Advent of Computing - Episode 48 - Electric Ping-Pong

Episode Date: January 25, 2021

Sometimes an idea is so good it keeps showing up. Electronic ping-pong games are one of those ideas. The game was independently invented at least twice, in 1958 and then in 1966. But, here's the thing, PONG didn't come around until the 70s. What were theses earlier tennis games? Did Atari steel the idea for their first hit? Today we go on an analog journey to find some answers. Like the show? Then why not head over and support me on Patreon. Perks include early access to future episodes, and bonus content: https://www.patreon.com/adventofcomputing

In 2021, the video game industry is massive. The overall market is measured more in the hundreds of billions of dollars. That's a whole lot of money. Most of that money comes from software. In fact, if we discount things like merchandise and licensing, then basically all of that money comes from software. So it may be tempting to assume that games and software have always been synonymous. Just trace a simple line from Space War to Doom to Skyrim to today, and you're done. But that wouldn't be very interesting. And it also wouldn't be entirely accurate. The story has a little bit of complication to it, which lands it firmly in my wheelhouse. Mainframe games played

a big role in the history of the medium, but they weren't the only show in town. Some of the earliest games weren't on computers. In fact, they weren't even digital at all. Before fancy graphics, even before color graphics, there was tennis. But be careful, because I'm not talking about Atari's breakout hit. Pong was a relative latecomer to the electronic ping pong game. By the time Atari formed, at least two games that were eerily similar to Pong had already been produced. Both were purely analog devices. Both simulated Tennis with nothing but electronic components. And both were totally free from any type of software. If you're anything like me, then that realization should leave you with two big questions. Why did

Tennis find its way into the electronic realm at all. And on a different note, does this mean that Atari stole the idea for Pong? Welcome back to Advent of Computing. I'm your host, Sean Haas, and this is episode 48, Electric Ping Pong. Today, we're going to be taking a little diversion back into the realm of video games. We're going to be looking at what's been called the first quote-unquote video game, and really what that term actually means in regards to this creation. But most of this episode isn't going to be about Atari's Pong. That's not the first video game, not by a long shot.

As it turns out, Pong wasn't even the first tabletop tennis emulator out there. Now, I think my favorite phenomenon, and one we're going to be dealing with for most of this episode, is the idea of parallel evolution. Some ideas are just so good that multiple unrelated people have them around the same time. I usually broach this topic with pretty big ideas. Email-like systems start popping up all over the place as soon as users are sharing computers. Networking follows a really similar pattern. It just takes getting enough computers to network,

and then we have multiple government and federal-level agencies all building their own networks with either very little or no knowledge of their contemporaries. Networking was just too good of an idea to only appear once. Same with email, and, well, for some reason this is also true of electronic tennis, believe it or not. This video game has been independently invented at least twice. Once at a physics lab in 1958, and then again at a defense contractor in 1966. Now, you should notice that I haven't mentioned Atari in that timeline. Perhaps the company most associated with ping-pong-based games.

Now, that's for good reason. There is evidence that Atari didn't actually invent ping-pong. Rather, they bootlegged one of these earlier games. This episode, we're going to be looking at these first two incarnations, Tennis for Two and the somewhat mysterious sounding Brown Box. Then we'll be closing things out with Atari, Pong, and the lawsuit that follows. Did Atari steal the idea for Pong? Well, that actually gets complicated.

So grab a quarter and buckle in, because Pong's history isn't as straightforward as one may hope. The beginnings of the very first video game start in, surprisingly enough, the same place as the first electronic computers. That's the Manhattan Project. However, the story takes place in a slightly different section of that project than we've explored before. Machines like ENIAC were more on the theory and physics side of things. The MIMEX would be designed as a response to the administrative issues faced at Los Alamos.

But the start of Tennis for Two is a lot closer to the physical bomb itself. During World War II, William Higginbotham headed up the electronics division at Los Alamos. Specifically, he designed the circuits that controlled the bomb's detonation, plus testing and monitoring instruments. After years developing radar systems at MIT, Higginbotham was perhaps the best person for this kind of sensitive work. He knew his way around analog circuits really well, and definitely knew how to get the most out of a CRT. And it must have been exciting work, really. The scientists in the Manhattan Project were breaking new ground on a daily basis.

But Higginbotham would never see his contribution as part of some grand achievement. By the end of 1945, he was frankly horrified at what he had been involved with. The sheer destruction caused by the atom bomb that he helped build horrified him on a very deep level. And he wasn't alone. A contingent within Los Alamos were sickened by what their research turned into. Many scientists believed in the importance of their research. Cracking the atom could have great consequences for humanity. But the Trinity test, the first atomic bomb test, made a lot of them reassess their work. The destruction of Hiroshima and Nagasaki broke the spell for many more. Higginbotham was among the most outspoken of this group. In the coming years, he would help found the Federation of American Scientists.

