They Create Worlds

13
Homeward Bound

On May 21, 1954, Texas Instruments executive vice president Patrick Haggerty approached electrical engineer Paul Davis with a challenge: develop a portable radio using germanium transistors in lieu of vacuum tubes. At the time, the transistor was struggling to replace the venerable tube in most electronic devices, and Haggerty reasoned that so long as it only served industrial and military applications, it would never achieve high-volume production. That required a consumer product, and no electronic consumer device was more widespread than the radio.¹

Four days after Haggerty approached him, Davis had a working prototype. As Davis’s team continued to refine the design over the next few months, Haggerty attempted to locate a radio company interested in manufacturing his transistor radio but was turned down by all of them. Ultimately, TI turned to a company called Industrial Development Engineering Associates (IDEA) that marketed antenna boosters under the Regency brand name. In November 1954, IDEA released the Regency TR-1, the world’s first transistor radio and the first consumer product of any kind based on solid-state technology. Although it never worked well and cost more than a tube radio of similar size and capability, the TR-1 sold 100,000 units within a year based on its novelty alone, while TI produced half as many transistors just for the radio as the entire industry had produced for all applications the year before. Although TI never earned much profit through transistor radios, the TR-1 demonstrated one important fact: the best way to spread and improve solid-state technology rapidly was to pair it with a consumer application.²

In the early 1960s, TI faced a similar problem with integrated circuits, which were not being embraced despite their obvious superiority over discrete electronics. Once again, Haggerty decided to jump-start the new technology through a consumer product, so in 1965 he tasked integrated circuit co-inventor Jack Kilby with creating a handheld calculator. By 1967, Kilby’s team had delivered a prototype called the Cal-Tech that contained all its functionality on just four ICs. Three years later, TI partnered with Japanese manufacturer Canon to release an improved version of the Cal-Tech as the Pocketronic.³ A year after that, a TI spinoff called Mostek succeeded in creating a pocket calculator using only a single chip, which was debuted by another Japanese company called Busicom.⁴

Unlike the transistor radios that were initially more expensive and less capable than the technology they were attempting to supplant, the pocket calculator was far smaller and cheaper than existing designs that provided the same functionality and was immediately successful. By 1973, 7 million handheld calculators were being sold a year.⁵ The first solid-state consumer electronics boom had begun. It ended nearly as quickly. In 1972, Texas Instruments broke with convention by not just suppling ICs to other companies, but actually manufacturing calculators itself. As a vertically integrated company, TI could undercut its competitors on price and touched off a ruinous price war over the next two years that cut into profitability for everyone and drove several high-profile companies out of the business entirely. By 1974, calculators that had cost $300 or more just three years prior could be purchased for only $19.95.⁶ The pocket calculator was here to stay, but it would never again be such a highly profitable business.

While the calculator boom ended poorly for most of the companies involved, it exerted an important lasting effect on the semiconductor industry. The high-volume production of ICs to meet calculator demand drove down the price of the devices dramatically and led to rapid advancement in large-scale integration (LSI) circuits, the first ICs with enough power to be useful in a wide array of consumer products at a competitive price. The boom also awoke the solid-state industry to the value of designing ICs specifically for consumer applications and to the desirability of entering the consumer electronics market directly under the right circumstances. Indeed, by the time the calculator market had run its course, the major semiconductor factories were already looking for the next big thing in consumer electronics, and many of them settled on the home video game.

***

In September 1972, Magnavox released the Odyssey in the United States, which proved a sales disaster when the company only sold 69,000 of the roughly 140,000 units produced. Magnavox nearly pulled the product but relented in the face of several positive signs. Magnavox had included a customer survey card with each Odyssey and held back a game called Percepts as an exclusive for anyone who returned the card as an incentive to fill it out.⁷ When these customer surveys began pouring in, Magnavox discovered most of their customers were happy with their purchase.⁸ Magnavox dealers were also pleased because they were able to sell through much of what they actually ordered and were interested in stocking the system again for Christmas 1973. Magnavox started a new production run on the system to build an additional 27,000 units before the end of 1973 and began offering the console at half price if purchased in tandem with a new Magnavox television.⁹ The product sold 89,000 units in 1973, which largely cleared out the unsold inventory.¹⁰

