in there.
There’s also a parallel development in telephone systems (landlines).
The original relay logic for dialing (electro-mechanical connection establishing) was being replaced by semiconductor electronic without movable parts.
Before quickly being replaced by digital circuits (digital/analog converters and analog/digital converters). That started in the 70s, I suppose.
I’m saying it this way, because digitalization wasn’t implemented the same everywhere, maybe.
Some countries adopted things like ISDN, for example.
The larger a country/nation, the slower the progress.
As a tween and teen, pinball machines- first electromechanical and then the early digital ones, cemented my growing knowledge of electronics. They are quite challenging, and quite educational, to work on. They also cemented my nickname after buying 373 old broken down machines at an auction. Cannibalizing some for parts, I fixed nearly 300 of them. Placed them in bars and convenience stores and made a bunch of $$$.
A lot work went into these video’s! Thanks for sharing. I used to pray a lot in the Church of the Silver Ball!
An absolutely excellent presentation and explanation of pinball operation. Thank you for the extraordinary effort these videos required – such a professional undertaking.
thats how old washing machine work, old vending machine and bunc of other mechanical computers, mechanical logic gates, you even had pneumatic and hydraulic logic gate machines…
A CAM “Computer”. ;-)
Casio made a number of full-on calculators, including some programmable machines, with relays. The Casio 14-A was their first: https://www.youtube.com/watch?v=tdxJ4HzJeA8 The dialog is in Japanese, but math is math. This machine is in Casio’s museum, and is fully operational. The 14-A was introduced, IIRC, in 1956 or so. And, it all fit in a nice desk. It’s a far cry from a pinball machine, but it does go to show what can be done with relays. Some of the later machines could extract square roots. The programmable machines used a removable spiky plastic wheel where the spikes could be broken off to form binary machine instructions. Kind of tough to describe, but it was essentially a read-only memory for program instructions. It wasn’t a stored program machine, e.g., it couldn’t modify its instructions under program control, nor could it use data as program instructions. Nonetheless, it was possible to program the machine to compute logarithms, trig functions, and other scientific and engineering functions.
I also forgot relay computers. The first were made in the US, mainly for military calculations. The Japanese made quite a line of fairly capable relay computers, mainly programmable by punched paper tape and punched cards. Here’s an example, courtesy of Curious Marc: https://www.youtube.com/watch?v=_j544ELauus . It is the Fujitsu FACOM 128B, built in 1958.
I used to work on them in the early 80’s, ingenious design, complicated until you understood things, follow the wires and you can fix anything on them.
Honestly I think they should teach the technology today, it helped so much when I was learning electronics, Gates, switches, transistors, etc.
Easier to visualize how things worked and how to put things together….
I used to refurbish old Gottlieb pin ball machines. In my opinion they are much more entertaining. One with Hawaiian genre had a girl on the top that would do a hula dance. Another was a soccer genre that you would rack up soccer balls in the top and kick them out to rack up points after the ball is lost. In my opinion they were much more enjoyable to work on. Repairing electronics is challenging and enjoyable, but electromechanical is more challenging.
Well, there is a whole other fork in the pinball heritage called The Bingo machine. They were completely electro mechanical but have very, very complex logic implemented with a randomizing unit that was at the heart of the particular machine. These were intended to be gambling machines and in fact were regulated by Federal regulations enacted in the 1950s.
When I graduated from high school, I worked for a pinball distributor refurbishing these machines for export. My boss at the time was a Dutchman, and was a great teacher and I learned a lot and probably refurbished over 40 machines before leaving for college.
At college in east Texas we were about 30 minutes from the Louisiana border. Since the bars in Texas closed before midnight during the week and 1:00 p.m. on weekends we would go to the bars just across the border for the music AND the gambling Bingo machines that you could win money on. In fact the bartender would actually pay you if you beat the machine. Having the knowledge of the circuitry, help us win more often then lose money.
I have 2 Bingo machines at home in my garage. They are a marvel in Electro Mechanical engineering. Also found a wonderful piece of software the emulates over 60 variation of these games ranging from early 1950s through 1980s. Have a look at the website. https://bingo.joopriem.nl/