This is a group that would work towards nuclear disarmament and eventually non-proliferation, essentially trying to undo the result of its members' earlier work. In 1947, Brookhaven National Laboratory was founded. It was a lab that specialized in the responsible peacetime use of atomic energy. Higginbotham became one of the lab's very first employees. In his new job, Higginbotham specialized in, what else, but instrumentation. That was his biggest area of expertise expertise and a really important, although somewhat hidden, field. In the 40s, this meant a lot of CRTs, a lot of oscilloscopes, and a whole lot of test equipment. Higginbotham was in the middle of almost any work done at Brookhaven,

since, you know, researchers have to have instrumentation to actually see their results. You can't do research blind. That's dangerous and unproductive. If that's not enough to have on his shoulders, there was one other responsibility that Higginbotham had to deal with, and really all researchers at Brookhaven had to deal with. Open houses. The lab regularly hosted open houses. Now, this is a pretty normal thing for labs and any research institutions to run. It's a chance for the public to see what

exactly is going on next door, and for researchers to prove how important their work is to lay people. You get to mingle with neighbors, talk shop, maybe share a drink and some hors d'oeuvres. It's a good time for all involved. For the most part, it's pretty easy to get everyday people excited about big ideas like nuclear physics. Just to have a short talk about atomic reactors, say something about the power of the atom, and presto, you have people in awe at your work. But when it comes to the actual nuts and bolts of nuclear physics, things get more difficult, to say the least. Maybe you could do it in a classroom, but definitely not in a few hours of milling around a lab. This was a common problem for the open houses at Brookhaven. They were composed mainly of poster boards, demonstrations of equipment,

and then a pile of PhDs roaming around talking to people. While interesting to some, it makes for a pretty one-dimensional experience. Most people can't really engage with an oscilloscope and a poster. In the fall of 1958, Brookhaven was gearing up for another open house, and it was planned to be just like any other. The instrument division was planning to show off their newest toys. A sodium ion detector, racks full of dials and switches, a new pulse analyzer. All things that are really, really important to researchers, but on their face, not all that exciting to outsiders. Maybe you can look at some flashing lights for a few minutes, but, you know, what does it actually do?

Higginbotham wanted to do something different, so he started looking for ways to get their guests more involved in the real action. The family down the block probably can't use a sodium ion detector, but Higginbotham figured there had to be some way to demonstrate what this kind of equipment could do. His idea was simple, but effective. He'd write later in a series of notes, quote, It occurred to me that it might liven up the place to have a game that people could play, and which would convey the message that our scientific endeavors had relevance for society.

and which would convey the message that our scientific endeavors had relevance for society. We used oscilloscopes in designing electronic circuits, and sometimes to display experimental data as it was being accumulated. End quote. It was a way to introduce the public to the broad idea of the research done at Brookhaven without getting too far into the details. An oscilloscope was a common sight around the lab. In fact, they're still common in labs today. The actual hardware of an oscilloscope is pretty simple. It's a CRT tube with some circuits for steering around the cathode ray. By driving the

beam around, you can display data or just show squiggly lines. Most commonly, these kinds of scopes are used to look at signals or outputs of analog circuits, but they can be configured to do just about anything you want. It's a really useful tool for researchers and the kind of cool visual gadget that the uninitiated love to look at. Most open houses already had a few scopes on display. The difference was that those scopes were kind of just there. At best, they showed some signals. Higginbotham figured, and correctly so, the guests would be more invested in those squiggly signals if they could actually control them somehow. Further,

he decided that the best way to do this was with a simple game. Now, this is where we get into some strange and new territory. Higginbotham decided pretty quickly that he was going to make a simple game. Now, this is where we get into some strange and new territory. Higginbotham decided pretty quickly that he was going to make a simple tennis game where two players bounced a ball back and forth over a net. According to him, the idea came while leafing through the manual for some lab equipment. The manual described how to wire up a scope to trace trajectories of projectiles. It was just a small step to add in some wires and buttons and make the circuit more interactive. Supposedly, the manual in question was for a Donner Model 30 analog computer. But here's the thing.

I can't find any information about the Donner Model 30 analog computer. There is a Model 3500, 3400, 10-20, but the Model 30 appears to either be a closely guarded secret or some kind of typo. From photos of Higginbotham's setup, it looks like he might have been using a 3400, but I'm not 100% sure. Ultimately, it doesn't really matter if it was a 30 or 3400. Either way, we're dealing with a radically different kind of computer. You see, Higginbotham's weapon of choice here was an electronic analog computer. Just for context, today we use electronic digital computers.

The key differences are how these computers represent numbers and how they perform math. Under the current digital system, we encode numbers as binary data, and we use logical operations to build up to math. Older mechanical analog computers usually encode numbers as some kind of rotation on a shaft or gear, and then use something like gearing ratios to operate on that data. Electronic analog computers are an entirely different beast. They store numbers as a voltage value on a wire, and then use analog circuit components to model equations. So a practitioner of this ancient art would build up a circuit using patch cables, capacitors, amplifiers, resistors, and transistors.