While Magnavox was happy to maintain the status quo, the company held no interest in expanding its video game product line further. In 1973, Bob Fritsche and his Odyssey marketing team envisioned a “lite” version of the console with only five games and a deluxe version with four controllers as well as a dozen or so new and updated games. In the end, no new console variations were released and only four new games came to market: Interplanetary Voyage, Basketball, Brainwave, and WIN. All four were designed by Odyssey assistant product planner Don Emry, though WIN and Brainwave were started by Bradford/Cout, the contractors that developed the initial set of games. They all used existing circuit cards save Interplanetary Voyage, which used a new card that added momentum to the spot. These were the last new Odyssey games released.¹¹

Meanwhile, Magnavox itself was falling apart. The company’s failure to transition to solid-state television technology fast enough found it losing market share by the start of the 1970s. Furthermore, a Federal Trade Commission (FTC) consent decree in 1971 forced the company to end price maintenance policies in non-fair trade states and caused price erosion in most of the United States that destroyed the company’s margins.¹² In late 1972, Magnavox president Robert Platt forced Consumer Electronics Group president George Fezell into retirement and replaced him with Alfred Di Scipio from the Singer Company, who was known to be a savvy marketer.¹³ Di Scipio then cleaned out most of the division’s senior executive staff including Gerry Martin, the vice president who had authorized the Odyssey project back in 1971. As the company’s difficulties continued, Magnavox became a subsidiary of Dutch electronics giant Philips in September 1974.¹⁴ Odyssey was quite simply lost in all the turmoil, and Fritsche proved unable to convince his new superiors that the product was worth supporting when sales were so far below those of the company’s televisions.

Nonetheless, Di Scipio’s new marketing approach benefited the Odyssey. In late 1973, Magnavox sponsored a Frank Sinatra television special as a springboard to advertise its complete line of products and subsequently hired baseball player Hank Aaron, one of the most recognizable men in America as he prepared to break Babe Ruth’s career home run record, as a company spokesman.¹⁵ The increased exposure and publicity the Odyssey experienced through these marketing campaigns helped fuel a banner sales year in 1974 during which the company sold 129,000 units.¹⁶

In 1972, the production cost of a single Odyssey unit came to $37.00, of which roughly $5.00 represented the cost of the overlays and additional game accessories, and the rest was the cost of the electronics. Magnavox sold the system on to dealers for $65.00, who then marked it up to $99.95.¹⁷ By 1975, rising inflation in the United States had increased the production cost of the Odyssey to $47.00, but the company could not afford to raise the retail price.¹⁸ Due to the risk of the system becoming unprofitable, Magnavox began looking for cheaper parts alternatives and ultimately signed a contract with Texas Instruments in May 1974 to provide four MSI ICs to replicate the functionality of the dozens of discrete transistors and diodes in the original system.¹⁹ Recognizing that the ball-and-paddle games that had since taken the arcade by storm were far and away the most popular games on the Odyssey, Fritsche and his team dropped the majority of the games on the system to focus solely on table tennis, hockey, and handball.²⁰

Magnavox ceased production of the original Odyssey in fall 1975 and sold another 80,000 units for the year,²¹ bringing total sales over five years to roughly 367,000 units. At the same time, the company introduced its new IC-based system in two configurations, the Odyssey 100 and the Odyssey 200. Both systems consisted of a single unit that attached directly to the television with no separate circuit cards or controllers.²² The Odyssey 200, which retailed for $109.95, was the complete version of the system and shipped with three games, Tennis, Hockey, and Smash, the last of which was a renamed version of Handball. Unlike the original Odyssey, it also featured sound and a primitive form of on-screen scoring in which a rectangle advanced across the screen when a player scored a point. The Odyssey 100 was designed to prevent Magnavox being undercut in the market on price and omitted both the Smash game and on-screen scoring so it could retail for only $69.95.²³ Despite a late delivery on the ICs from TI that pushed the introduction of both systems back until November,²⁴ Magnavox sold over 100,000 units between them to stay on top of a suddenly competitive marketplace.²⁵ Total sales of the Odyssey 100 and 200 systems over their lifespans were just under 100,000 and 200,000 units, respectively.²⁶

For three years, Magnavox essentially had the home video game market to itself. Only one other company, Paddle Battle creator URL, marketed a home system in the United States in 1974, and that was merely to clear out unused game components after the collapse of the ball-and-paddle market in the arcades. Called Video Action, URL’s game played the three most common arcade ball-and-paddle variants, Tennis, Soccer, and Hockey. The inclusion of a 12-inch black-and-white television with the system pushed the price to a decidedly non-consumer-friendly $499, and it sold poorly. URL tried again with the Video Action II sans TV in 1975, but the system still retailed for a cost prohibitive $299.²⁷

