The Age of Atari - the beginning

When Steve Russell and his friends completed the development of Spacewar, there was no market for video games. So, they allowed anybody to freely copy it. Furthermore, DEC used it as a testing tool in the production line of PDP-1 because it made use of a variety of functions of the machine. After checking the products, workers of DEC didn't bother to erase it. Hence, the game became a free gift for buyers.

One of those buyers was the University of Utah, where Nolan Bushnell, a co-founder of Atari, learned electrical engineering.

--

He was born in Utah in 1943. In his childhood, amateur radio captured his enthusiasm. He earned money to buy radio parts by fixing his neighbors' home electronics like TVs and washing machines. This experience helped him when he later made prototypes of arcade video games by himself.

After growing up, he entered the university in 1963. Nowadays, it is regarded as one of the top US universities in the field of computer science, especially computer graphics. But they didn't have a department for computer science at that time. So, it took a few years before he touched a PDP-1 there. However, during the days before then, he learned the basics of entertainment business from his working experience in a nearby amusement park.

During weekends and summer breaks, he worked as a game operator at Lagoon Amusement Park in Salt Lake City. Although there were no video games yet, he was able to learn how to induce people to spend money for amusements. In addition, as he was good at repairing machines, when some amusement machines broke down, other workers would call him for help. So, he was able to observe how those machines worked.

After having worked there for years, he was promoted to the manager of all games in the park and learned the business side of entertainment.

Nolan Bushnell (2009):

I was actually the manager of the games department of an amusement park when I was at college, so I understood the coin-op side of the games business very well.

-"Meet Nolan Bushnell, the man who created the videogames industry" on The Guardian
https://www.theguardian.com/technology/gamesblog/2009/mar/18/nolan-bushnell-interview-atari


Actually, he was not the only person who tried to make commercial video games. But among them, he was probably the only one who correctly understood both technical requirements and business aspects of video games.

--

Now let's back to the story of his academic life.

During the last half of the 1960's, the University of Utah hired several talented computer science researches including David Evans and Ivan Sutherland - the duo later became known as pioneers of modern CG technology - and allocated much money and research effort to the area. Although the newly created computer science department owned some computers, undergraduates like Bushnell were only allowed to use them during computer science classes.

Nevertheless, he visited the computer room almost daily and became friends with some of teaching assistants of computer science. Thus, he was granted access to the machines in off-peak hours.

There, he learned the basics of computer programming and played some of the earliest video games including Spacewar, which became his favorite. He and his friends also developed some video games, ranging from a primitive one like Tic-Tac-Toe to a little bit more sophisticated ones, but none of them were in the same league as the revolutionary shooter.

From his working experience in the amusement park, he knew what kinds of games would attract how many customers. He thought this exciting video game could attract large crowds if he could make it an arcade game. But he soon noticed that it was almost impossible to make any profit from the game because a PDP-1 cost around a million dollars (in today's dollars).

So, he had to dismiss the idea at the time. But the day to test his thoughts had arrived unexpectedly soon.

The video game which changed everything

Spacewar was developed by some members of MIT's Tech Model Railroad Club (TMRC). Despite its name, their main interest shifted from the hobby to digital computers after they found a TX-0 in the campus.

As I wrote in the last article, the machine was a successor of the Whirlwind computer, the world's first computer with real-time processing capability. It ended its role in the Lincoln Laboratory when a more powerful computer called TX-2 was built. So, the lab transferred the outdated machine to the university's campus, anticipating it to be used for research and education purposes.

The machine was managed by its Department of Electrical Engineering and Computer Science. Although MIT also owned some large computers like IBM 704, their use was strictly limited to serious research projects and ordinary students like the members of the club didn't have right to use them.

By contrast, the newly installed machine was managed under a much more open access policy. Moreover, the machine's administrator was an assistant professor named Jack Dennis, who was a former member of TMRC. So, they were allowed to use the machine unless it was used by researchers for important works. They were soon amazed by the advanced features of the machine and almost hooked on it.