The other fun aspect of these machines was their strange power supplies. Now this is a little in the weeds, but I think it's interesting enough to mention. Since voltage is used to denote values, you end up needing a wide range of voltages. That translates into pretty high voltage supplies, and very highly accurate power supplies. The Darner 3400, for instance, supplied plus 150 and plus 300 volts. Modern digital circuits usually max out at plus 5. All this is to say that Higginbotham was working with very specialized equipment, and definitely not something that we'd call a computer in the modern day. To that end, the game he developed, called Tennis for Two, was not a program.

Instead, it was a circuit built up on the plug board of a Donner computer. This was a place where Higginbotham's experience factored heavily. From his work at Los Alamos and his earlier experience at MIT, he knew how to get a scope to do what he wanted. In all, it only took him a few hours to lay out Tennis for Two on a piece of paper. It took a few more days of work and help from a coworker to get the game up and running in reality. This is more personal speculation, but I also think that Higginbotham's horror following his involvement in World War II contributed to Tennis for Two. His work to prevent nuclear proliferation was a key aspect of his life

that has to have some kind of effect on his career in general. And I'd wager it made him view public relations in a different light than the rest of his colleagues. Helping the public understand his work and what nuclear research was really about was a really important step in his larger goal. nuclear research was really about was a really important step in his larger goal. Public opinion mattered a whole lot. It can affect how a government approaches new technology, and possibly affect nuclear regulation policies. Tennis for Two could serve as a way to get more citizens in the door. But that's just my own speculation.

If you had your own sodium ion detector, you'd probably find a pretty big grain of salt to take with this one. Anyway, Tennis for Two was finished up in time for the fall 1958 open house. So what did the game look like? Well, it wasn't 100% the Pong you may recognize. Higginbotham rendered the game with a side view. The net was a vertical line in the middle of the scope, with another line on the bottom for the ground. The ball was just a single blip on the screen, and actually the only moving object on the screen at all. There were simple and well-known circuits for tracing trajectories using a Donner computer.

So the ball observed all of the expected physics. It bounced off the net, bounced off the ground, and it flew in arcs across the scope. Those are all the graphics you really got. There were no paddles on the screen, just a bouncing ball and a static court. Tennis for Two had some strange gameplay to match. Each opponent had a controller with a dial and a single button. This is a similar arrangement to later paddle controllers sold by Atari and others. But the dial didn't control the location of the paddle. There wasn't a paddle to control. Instead, it changed the angle with which you hit the ball. Pressing the button would register a hit. Tennis for Two wasn't really about movement so

much as timing. If you hit the ball at the wrong time, then it flies off the scope and, ostensibly, you lose the match. The simplicity of Tennis for Two can't be understated, but it is an effective demo. Rotating the dial and then pressing the controller's button set the parameters for the ball's new trajectory. From there, the game simulates a bouncy projectile flying through a small space. There you have physics, instrumentation, and application all tied up into a nice little bundle. The other wonderful part of the setup is how Donner computers were wired up. Everything is plugged into a quote-unquote problem board. It's a grid of jacks that are waiting for patch cables and components. The model that Higginbotham was working

on displayed the problem board right in front of the machine. So if someone asked how Tennis for Two worked, then Higginbotham could bring them over and expose the circuit itself. By all accounts, Tennis for Two was an unqualified success, and the highlight of Brookhaven's open house. For perhaps the first time, visitors were actually lining up to see one of the demonstrations. Higginbotham had tapped into something that fascinated onlookers. He found a way to pull back the lab's veil just a little. Sure, you normally use all this equipment for serious research, but you can just as easily wire it up for fun.

The big question is, though, should we consider Tennis for Two the first video game? And I think that gets a little tricky. It's interactive, you play it, and there's a definite way to win and a way to lose. You have the ultimate goal of getting the ball past your opponent. But there isn't a scoring system and when the ball goes out of bounds, everything just resets. The other big factor we have to weigh are earlier games. The most commonly brought up example is usually Birdie the Brain. This was a purpose-built machine that played tic-tac-toe with a human opponent. There were actually a number of similar machines designed in the very early 1950s. They were all for demonstration purposes, all designed for one task,

and all replicated some existing game using circuits. We could get into some nitpicky parts of the argument here. Tennis for Two was using something similar to a video display after all, but I think there's a more relevant angle to take. The benchmark that I usually like to use when talking about early games is, does the electronic-ness of it matter? Is it doing anything different with the medium, or is it just an electrified version of an existing game? Birdie the Brain doesn't really do well by this test. It's electronic tic-tac-toe. Sure, the machine can beat a human, but you could have accomplished the same thing with pen and paper and a printout. The video part doesn't really change the game. But Tennis for Two actually

passes this test handily. On the most simple level, it lets you play tennis where you normally aren't supposed to, in a nuclear physics laboratory. But everything becomes a lot more clear with the 1959 open house. After the huge success in 58, Higginbotham upgraded his game. He added a larger display and, importantly, variable gravity. The improved demo could be switched to play on Earth, the Moon, or Jupiter. Now, unless I've missed some pretty big developments lately, you can't play tennis on Jupiter. For me, this is a really clear case of the medium of video games being exploited. And Tennis for Two was really one of the first times this happens. While successful at Brookhaven, Higginbotham's creation never left the lab. He also never patented it. In his view,