In 1975, two new competitors emerged from an unlikely source: the novelty product business. Over the previous year, several companies in this field had experienced success bringing popular coin-operated products like pinball and air hockey into the home, and now they were ready to do the same with video games. Executive Games, established in 1968 to create novelty versions of board games like chess and backgammon, decided to capitalize on the burgeoning popularity of both Air Hockey and Pong by producing a limited run of home versions of both games in 1975.²⁸ The video game was developed by a group of MIT students who were part of an initiative called the MIT Innovation Center and was released as Television Tennis in November.²⁹

In December, Executive was joined in the market by former Nashville firefighter Norvell Olive, who in 1971 established a company called General Advertising Corporation to sell novelties like key rings through bank credit card system. To enter the video game market, he established a company called First Dimension in 1975 and contracted a Massachusetts firm to build a ball-and-paddle game called the FD-3000W.³⁰ Both Executive and First Dimension kept their production runs small, but the video game market ended up growing well beyond their expectations due to the introduction of a new system by one of the leading companies in coin-operated video games, Atari.

***

In late 1973, Atari hired a man named Harold Lee as a production engineer, a role in which he took wire-wrapped prototypes coming out of Cyan in Grass Valley and turned them into production model PC boards that could be wired to the various other components in the arcade cabinet. Before coming to Atari, Lee had worked for Standard Microsystems, which like so many technology companies entered the calculator chip business in the early 1970s. Lee personally designed 12 chips at the company, several of which entered production.³¹

After about a year at Atari, Lee was burned out on arcade games and decided to quit.³² Due to his chip experience, Al Alcorn asked him to continue working for the company as an independent contractor and to deliver a custom sync chip to help combat the rampant copying of Atari products by its competitors.³³ Lee decided this project was a bad idea, for by the time he completed the nine-month development cycle on the chip, Atari would have updated its technology, and the chip would be useless.³⁴

While mulling over the sync chip, Lee turned to a friend from his Standard Microsystem days named Bob Brown.³⁵ An electrical engineer with a PhD from Stanford, Brown worked on speech recognition for five years at Rockwell before joining Fairchild Semiconductor to lay out ICs using new computer-aided design (CAD) techniques. Brown next worked on high-speed modems with a startup called Modex that was subsequently acquired by Standard Microsystems, where Brown managed CAD and IC testing.³⁶ At the time Lee approached him, Brown had recently become fascinated by a product called the Go Scope that generated colorful patterns on a TV when music was played into it. Using a Go Scope required a state-of-the-art hi-fi TV system that retailed for around $1,000, so Brown became interested in producing something cheaper that could be hooked into a regular television set. Combined with his work on the sync chip with Lee, this led him to ponder if Pong could be placed on a single chip.³⁷ Brown brought his idea to Lee, who decided it could be done and offered it to Alcorn in place of the sync chip.³⁸ Alcorn pitched Bushnell, who had indicated as far back as August 1973 that he wanted Atari to enter the home market and felt pressure to diversify after the difficult 1974 fiscal year, and the project received the green light.³⁹

Lee set to work on the chip in late 1974 under the umbrella of his contracting firm, MOS Sorcery, which was based in a cabin on a Christmas tree farm he had recently purchased in the hills outside Los Gatos. Over the next few months, Lee diagramed the chip, Alcorn’s wife Katie wire wrapped it, and Alcorn debugged it and sent corrections to Lee. Once they had a working chip prototype, Lee rented time on a CAD system in East Palo Alto and spent his nights – rent being cheaper than during the day – laying out the chip with the help of Bob Brown, whom Alcorn hired into Atari from GTE Sylvania to write the testing software necessary to make sure the chips were functioning properly. By July 1975, the group had completed the chip, so Atari turned its attention to locating a manufacturer. Four local semiconductor companies were approached, and Atari chose American Micro Systems, Inc. (AMI), a spin-off from electronics giant Ford-Philco established in 1966 that specialized in creating custom circuits for outside clients.⁴⁰

As Lee and Brown continued designing the final chip, Atari brought a wire-wrapped prototype of their Home Pong unit to the American Toy Fair in January 1975. Sponsored by the Toy Industry Association since 1903, Toy Fair was the premiere trade show of the U.S. toy industry and attracted thousands of buyers to the show rooms in the Toy Center on New York City’s Fifth Avenue to examine the toy lines planned for the coming year and decide what and how much their stores should buy. Atari set up a small booth on the main floor of the show but failed to entice any toy buyers with a suggested retail price of $99.95 that was unpalatable to an industry that never sold a product for over $30 other than bicycles.⁴¹ The only company that showed any interest was Tandy, proprietors of the Radio Shack chain of electronics stores, but when the buyer asked for a standard term in the toy industry called an “anticipation discount,” Atari’s management team, which had no experience with the customs and mores of the industry, turned the offer down.⁴²