Thus, those model railroad fans became probably the world's first computer geeks.

A few years later, they heard the news that an even more advanced machine, PDP-1, was coming to their campus. They were so thrilled that some of them soon began to make plans about what kinds of programs to write. Among them, Steve Russell and his friends made the most ambitious plan - they dreamed of making a video game in which you can experience a war in space.

Actually, even TX-0 had some entertaining programs like "Mouse in the Maze", "HAX" and, of course, a Tic-Tac-Toe. But they were developed only to demonstrate its high computing power and advanced features.

What the geeks of TMRC was going to do was the opposite. They used high computing power of PDP-1 to develop an entertaining program based on science fiction stories, which was their common hobby other than model railroads and computers. Especially, they loved "Lensman", a space opera fiction series written by E.E. Smith, and wanted to make a game where they can become a pilot of a spaceship.

That's why Spacewar became an innovative game. Unlike other "games" developed to demonstrate latest computers - and much similar to the case when William Higinbotham created the world's first video game, Tennis for Two - Steve Russell and his friends used the latest computer solely to entertain people including themselves.

I'll not write the details of the development history because it's so famous. Rather, I'd like to point out an important factor which enabled Spacewar to become a successful video game - MIT's open and cooperative culture.

Although it was long before the word "open-source" became commonly used by software developers, the source codes of the game were completely open to anyone. They were recorded on tapes, which were stored in drawers in their clubhouse. So, those who wanted to know its software architecture had no hurdles to access. And some of them in turn made great contributions.

It's a fact that Steve Russell constructed the basic structure of the game. But what really made the game a masterpiece were additional programs written by his friends. Without their contributions, the game would end up as just a technical demonstration of the latest machine, which in turn would delay the birth of the video game industry considering Nolan Bushnell, a co-founder of Atari, made video game his business because he saw a huge business opportunity in Spacewar.

Among them, the most important improvement was made by Dan Edwards. When he saw an early version of the game,  he complained to Russell that controlling a spaceship was just boring because its movement was too linear. So, he made a suggestion to apply some gravitational effects of the sun, but Russell dismissed it by saying the machine didn't have enough power for such calculation. Edwards was so confident in his idea that he decided to write the program by himself. When it was completed, Russell saw no value in it and only reluctantly merged it with his main program.

Steve Russell (1997) :

Dan Edwards' works added nothing other than more complexity to the existed program - it didn't improve the game.

Shin Denshi Rikkoku 4 (ISBN: 414080274X, p.60) NHK Publishing in Japanese, 

This comment is very interesting because it shows that even the lead programmer of this incredible game didn't understand what makes a video game fun. Contrary to his understanding, the gravitational effects is the key factor of its success because it introduced the concept of "tactics" into video games.

While it's true that the gravitational effects increased the difficulty of controlling a spaceship, mastering it is an enjoyable activity; skiing and skating also require some practice before you actually experience the fun part, but millions of people enjoy them.

And it also means that your opponent can't easily predict your moves, which increases your chance to beat more skilled players by chance.

Actually, you can make absolutely unpredictable moves in the game. When you are in a desperate situation, there's a last resort function for you called "hyperspace".

It was developed by another member of the club named J. M. Graetz. By using it, your spaceship will disappear and later reappear at a randomly chosen point in the game field. Of course, there's some limits. You can only use it a few times for a single play and your situation may not get better - your spaceship may reappear in front of the opponent or a point very close to the sun, which destroys nearby spaceships.

Although those two additional features were not very well-designed, they did add some tactical depth to the game, and people loved them.

However, in the case of video games, "deep" doesn't always mean good; it's a double-edged sword. That's the key to understand why Nolan Bushnell's attempt to make money from the game failed.