he hadn't created anything new, just strung together existing technology. That means that after 1959, the game disappeared. Only a small number of people ever knew it existed. And with that, we're set up for the second invention of electronic ping pong. To get up to the next game, we actually need to go back in time to 1955. up to the next game, we actually need to go back in time to 1955. That's the year that Ralph Baer, an engineer at Laurel Electronics, started thinking about video games. Baer's latest project was to design a new television, a product that was fast becoming boring. By the middle of the 50s, televisions were somewhat commonplace. Just about every electronics firm was creating their own

model, and they were all nearly the same. You have a CRT connected up to some circuits for tuning to channels and receiving video signals, and then everything's just packed into some wooden cabinet. Broadcast signals were all standardized by 1955, so there wasn't room to change there. A manufacturer could really only compete in terms of price point and maybe the design of the TV's case. There wasn't much space to innovate without making sweeping changes and taking huge risks. Baer looked at things from a slightly different angle. While building up the circuits for his new TV, he came to an important realization. You can actually drive a TV tube with more than just a broadcast signal.

It's just a CRT after all, so with some careful wiring, a TV can be driven just like an oscilloscope. Bayer figured it would be possible to build a game into the new TV. It would just be something simple, mainly a gimmick to try to make his new television stand out from the crowd. When he brought the idea up to his boss at Laurel, he hit a barrier. It seemed he was the only person at the company that thought a television game was a good idea. So he tucked the rough idea away for some later use. By 1966, Bayer was in a much better position professionally. Since leaving Laurel Electronics, he had gone through a series of jobs, eventually

being hired by Sanders Associates as the head of their equipment design division. In this new position, Baer managed a team of engineers who worked primarily on government defense contracts. a team of engineers who worked primarily on government defense contracts. It was pretty serious work, but Baer had a lot of autonomy. It was enough space to decide what was serious enough to pursue and what projects were unimportant. 1966 would be the year that his long-forgotten television game came back to the focus of his mind. Quoting from Baer, quote, came back to the focus of his mind. Quoting from Baer, quote, During a business trip to New York City on the last day of August in 1966, while waiting at a bus terminal for another Sanders engineer to come into town for a meeting with a client,

I jotted down some notes on the subject of using ordinary home TV sets to play games. I distinctly recall sitting there on a sunny day and writing on a small spiral notebook perched on one knee." Sometimes it really does pay off to just step out of the office, right? Baer knew that this was one of those projects that he did want to pursue, and that he could probably get away with it. On his return to his office, Baer drafted his notes into a more formal document. That paper works out to just two pages of notes and provides a roadmap for going from idea to a full implementation. The broad technical strokes are similar to Tennis for Two,

but Baer and Higginbotham are coming at the problem from very different places. Higginbotham wanted to make a compelling demo. Baer wanted to make a compelling product. The net result is that we get two independently invented electronic tennis games. Baer's design had shifted considerably since his days at Laurel. His new creation wouldn't be part of a television, but rather a separate device that plugged into a TV. This magical game box would create an image, convert that into a broadcast signal, then feed that into a standard television. The really smart part here is that Bear is leveraging existing technology to do something totally new. The plan was to create his own little broadcast channel,

essentially hijacking a normal TV signal. Instead of showing a video feed, Bear would just present some kind of interactive graphics. Bear's device would eventually be called the Brown Box, since it was housed in a faux woodgrain case. So I'm gonna stick with that naming convention even though earlier prototypes weren't all brown. Anyway, the heart of the brown box was its RF modulator. This is what actually creates the radio frequencies of a normal TV broadcast. It was the bridge between Bear's circuits and a user's television. RF modulators weren't really new technology, so that part wasn't going to be an issue.

The bigger task was figuring out how to drive that modulator based off user inputs. Now, running current directly into an oscilloscope is pretty easy. From there, it just takes a few steps to build up an image and then vary the current to vary the image. But creating a signal that a TV understands, and one that actually displays an image and not just static, that's a little bit more tricky. Baer would approach this problem in steps, working up in complexity until he had an interactive and intelligible image. The 1966 proposal goes into pretty deep detail on each of these steps, and this is another thing that really puts the brown box in a league above Tennis for Two.

This was a sustained, planned, and well-organized project. Each of the steps in Baer's proposal were actually games in themselves. That's one of the interesting things about his overall plan. It wasn't just a matter of generating increasingly complex images. It was more like a directed exploration of the new video game medium. No one really knew what a video game would look like. There just wasn't any reference material. So they were kind of out on their own.

The only way to see what worked was to try a bunch of different ideas. And why not start at the simplest one possible? For a quick taste, let's just look at the first step in this explorative process. Baer described a simple oscillating circuit that modified the color of the broadcast signal. The graphics were about as simple as you can get, just a flat screen which could switch colors. As the screen flashed, a player would spin a flywheel, and as it rotated, the screen would cycle through a series of colors. The goal of the game was simple.