With the toy industry uninterested in Home Pong, Atari turned in desperation to Sears Roebuck, the largest retailer in the United States. Atari cold-called the company’s television department at the Sears Tower only to be turned down. The television buyer did, however, connect Atari to another person at the company who had previously expressed interest in home video games, a sporting goods buyer named Tom Quinn.⁴³ In the winter months, the sporting goods department largely became a purveyor of indoor pastimes like table tennis and pool, and Quinn had decided that digital depictions of sports like Table Tennis and Volleyball made a video game console a natural part of the family rec room.⁴⁴ In 1974, he convinced Magnavox to let him sell the Odyssey through the vaunted Sears Catalog, but the company refused to let Sears sell the system in its stores due to its policy of dealer exclusivity. As a result, Quinn was perhaps the only retail buyer in the country actively searching for new video game product. Quinn invited Atari to demonstrate the game for the head of his department at the Sears Tower. The demo almost ended in disaster when the antenna on top of the building interfered with the signal from the console and forced Alcorn to undertake a quick modification.⁴⁵ Despite this hurdle, Atari and Sears signed a deal on March 17, 1975, in which Atari would provide Home Pong to Sears for the 1975 holiday season.⁴⁶

Manufacturing a consumer product on such a large scale would require a new facility and a large influx of capital. The facility proved relatively easy: in spring 1975 Atari and Kee Games consolidated their operations into a new 65,000 square foot facility on Martin Avenue in Santa Clara, California, which left the old Kee Games factory idle. This facility was transformed into a manufacturing plant for Home Pong.⁴⁷ Money proved harder to come by. Despite having a purchase order from Sears to use as collateral, banks refused to lend Atari the money it required to establish a production line due to its small capital base. Atari needed an infusion of cash quickly and turned to venture capital to provide it by approaching one of the newest investors in Silicon Valley, Don Valentine.

Born in Yonkers, New York, Valentine studied chemistry at Fordham University and spent time in the Army as an electronics instructor before transferring to a naval base in California, which fueled a desire to spend the rest of his life in the state. After leaving the military, Valentine worked for electronics giant Sylvania from 1957 to 1960 – interrupted by a brief stint at Raytheon – before joining Fairchild Semiconductor as a salesman in the Los Angeles area. After increasing sales significantly in his territory while simultaneously taking courses at the UCLA business school, Valentine became sales manager for the company in 1962.⁴⁸

In 1967, Valentine followed Fairchild general manager Charlie Sporck to a struggling company called National Semiconductor. Established in Connecticut in 1959 by former Sperry Rand employee Bernard Rothlein, National was a small, but generally successful semiconductor firm until a patent lawsuit filed by Sperry Rand depressed the stock price in the mid-1960s. Investor Peter Sprague took advantage of this situation to purchase a large stake in the company and become chairman of the board in 1966. Sprague wanted National to become a major player in the semiconductor industry, and he hired Sporck to transform his vision into reality. A manufacturing guru crucial to transitioning Fairchild from small-scale defense contract work to mass production for the consumer market, Sporck moved National to Santa Clara in 1968 and commenced high-volume production of linear and TTL circuits at a low cost. In the process, he instigated a price war that drove many companies out of the business and left National one of the largest semiconductor manufacturers in the world. Valentine led the company’s sales force while beginning to make private investments on the side.⁴⁹

In 1972, Los Angeles-based mutual fund company Capital Research and Management Corporation approached Valentine to manage a venture fund. With Capital’s backing, Valentine left National to form Sequoia Capital. He spent the next year and a half raising money, then spent that much time again looking for suitable investments. He was determined to invest only in Northern California high-technology companies, which limited his options, but he eventually settled on Atari.⁵⁰ Although initially repulsed by its coin-op business, which continued to be associated with organized crime in the minds of many people, he loved the home business, which would be wholly dependent on the semiconductors Valentine felt would be the future of all technology businesses.⁵¹