Primary reference:

"Spacewar and the Birth of Digital Game Culture" on Gamasutra
https://www.gamasutra.com/view/feature/129861/down_the_hyperspatial_tube_.php

The birth of the personal computer

Computers and electronic gadgets can be categorized in several different ways.  In a standard way, we categorize them like Supercomputers, Desktop PCs, Notebooks, Tablet PCs, Smartphones, Smart-watches, VR devices, Smart home devices and so on.

But ultimately, there's only two distinctive categories of computers: the ones meant to be operated by organizations vs. by individuals.

Needless to say, supercomputers belong to the former category and most of the others on the list above belong to the latter.

To be clear, if you are a business tycoon or someone like that, you might be able to own a latest supercomputer but will need specialists to "operate" it. And, of course, almost all large organizations in the world owns PCs but operating a personal computer doesn't require more than one person.

The important thing here is, those two categories represent not their capabilities but their suitable applications. Even in today, playing a FPS on a supercomputer is not a good idea - it's capable but not suitable. So, we needed some kind of a personal computer with real-time processing capability to develop and play high-quality video games.

--

The invention of the personal computer was not an evolution, but a revolution.

In the industrial revolution, we found a smart way to extend our physical abilities. Industrial machines powered by fossil fuels had multiplied the productivity of individual workers and new kinds of transportation, most notably steam locomotives, had reduced the time required to move goods and people around the world.

Similarly, personal computers extended our mental abilities - I think you don't need clues.

--

The history of the revolution began in 1955 at the same place where Whirlwind computer was developed - MIT's Lincoln Laboratory. Until then, practically all digital computers were supercomputers - they were meant to be operated by organizations like governments and giant companies.

The revolutionary, Ken Olsen, was a relatively junior member at the lab. Although he was not deeply involved in the development of Whirlwind, he was very much impressed by its high capability and potential.

His main role there was to evaluate usefulness of new technologies and develop experimental machines using them. One of those works, TX-0, changed his fortune and opened a new frontier in computing. As its extended name "Transistorized eXperimental computer zero" represents, it was the first transistor computer developed at the lab. Although it inherited the core architecture of Whirlwind series, which consisted of thousands of vacuum tubes, it was much smaller and required less maintenance work.

YouTube: TX-0 (MIT Lincoln Laboratory, 1955)

Back then, most research effort was focused on making more and more powerful computers. For example, IBM was also an early adapter of transistor technology and they used it to develop more powerful supercomputers (they preferred to call them "mainframes"). For them, it just meant that they could cram more electronic parts into the same space once occupied by large vacuum tubes.

But Olsen saw the technology from a completely different perspective. He thought he could make a more affordable and easy-to-operate computers by combining it with real-time processing capability and intuitive user interface of Whirlwind series.

To realize the vision, he decided to leave the Lincoln Laboratory for founding his own company, DEC.

--

Here, I assert that PDP-1, its first computer released in 1960 and the machine on which Spacewar ran, was the world's first "personal computer" in a practical sense.

Although it cost around a million dollars (in today's dollars) and you couldn't move it alone, it was meant to be operated by individual users and did provide features and I/O devices which later became the industry's standard.

For example, its users would write a program with a typewriter-like keyboard. Then, the machine would load the program and ran it in real-time. They could see the results on a CRT display and could control it with a keyboard or pointing devices. When they found a coding error, they could debug it on the machine itself.

All of those facts are hardly surprising because they're what we usually expect from modern PCs, which in turn supports my assertion.

Therefore, although DEC officially introduced PDP-1 as a "minicomputer", it was undoubtedly a forerunner of today's PCs.

YouTube: PDP-1 (Digital Equipment Corporation, 1960)

But like other innovative products, it didn't become popular overnight. First of all, there was no market for such kind of product. Moreover, it lacked a killer application software to prove the usefulness of real-time computing.

So, it was not just a coincidence that the innovative video game, Spacewar, was developed for the world's first personal computer - they needed each other.

To motivate and assist software developers, PDP-1 provided a simple and useful command set and various kinds of optional I/O devices including "light pens". Also, its hardware architecture was so simple that its users could connect homemade devices - some members of Spacewar team built a gamepad by themselves.