Stop the flywheel on a certain color to score points. Eventually, this would build up to full raster graphics and two-player games. Each step used experience gained from the previous games. It was a great plan, with very tangible and fun progress each step of the way. But it was just that, a plan. The real start of work went something like this, from Berrigan, quote, I put a technician on a bench and had him build television game number one. We didn't call it that then, but that's what it was. It was basically a demonstration of how to put a spot on a screen, how to move it laterally, horizontally, and vertically,

and how to color it, how to color the background. Once that was done, really nothing happened for several months. It was not until early next year that I conferred with a director of R&D, suggested that maybe he ought to put a few bucks into it and make it a legitimate project." Maybe a little less spectacular than going from zero to game, but it was a start. Game number one was really the bare minimum tech demo, but it covered basically everything the bare needed to make more complex video games. Creating a moving pixel and changing colors

really covered the whole spectrum of possibilities. It was just a matter of making more and more moving pixels. But the recurring issue would be getting funding to make more moving pixels. While Bear had a lot of control over his division, he didn't have control over its finances. And this is where Bear's plan really started to show its brilliance. At each step along the way, there was a playable game. That's fun, and it adds a certain sense of momentum, but there's a more savvy purpose. Each step has a demo to show to management. Each step was a chance to get more funding. So even though the initial six-step program laid out in 1966 didn't come to pass,

a very similar progression occurred. Eventually, Bear was able to get some R&D funding by showing off Game Number One. The following year, 1967, saw the creation of Game Number Two. Really, a poetic name for a sequel. Armed with a little bit of cash, a lab full of equipment, and experience gained from game number one, a new series of circuits formed. Bear pulled the same maneuver here, he brought on a second technician to assist. That technician was named Bill Harris, a circuit designer who would continue with the project for some time. Analog circuits handled generating the actual signal,

while a series of transistors were added to control parts of the operation. But by and large, the circuitry was analog. And this is where a more interesting game starts to develop. Bear called this first operational TV game, Chase. Game, maybe a bit of an overstatement here. Two players used controllers to move two dots on the screen. The goal was to reach and catch the opponent's dot. Just about as simple as you can get. Two other demos were built for game number two. The earlier color wheel game was finally implemented onto circuits, and a light gun-based game was also developed. We're still outside the realm of computing, so these games weren't programs per se. Each game was essentially a small control circuit that drove the RF generator based on a

series of inputs. A switch was used to change which circuit was running the show. So no cartridges yet, no code, and nothing fancy. However, game number two was showing promise. With three simple demos, it was time to continue the cycle. Bear scheduled a meeting to show off his work to management at Sanders Associates, with the end goal of securing more funding for the next bootstrapped phase of the project. The meeting ended up including the company's stakeholders and even its president, Royden Sanders himself. Needless to say, it must have been a high-stress situation. Here was Baer trying to convince a roomful of his bosses that, you know, a military contractor

should really be investing in these new television games. So much shockingly, there was only one road bump to the demo. From Berrigan, quote, The demo went pretty well, but after it was over, Royden Sanders said a natural thing. What are we going to do with it, right? How do we make money out of this? This is a military electronics company, you know. How the hell are we going to handle this? So I just said, let me worry about that, end quote. And somehow, that was enough of an answer for Sanders. Baer was able to kick off the final stage of his project, turning prototypes into a product. Back in the lab, Bear and Harris set to work, but an issue came up almost immediately. They had been working in the realm of prototyping.

The primary concern had just been if a television game was even possible. The shift to a sellable product really brought up a new concern. Cost. After ironing out some kinks and building up a brand new board, the two started to run up a bill of sales. Parts and fab came out to about $25, adding in margins that would sell retail for around $50. Now, crunching some numbers, adjusting for inflation, that's just shy of $400 to plot some dots on a screen if we're using 2020 money. A third member, Bill Rush, was brought on the team to try and either get costs down or value up. Rush spent a few weeks jotting down notes, crafting ideas, and really puzzling over the mess of wires that Baron Harrison had produced. And this is where a drastic change occurred.

One path the team was exploring was to expand what their television game could do. Basically, if you can't make it cheaper, then make it more interesting. The end of that path was ping pong. A working prototype of their electronic ping pong game was created, and Bear realized that they had struck gold. The reinvention of the game differed from Tennis for Two. The game was played from a top-down view. Players used a dial to move the racket up and down the court.

A ball bounced between each racket with the end goal of getting it past your opponent's paddle. The final complication was the so-called English control. A second dial on the controllers allowed players to adjust the angle of the racket. You couldn't see this on the screen, but the adjustment affected how the ball rebounded. Once everything was finalized, the entire design of the brown box was based heavily around this one game. Each controller had three dials, one for vertical movement, one for English, and then another for horizontal position in some games that supported moving horizontally. There was also a button used to serve the ball.

It's a simple controller, but it's pretty effective for playing ping pong plus some other games. it's pretty effective for playing ping pong plus some other games. Bear's ping pong game worked for the same reason that Tennis for Two did. It was just fun. The more I read, it seems that electronic ping pong is somewhere close to the minimum complexity needed for an engaging video game. There's something captivating about seeing a dot bounce around on the screen. Even without your input, you can really tell the game is doing something. Sure, it's doing something exceedingly simple, but it's something that you haven't seen before.