Valentine invested $600,000 in Atari in the summer of 1975, solicited a matching contribution from Time Inc. and the Mayfield Fund, and pulled in an additional $300,000 from the Boston-based firm Fidelity Venture Associates. This gave Atari a capital base of $4.5 million and allowed the company to secure a $10 million line of credit to put Home Pong into production. Marketed by Sears under the “Tele-Games” label, the system incorporated the most complex LSI yet designed for a consumer product and featured both on-screen scoring and full-color graphics to provide a technological edge over competing systems from Magnavox and the novelty manufacturers. The combination of America’s largest retailer and the most recognizable name in video games proved irresistible to the public, and Sears was swamped with more orders than it could fill as customers waited in long lines during the holiday season for the chance to put their name on a list that would guarantee them a system. Sears hoped to sell as many as 200,000 units before the end of the year Atari was only able to supply 85,000, which quickly sold out.⁵²

In 1975, roughly 310,000 home video games were sold in the United States, and most of them were dedicated Pong systems.⁵³ Magnavox sold the greatest number of units, but Atari generated much of the consumer excitement with its technologically superior system. The market was incredibly supply constrained, so even small companies like Executive Games and First Dimension that only managed to sell a few thousand units between them took back orders for hundreds of thousands more.⁵⁴ With video games generating so much excitement from both retailers and consumers, analysts predicted a big year for the category in 1976.

Despite the apparent popularity of home video games, however, the toy industry continued to shy away from the category due to the high cost of the products. Instead, a host of consumer electronics companies that had profited from the calculator boom by importing cheap models from Taiwan and Hong Kong rushed to fill the market. These companies did not have chip design expertise, so they needed to partner with existing semiconductor companies to create their games. The early entrants into the market like Magnavox and First Dimension relied on discrete components and MSI circuits to power their products, but once Atari deployed a console built around an LSI that offered superior gameplay at a competitive price, any company that hoped to remain viable in the rapidly expanding market would need to follow suit.

National Semiconductor was the first chip company to explore an LSI for video games. Like its main competitor in circuits, TI, National moved aggressively into building its own calculators in the early 1970s by establishing the Novus consumer products division in 1971. To run the division, Charlie Sporck tapped a salesman with deep experience in the office equipment and electronics businesses named Gene Landrum. Under Landrum’s watch, National became the leading producer of calculators in the world for a brief time,⁵⁵ but it ultimately scaled back its consumer efforts when that market fell apart. Landrum departed in September 1975, but before he left, National made a bid to win the Magnavox game business with a single-chip solution superior to the multi-chip system proposed by TI in 1974. Magnavox rejected the proposal due to the larger upfront cost of the chip,⁵⁶ so National demonstrated its technology at the same 1975 Toy Fair at which Atari debuted Home Pong.⁵⁷ Like Atari, it found no takers.

National had initially planned to mirror the original Odyssey by using plug-in cards to allow new game designs for its basic system, but with the success of Home Pong and the scaled down Odyssey systems in 1975, the company chose to refocus its efforts on a dedicated unit.⁵⁸ To that end, National engineers designed a chip called the MM-57100N that played the same three games as the Odyssey 200, Tennis, Hockey, and Squash, but with eight-color graphics and on-screen scoring. The plan was to make the chip available to any interested company, but National also released a system called the Adversary to showcase the chip that proved a modest success in 1976 with sales of 200,000 units.⁵⁹ The firm began marketing the MN-57100N in July 1976,⁶⁰ but it was not successful because by then the home game market was dominated by a different chip originally developed in Europe.

***

As in the United States, the Magnavox Odyssey was the first home system available in Europe. It was first released via a small test market in Germany in 1973 through ITT Schaub-Lorenz before becoming widely available the next year.⁶¹ For the general European release, Magnavox altered the game mix by removing several less popular games or games that would not work as well for an international audience and adding a few games that were sold separately in the United States.⁶² While it was well received in certain markets, it ultimately failed to have much impact.

As arcade video games continued to experience popularity into 1974, the European markets birthed their own home video game manufacturers, but adoption of the new technology proved uneven. While Europe is often treated as a single entity when discussed in passing, each country is its own market with differing regulations and tastes, which makes introducing a new product like the video game difficult. These problems were further complicated in the mid-1970s by inadequate distribution systems because most Western European countries still relied upon small mom-and-pop businesses to drive their economies, and achieving significant market penetration could be extremely difficult. Only the United Kingdom and West Germany were capable of mass market distribution, so these two nations became the largest European markets for video games. France, the third largest market, was hindered by fragmented distribution, while sales in southern European countries like Spain and Italy were hampered by significantly lower television adoption rates than Western and Northern Europe.⁶³ Most systems distributed in the European countries were manufactured or marketed from Britain, Germany, or the Netherlands, home to Europe’s most important consumer electronics firm, Philips.