In short, it was a perfect toy for geeks who had enough time and passion to learn new technologies - MIT have never lacked such kind of people.

So, let's resume the development story of Spacewar.

Primary reference:

"Interview with Ken Olsen" on The National Museum of American History
http://americanhistory.si.edu/comphist/olsen.html

The beginning of real-time computing

One of the most essential technologies for video games, "real-time processing capability", was born from a fear during the Cold War era.

In 1949, The Soviet Union successfully carried out its first atomic bomb test. Combined with its capability to send bombers to the mainland US, the nuclear threat suddenly became serious.

Having recognized the shifting power balance in the Cold War, the Department of Defense (DoD) decided to replace their outdated air defense system. They feared hostile aircraft could sneak into the airspace of the US as their radar system's coverage area was not enough.

The most obvious way to improve its capability was to dramatically increase the number of radar sets. And since the ultimate objective of the air defense system was not just to detect but to intercept hostile aircraft before taking any military action, it was almost meaningless unless it could provide real-time airspace information.

Therefore, the system had to receive and process data coming from hundreds of radars without delay, but there were no computers which meet such high demand.

At that time, there was an associate professor of Physics at MIT named George Valley. He also had concern about Soviet nuclear strikes and had been making suggestions to the Air Force about improving air defense.

So, the DoD launched Air Defense Systems Engineering Committee and appointed him as its chairman. Then, he consulted several computer manufacturers about the possibility of realizing real-time computations, but none of them had a good solution.

Fortunately, he found the answer lying in his own campus. During the WWII, MIT was involved in various military research projects. Among them, a project named Whirlwind attracted his interest.

Whirlwind computer, a resulting product of the project had a massive computing power, which was enabled by very efficient and powerful memory units invented by one of its members. The project had been receiving funds from the Office of Naval Research and the machine was meant to be used for a flight simulator. But after years of delay of completion, the sponsor lost interest and decided to abandon it.

It was at just this time that Valley heard from one of his colleagues about its technological progressiveness. So, he decided to invite members of the project to join his new project and let them continue developing more fast and powerful computers.

Having found a promising computer technology, he was given a green light to advance the project to the implementation phase. Thus, the committee was dissolved and a large project currently known as Project Lincoln and the construction of MIT's Lincoln Laboratory was started. After just a few years, they hired thousands of talented researchers and engineers and the lab became an epicenter of innovation.

--

For example, the Internet has its origin here.

To gather information from radar sets installed in all parts of the nation, they required a reliable and efficient data transfer network. A centralized network which covers the entire nation is easy to build, but apparently vulnerable to destruction as its central hub becomes the single point of failure. Therefore, they divided the nation into 24 areas, each of which was managed by a direction center, and constructed a distributed network among them.

Although today's Internet was evolved from another military-funded network called ARPANET, its concept was thus born in the Lincoln Laboratory.

--

Anyway, having received abundant funds, the Whirlwind computer was much improved in speed.

By 1951, it had evolved enough to process incoming data in real-time, so they carried out their first live test of an interception.

It went perfect. The machine was able to process data without delay and the location of the aircraft which acted as an enemy was successfully transmitted to the interceptor aircraft.

So, this is the beginning of real-time computing.

Eventually, this technology became the driving engine behind video games but it was not a foreseeable future for people back then.

AN/FSQ-7, the deployed version of Whirlwind, not only cost an enormous amount of money, but also required a dedicated building with expensive cooling equipment to prevent it from overheating. Besides, since it was composed of thousands of very fragile and unreliable vacuum tubes, specially trained maintenance workers had to be standing by 24/7.

Therefore, the computer itself had little impact outside the military, but it indirectly caused the creation of two distinct categories of computers, which have been affecting our lives in many ways since then.



Primary references:

"History" on MIT Lincoln Laboratory
http://www.ll.mit.edu/about/History/history.html

YouTube: Military promotional video of "SAGE" air defense system