To become a pro, you need reaction time and dexterity, but not all that much. You also need to plan ahead, but not that far into the future. There's a certain amount of skill needed to become a ping pong champ, but it's not unattainable. And perhaps most importantly, electronic tennis is just complicated enough to make you wonder how someone made this happen. After some tweaks and additions, we arrive at the final game, and it was packaged up lovingly in a woodgrain box. Thus, the brown box was complete. In both form and function, Bear's device is pretty close to more recent video game consoles. A large box houses all the important hardware, and the case was an RF generator and a series of circuits for each game. The box connected up to a standard

TV set to display graphics, and two controllers were wired up to the box. There were also connections for a light gun. Games were selected via a series of switches on the front face of the brown box. Now, ostensibly, there was a large set of games, but they were all really variations on ping pong. The brown box couldn't render fully rasterized graphics. It just wasn't sophisticated enough to control a whole lot of pixels. This just came down to technical constraints, really. To display complex graphics on a TV, you need somewhere to store that image. In the world of computers, that's easy. You use this little thing called RAM. But Bear was working totally analog, or at least mostly analog, so that wasn't an option. So instead, the brown box could display a grand total of four pixel-like

objects. May want to call them sprites, but that's still a little grandiose. You get a vertical line, two paddles, and a small box. So any game on the system looked a little bit like ping pong. As 1967 turned into 1968, the Brown Box team started looking at ways to actually make money off their creation. You see, the whole project was conducted a little bit backwards. Bear had developed a video game console, the first home video game console at all. And he did it because he thought it was a good idea. His co-workers and even management agreed that, yeah, it's a cool idea. But, well, they weren't solving a problem. There wasn't a problem to solve. There wasn't a niche to fill. They just had the school video game thing that you could use to play ping pong. To make matters worse, Bear had taken up

the responsibility of making money off the brown box. Baer, the electrical engineer, was trying to sell a totally new kind of device. Baer described it best like this, quote, Do I know anything about marketing? No. Do I have any connections? Do I have any idea how the commercial consumer product business worked? Heck no, end quote. His plan was to license the brown box to interested parties. So in early 68, Bear applied for a patent on his video game console. While the application was being processed, Bear started making calls and giving demos of the brown box.

It was kind of an on-again, off-again process for a while. Games weren't Sanders Associates' main product. Defense work took precedent over everything. Some early leads would eventually drop out due to financial issues. Things were just slow, to say the least. The silver lining was that everyone who saw the brown box was impressed. It was impressive technology, and more importantly, it was fun technology. After a critical mass of demos, something was just bound to happen. So in 1969, Magnavox, a TV and radio manufacturer, became that something that happened. They entered into an exclusive licensing agreement with Sanders Associates. Magnavox would adapt the brown box design slightly, package it up in a nicer,

not-quite-so-woodgrain case, and sell it as the Magnavox Odyssey. In 1972, the console hit shelves at a whopping $99, which inflates up to $600 or so today. It was a hefty cost, but there was nothing like it, so they could kind of command their own price. The Odyssey was the first home video game console money could buy. There were a handful of arcade games, but they had yet to break into the mainstream. The Odyssey was very much in a league of its own. Over the next three years, over 300,000 consoles were sold. Each sale netted Sanders Associates a slick licensing fee. Bear had successfully taken an idle idea all the way into a revolutionary product, but the Odyssey wouldn't be alone in the market for very long. For an episode that has Pong in the title, I've spent a lot of time not talking about Pong.

The big reason for that is that Pong wasn't some third, totally independent invention of electronic tennis. Atari didn't come up with the game out of whole cloth like Bear or Higginbotham. Instead, Pong was a remake of earlier games. Call it the first remaster, if you will. To explain this final twist, we need to get into how Atari formed and how Pong became a contentious video game. In 1972, Atari was incorporated by Ted Dabney and Nolan Bushnell. They each had a bit of a winding path on the way to forming Atari, but I want to focus on Bushnell in particular. In the 60s, he was an engineering student at the University of Utah. Importantly, that university owned a

DEC PDP-1 mainframe, a pretty hot commodity at the time. The machine was used primarily for research, but there was something more exciting kicking around the computer lab. A little something by the name of Space War. This was the first computer video game in the entire world. It had been developed in the very early 60s at MIT and really quickly spread to other campuses. Bushnell was one of the many students exposed to the game, and one of the many students fascinated by it. By our standards, Space War might be a simple game, but it was actually an amazing piece of software for the day. Two players fly around spaceships trying to shoot at each other. Ammo and fuel limits plus an environmental hazard complicate matters.

Everything's played over a slowly moving starscape. For the 60s, this was an amazing feat of programming. Most of the time, the university's PDP-1 just sat around crunching numbers. But with the flip of a switch, a user could be transported to outer space. It's the kind of transformative experience that will really stick around with someone. At the same time that Bushnell was being exposed to possibly dangerous digital ideas, he was also doing something a lot more mundane. Like a lot of students, Bushnell had to work his way through college. He spent years employed at a nearby amusement park, eventually working his way up to managing the Midway section.