The first European-designed video game system was introduced in 1974 by the British company Videomaster established by Cameron MacSween and Richard Fairhurst. A natural marketer, MacSween started in the food industry before switching to hiring out models for promotional purposes. He later teamed with Fairhurst, a market researcher for the Milk Marketing Board, to promote Barclays travelers checks in the United States. The duo beheld their first video game when they ducked into a seaside arcade in 1973 to escape the rain. Impressed by the amount of money the machine appeared to be earning, they began installing the machines themselves and became one of the largest operators of video games in British pubs.⁶⁴

Later in 1973, an engineer named Robert Palmer who had built a ball-and-paddle game that could plug into a television approached the duo. Although they were unimpressed at first, Palmer eventually won them over and joined the company as technical director in 1974. That same year, Videomaster released Palmer’s game as the Videomaster Home TV Game, which played Tennis, Football, and Squash games similar to those found on the Odyssey and retailed for £70. It sold roughly 4,000 units.⁶⁵

In 1975, Videomaster released four more systems that played varying combinations of ball-and-paddle games. The same year, a Cologne-based electronic manufacturing company founded in 1962 by Hans-Herbert and Hellmuth Türk called Interton released a system called the Interton Video 2000 that played ball-and-paddle games called Sparring, Badminton, Tennis, and Super Tennis and a spot-chasing game called Attacke [sic]. Like the Magnavox Odyssey, these games came packaged on discrete circuit cards, though unlike that system certain of these cartridges added graphical elements not found in the base system. While neither company sold more than a few thousand units, their entry into the market signaled the beginning of a small cottage industry of European video game manufacturing that birthed the most significant product of the dedicated console era, a ball-and-paddle LSI developed by the Scottish branch of an American electronics company called General Instrument (GI).

Established in New York in 1923, GI remained a small electronics manufacturer until the 1950s, when it began a massive acquisition spree to become a major player in a wide variety of electronic fields, including transistors. The company’s main business consisted of selling electronic components to manufacturers of consumer electronics like radios and televisions, so when managers at GI’s lab in Glenrothes, Scotland, learned in 1974 about the ball-and-paddle video games gaining popularity in American bars, they were intrigued.⁶⁶

Management asked an engineer who had previously worked on television remote control chips named Gilbert Duncan Harrower if it might be possible to design a chip to play a tennis video game on a television. Harrower thought it was and agreed to work on the project on his own time because GI was not willing to invest any money. Once Harrower and his assistant Dave Coutts had a basic prototype running, GI embraced it as an official company project and put out feelers to customers about additional features.⁶⁷ These ideas were incorporated into the finished product, designated the AY-3-8500, which could play six games: four ball-and-paddle variants and two target shooting games similar to the basic rifle games found on the Magnavox Odyssey. The chip could only output black-and-white graphics, but it sold for the incredibly low price of $5.00 per chip if purchased in bulk. GI was in the process of adapting the chip to work with American televisions when it was approached by one of the few toy companies that were looking to enter the video game business, Coleco Industries.

Coleco founder Maurice Greenberg immigrated to the United States from Russia in 1911 when he was just a boy and worked for his older brother’s moving company in New Haven, Connecticut. Eventually, he transitioned into the shoe findings business before moving to Hartford in 1932 and establishing the Connecticut Leather Company to supply leather products to shoe manufacturers.⁶⁸ While the company served primarily as a distributor, in 1944 Maurice’s 16-year-old son Leonard convinced his father to let him open a small shop to sell handcrafted leather items. After graduating from Trinity College in 1948 with a degree in mathematics, Leonard took a job as an engineer at United Aircraft in Hartford, but quickly grew bored and joined his father’s business full time in 1949. The next year, he convinced his college friend and fellow engineer Melvin Gershman to join him at the company to help fashion and operate a leather-cutting machine to produce spools of leather lacing. This initiative brought the Greenbergs into the manufacturing business.⁶⁹

In the early 1950s, the Connecticut Leather Company took over an abandoned glove factory in Mayfield, New York, to manufacture leather moccasin kits. When the kits proved extremely popular at the 1954 Toy Fair, the company committed fully to the toy business by offering a variety of leather kits in the late 1950s backed by major licenses including Davy Crockett, Howdy Doody, and Mickey Mouse. In 1956, Leonard expanded the business into small plastic toys and then into plastic wading pools. The pools proved immensely successful, so the Greenbergs sold off their leather supply and manufacturing departments in 1961 and renamed their company Coleco Industries. The next year, Coleco went public on the American Stock Exchange, and Maurice ceded the president and CEO roles to Leonard, retaining only the title of chairman for himself.⁷⁰

Over the next four years, Leonard Greenberg expanded the company aggressively through several acquisitions, including inflatable backyard pool maker Kestral Corporation in 1963, leading doll carriage producer Playtime Products in 1965, and Canadian tabletop hockey and football game maker Eagle Toys in 1968.⁷¹ During this process, Leonard leaned heavily on his younger brother Arnold, a 1958 graduate of Harvard Law School who worked at the law firm that had represented Coleco since its founding in 1932. As Coleco began to eat up more and more of Arnold’s time, he decided to join the family business himself in 1966. Officially appointed Coleco’s chief legal counsel, he immersed himself in all aspects of the business and discovered a knack for marketing, rising to the rank of executive vice president in 1970.⁷² In the late 1960s, Coleco became the largest maker of aboveground swimming pools in the world, enjoyed record earnings, and listed on the New York Stock Exchange in 1971.⁷³

Although Coleco enjoyed great success in the 1960s and experienced 12 consecutive years of record growth through 1972, the Greenbergs worried they remained too reliant on swimming pools, which as late as 1968 remained 75% of the company’s sales. An acquisition spree of roughly 20 companies followed over the next five years that Coleco largely financed with short-term debt. Integrating the companies efficiently proved impossible, however, and management found itself stretched too thin.⁷⁴ The final straw came in 1973 when a snowmobile company purchased the year before proved unprofitable during a mild winter and an attempt to move into dirt bikes failed. The company posted a loss of $1.1 million for the year, the first loss in company history.⁷⁵ An executive reshuffle followed as Maurice retired, Leonard replaced him as chairman while retaining the CEO position, and Arnold took over Leonard’s former role of president.⁷⁶

After two years of retrenchment, Coleco prepared to expand again and diversify into categories that sold well during the Christmas shopping season to balance its outdoor line that heavily skewed toward the spring and summer. Arnold Greenberg took note of the growing consumer electronics field and the emerging video game business and believed that if Coleco could produce a system for around $50, it might appeal to toy industry buyers.⁷⁷ The company employed no electrical engineers capable of designing such a game, so Coleco’s head of product development, Bert Reiner, formed an alliance with a small Connecticut company called Alpex Computer Corporation to develop a video game hardware system.⁷⁸

When it came time to engage a computer chip company to develop the heart of its new video game, Coleco was naturally drawn to GI because it was one of the few companies located on the East Coast rather than in Silicon Valley. Reiner arranged a meeting with representatives of the chip company, who revealed partway through his presentation that they were already developing a Pong chip of their own. Alpex cautioned against a GI partnership due to the company’s generally poor reputation, but the AY-3-8500 sealed the deal.⁷⁹ Alpex and Coleco developed a system around the chip called the Telstar – named after the famed communication satellite launched in 1962⁸⁰ – which included only three ball-and-paddle variants out of the six games included on the chip to keep costs down so the system could retail for roughly $70.00. Coleco planned to release the system in June to coincide with Father’s Day, but these plans were almost wrecked due to problems with the Federal Communications Commission (FCC).

As part of its mission to regulate the airways in the United States, the FCC requires that any device broadcasting on a radio frequency not interfere with other devices as outlined in the Federal Communications Act. Because early video game systems used an RF modulator to broadcast a video signal to a television set, they fell under the category of devices required to comply with the act and had to be tested by the FCC to ensure compliance. For most of the toy and consumer electronics companies now clamoring to enter the video game business, this was unfamiliar territory that involved navigating confusing interference standards and investing in an expensive electromagnetic interference (EMI) laboratory so as not to risk a failed test.