He was in charge of pinball machines, carny games, and all the various games and fun that you find at a local carnival. It's easy to imagine a young Nolan Bushnell pulling coins out of a pinball machine while just dreaming about the next game of space war. It was during this period of his life that Bushnell hit upon a really big idea that would change his future. Ever since he saw his friends zapping away at space war, he realized there was something bigger there. Sure, space war was a lot of fun, but there could also be a lot of money in it. There was a product here.

He knew consumers were happy to pay for entertainment, and space war was one of the most entertaining things he had ever seen. The issue came down to the market. Most folks, believe it or not, had no idea what a computer actually was. Even fewer had access to one of these digital behemoths. Most computers were in research labs, offices, government institutions. You know, very serious places with a lot of money. While users were happy to play a computer game during stolen minutes, it would be a really hard sell to get government contractors to buy a video game. But Bushnell saw a possible way around the problem. Coin-op machines. People would gladly drop some quarters for a game of pinball,

even though most players didn't own their own pinball machines. So why not up the ante and make a coin-op computer game? In 1971, Bushnell and Dabney founded a startup called Syzygy, which is honestly an awful name for a company. The duo built coin-op cabinets that played Space War, or at least something similar to Space War. The venture ultimately failed. They got a contract to build cabinets, but weren't able to move many units. So the next year, Bushnell and Dabney picked up the pieces of Syzygy and formed Atari. What's important about the Syzygy days is the type of hardware they built. Their space war clone, called Computer Space, well, it didn't actually use computer hardware per se. That was still beyond expensive, you can't pay for a mainframe with quarters.

But computer space was digital. The only analog parts were the drive circuits for an internal TV display. The guts of the game were built using transistor-transistor logic chips. Usually just called TTL, these amount to packings of standard logic circuits in compact chip form. So already, we're in a different realm than earlier games. Computer space wasn't programmed, its logic was hardwired into a circuit board, but it's a very different beast than things like the brown box or tennis for two. The initial plan for Atari was to do everything that Syzygy was attempting, just better. Bushnell and Dabney basically had a dry run before their main act. As Atari got running, they brought on new talent, and one of their first employees was Al Alcorn,

a graduate from UC Berkeley who studied electrical engineering and computer engineering. Berkeley who studied electrical engineering and computer engineering. So basically the right person for Atari's goals. To get Alcorn started, Bushnell assigned him a fake project. Basically an exercise to get Alcorn up to speed with game development. And this is really where Bushnell did something that I think is dangerous. In the spring of 1972, Bushnell had caught a demo of the Magnavox Odyssey. He signed a guestbook, sat down to a presentation, and then played himself some ping pong. Bushnell would later say that he thought the game was interesting, but underwhelming. When drafting up an initial bogus project for Alcorn, Bushnell decided to throw ping pong his way. Alcorn recalled it this way, quote,

It's the simplest game. One moving spots, two paddles, score digits. That was one of the harder things to actually make, to try to get those seven segment numerals, you know. And that was it. It was an idea, but he told me he had a contract from General Electric for a consumer product, which meant that it had to be, like, maybe 10 chips, and I was up to 70 chips. But he told me that to get me to work hard at it and do a good job. So I tried to make it a good game." And now we arrive at Pong itself. Working under this fake crunch and designing to fake specs, Alcorn built a

prototype for the game. It was really similar to the ping pong game on the brown box. The court was displayed from a top-down view. A line broke the screen in half. Rectangular paddles on both sides of the court were controlled via dials, and a square ball bounced back and forth, but from there we reach something different. Pong didn't use an English dial to curve the ball. Instead, the trajectory was dependent on which part of the paddle the ball hit. The center of the paddle bounced the ball straight forward, while the extreme edges caused the ball to bounce off at steeper angles. This was just one tweak that Alcorn made to the game to make it more interesting. He also added dead zones to the court.

That's a small border where the paddle couldn't quite reach. A savvy player could, at least in theory, aim for one of these boundaries and score an easy point. Finally, the ball would increase in speed the longer it was in play. Taken as a sum, this leads to a pretty compelling game. It's fast-paced, requires a little bit of dexterity, and it's the kind of game that you can actually improve through practice. But the largest visible change, and the one that Alcorn talked about agonizing over, was the score counter. Each side of the court had a number displayed in its upper corner. This started at zero, and every time a player missed the ball, their opponent's counter ticked up.

Earn enough points, and you win the game. It's a simple addition, but it meant that there was a definite end goal. You didn't have to keep track of who was winning, the game itself handled that logic. Pong's internals were also totally different than other electronic tennis games. Atari was a digital house, and Pong was a digital game. Everything was built using TTL chips, so it's still not a program, but it's definitely not analog either. When all was said and done, Alcorn had created an impressive video game. Sure, it wasn't for a real contract, but Al came through with something spectacular. Bushnell figured that it would really be a shame to let such a good game go to waste. So the prototype board was packed up in a cabinet. A coin mechanism and TV were added,

and the first Pong cabinet was installed in a nearby bar. Bushnell just wanted to see if Pong could actually be a viable product. After a few days, Alcorn got a call from the owner of the bar complaining that their cabinet had stopped working, so he headed over to see what the issue was. After cracking open the cabinet, the problem became clear. The machine had gotten jammed up with too many quarters. Pong was shaping up to be a huge success. From there, it was pretty easy sailing. By the end of 1972, the first commercial Pong cabinets were rolling out of Atari and into arcades across the country. It became Atari's first major success. Suddenly, the tiny startup was becoming a major player.