Coleco had neither the expertise nor the facilities to ensure the Telstar met the FCC standards, and the unit consequently failed its test. The FCC told Coleco it could resubmit the product at the end of the week for a second round of testing, but if the Telstar failed again, it would move to the back of the line of the dozens of systems awaiting approval and Coleco would miss its launch window. Desperate to release on time, Arnold Greenberg turned to Ralph Baer, who had started a side business of consulting on electronic game designs.⁸¹

Baer had access to an EMI lab within Sanders Associates and was able to isolate and correct the interference problem so that the Telstar could pass its second round of FCC testing and release on schedule.⁸² Even better, GI experienced such huge demand for the AY-3-8500 that it proved unable to fill all its orders in a timely fashion. Because Coleco had ordered its chips first, however, it received a full allotment.⁸³ Releasing into a wide-open market, the Telstar rocketed Coleco to the top of the nascent home video game industry as the company sold just under 1 million systems before the end of the year.⁸⁴

Although Coleco was widely acknowledged as the market leader in the United States in 1976, Atari may have actually come out on top, as the company claimed it sold over a million units combined of several console models. These were now marketed directly by Atari’s new consumer division in addition to its continuing relationship with Sears.⁸⁵ The main product offered by the company was an upgraded version of its original system called Super Pong that included full-color versions of both Pong and the Atari arcade game Super Pong and two original variants developed by Harold Lee.⁸⁶ A second console, Super Pong 10, added four-player variants of the games found in the Super Pong unit as well as a Handball game similar to the Smash game in the Odyssey 200.

Meanwhile, previous market leader Magnavox concentrated on refining the systems it released in 1975 with new low-end and high-end products. The Odyssey 400 contained all the functionality of the 200 model but with the addition of full on-screen scoring, while the Odyssey 500 sported full-color graphics and replaced the rectangular paddles with stick figure representations of people holding tennis rackets and hockey sticks. The two new systems retailed for $100 and $130, respectively. The company also embraced the AY-3-8500 through a system called the Odyssey 300. In addition to producing this console itself, the company made it available to other organizations through its Sentinel private label division.⁸⁷ After enjoying a relative lack of competition over the previous four years, Magnavox had to settle for third place after moving around half a million units.⁸⁸

Behind the market leaders lurked another dozen or so companies that experienced varying degrees of success. The most prominent of these firms was a New-York-based importer of Asian products called APF Electronics established by brothers and veteran Asian importers Albert and Philip Friedman in 1968.⁸⁹ The company went public in 1972 after riding the calculator boom and jumped into video games in 1976 with an AY-3-8500-based system called the TV Fun that retailed for $90.00 and moved 400,000 units.⁹⁰

On the other end of the spectrum was Allied Leisure, which sought to combat flagging sales of its arcade games by releasing a home system in 1976 called Name of the Game powered by a chip from GI competitor MOS Technology that could play four ball-and-paddle games and shipped in both a two-player and a four-player configuration. With video games in such high demand, Allied secured orders for 60,000 units, but difficulties in obtaining FCC approval delayed the system until December and resulted in cancelled orders and unsold inventory. Allied lost $3 million in the 1976 fiscal year and teetered on the edge of bankruptcy.⁹¹

The novelty companies that jumped into the market in 1975 continued to sell systems in 1976 as well, but neither of them survived the year. Executive Games experienced some success as Television Tennis reached 65,000 units sold and a follow-up game called Face Off moved another 18,000, but what would have been huge numbers in 1975 were only modest in 1976. The company closed up shop in early 1977 rather than try to continue in the increasingly competitive business.⁹² First Dimension fared worse: the company placed a large order for the discrete components to manufacture a new system in 1976 right before GI announced the AY-3-8500 and rendered the system obsolete. Norvell Olive sold out to a Tennessee businessman and politician named John Hooker, who saw an opportunity to make some quick cash, but after the purchase he learned that First Dimension had been acquiring most of its parts inventory via debt, and the company simply ran out of money before the end of the year.⁹³

Overall, 3.2 million dedicated consoles were sold in the United States in 1976 with a market value of $125 million,⁹⁴ and video games were declared one of the hottest toys of the holiday season. Another 125,000 games were sold in Europe, where the United Kingdom was the largest market with sales of 50,000 games,⁹⁵ or 40% of the European total.⁹⁶ Videomaster remained the leading company and controlled roughly 80% of the British market while also exporting to the Continent.⁹⁷ Like Coleco and Magnavox, the company embraced the GI AY-3-8500 chip for its 1976 model, the Videomaster Superscore, which played all six games on the chip.⁹⁸

With no sign of waning popularity, analysts estimated that sales in the United States could grow as high as 10 million units in 1977. Ultimately, the industry would not come close to reaching those heights because the market rapidly shifted to new platforms for games. These included not just more sophisticated consoles based on advancing chip technology, but also smaller and cheaper computers intended for personal use in the home. When the first of these so-called “microcomputers” arrived in quantity between 1975 and 1977, they were immediately capable of playing a diverse array of games developed over the proceeding decade at schools and researched labs equipped with mainframes and minicomputers useable by multiple individuals simultaneously through time-sharing.