But the success would lead to a little bit of a problem. It would get them noticed by Magnavox and by Sanders Associates. The licensing agreement between Sanders and Magnavox had a bit of an interesting provision. Magnavox was supposed to handle sub-licensing the Brown Box patent, and that included pursuing those who were violating that patent. Magnavox had been a little lax in the second part of their duties. With Pong becoming popular and a handful of other clones showing up on the market, Sanders was getting pretty antsy, showing up on the market, Sanders was getting pretty antsy. So, they pushed Magnavox to start suing. The way Sanders saw it, they were being stolen from. Eventually, they got Magnavox to

go along with their idea. As Bear recalled, quote, They finally committed to dropping notices of infringement on a group of arcade game manufacturers. Sanders and Magnavox went after the first set of infringers. The lawsuits began and would go on for the better part of 15 years, end quote. One of the main targets was, naturally, Atari. Now, this is a case of a small company getting a little bit too close to the big leaguers. Sanders Associates was a defense contractor. They were used to keeping tight control over their technology. Every step of the development of the brown box and, by extension, the Odyssey, was very well documented and archived. The very logbook

that Bushnell signed to see a demo was in those archives. And that was a possible smoking gun. was in those archives, and that was a possible smoking gun. And as a much larger company, Sanders was able to do things in court that Atari just couldn't. This is where we run into the very unsatisfying field of patent law once again. The patent on the brown box, officially called Television Gaming and Training Apparatus, makes about 45 separate claims to new technology. None of those are on specific games that the console plays, but rather the video generation technology in use. That is, circuits for generating dots on a TV screen, circuits to control those dots, and all the support for those operations. What's important here is the patent is pretty clearly targeted at Bear's analog implementation. It was full of diagrams of analog circuits.

It covered moving dots around on a screen with analog components. And while it talked about dial and light gun inputs, it never outlined any of the actual games. The simple fact was that Atari wasn't using much of this technology inside Pong. Alcor was working with purely digital circuits. At most, maybe the controls were similar and a little bit of the RF signal circuits were doing similar things. But Pong was its own device, inspired by Bear's work, but not exactly a copy. If you want to get into a little callback, we could almost say that Pong was a clean room reimplementation. Al Alcorn hadn't seen Ping Pong on the Odyssey. Someone

else had. Bushnell and Alcorn, neither of them knew how the machine worked internally. Bushnell just told Alcorn, hey, make me some ping pong. Atari's lawyers were convinced they could win the case, but it was going to be an uphill battle. They were looking at a long and expensive process. Sanders and Magnavox had money to burn. They could file appeals. They could be in court as long as they wanted. Resources weren't an issue for them. Atari couldn't compete with that. So the case was settled out of court. Atari paid back licensing fees and signed an ongoing license agreement.

Of course, Atari survived this. Pong would make its way further into the market, and Sanders made some extra money from the trial. Alright, that does it for our dive into early ping pong games. I want to close out by coming back to what I started this episode with. Was Pong a ripoff? I started this episode with. Was Pong a ripoff? Was it just an imitation of earlier electrified tennis games? Sure, Bushnell was inspired by Bear's work. There's a paper trail and everything. But I don't think that's a very interesting answer. We can do a little bit better. One theme that kept coming up in my research was how simple tennis games are. When you get down to it, an electronic

ping pong game is probably the most simple game you can produce that's actually fun. You can make even more minimal games, but there's something about a bouncing ball that people just like. This is exactly why the game was invented twice. Higginbotham hit upon the idea as a fun demo, and then independently, Bear's team developed the idea as a product. Pong was definitely inspired by the Odyssey. There's no getting around that. But I'd wager that a third implementation could have happened on its own. Atari was working on a new way of creating games using new technology.

Ping Pong had already appeared twice before. I don't doubt that if things had gone differently, it could have appeared a third time. Thanks for listening to Advent of Computing. I'll be back soon with another piece of the story of computing's past. And hey, if you like the show, there are now a few ways you can support it. If you know someone else who's interested in the story of computing, then why not take a minute to share the show with them? You can also rate and review on Apple Podcasts or basically anywhere you listen to podcasts. And if you want to be a super fan, you can now support the show directly

through Adjunct of Computing merch or signing up as a patron on Patreon. Patrons get early access to episodes, polls for the direction of the show, and bonus content. Right now I have a poll up that should be closing soon for a February bonus episode, so if you want to get in on that, go sign up now. You can find links to everything on my website,

adventofcomputing.com. If you have any comments or suggestions for a future episode, then go ahead and shoot me a tweet. I'm at AdventOfComp on Twitter. And as always, have a great rest of your day.