On a Saturday in late April, in a desert landfill near Alamogordo, N.M., several hundred people gathered to watch an attempt to dig up history—and find out whether an industry legend had any truth to it.
Back in 1983, the story went, the videogame pioneer Atari, its meteoric rise having turned into a just-as-meteoric fall as the game industry crashed, took truckloads of E.T. game cartridges and buried them in the desert. The game, planned as a blockbuster hit, turned out to be unplayable by all but the most indefatigable gamers. Unsold and returned cartridges flooded back from retailers. Atari tried to make the problem go away by hiding it in the desert.
Howard Scott Warshaw, the computer engineer who designed E.T., didn’t believe the story. “It didn’t make sense for a financially failing company to pay to get rid of worthless stuff,” he says. Of course, Warshaw went on, “Anytime you think about something making sense your are losing touch with what was Atari. Atari wasn’t about making sense. It was crazy; fun crazy, but crazy.”
Warshaw was 25 years old in 1982, the year E.T. was released. A computer engineer with a master’s degree from Tulane University, Warshaw had been working at Atari for about a year and a half at the time. In many ways, it was his dream job. He had previously worked as a networking engineer at Hewlett-Packard, getting that job on the basis of some packet-switching work on the Arpanet he’d done in college. But he wanted to work on microprocessors, not big computers, so he begged his way into a job at Atari, taking a pay cut to get it. His first game for Atari, Yars’ Revenge, was a big hit—many now consider it to be the best game ever written for the Atari 2600 game machine. His second project, Raiders of the Lost Ark, was also a huge seller. The success of Raiders, he believes, is what led to the E.T. assignment.
Game development back in the early 1980s—when the very first generation of home videogame systems was dominated by the Atari 2600—was very different than it is today. Today, artists; animators; script writers; music writers; motion capture specialist;, and all kinds of programmers specializing in things like user interfaces, sounds, and networking work in large teams to produce a game. But back in 1982 game development was essentially a one-engineer show, with perhaps some input up front from a graphics artist and a sound programmer who might produce the theme music that plays at the beginning of the game. After that, though, the game developer was on his own, responsible for every image, beep, or movement on the screen. Of course, the software wasn’t nearly as big as it is today. “Yars’ was a 4K game,” Warshaw recalls. “Raiders and ET were big at 8K.” Today’s games typically fill 4 to 7 gigabytes of memory, he says—about a million times bigger.
It was Tuesday, 27 July 1982, when Warshaw first heard about the ET project. “I got a call from the CEO of the company. He said ‘We need ET by September 1st, can you do it?’ I said, ‘Sure,’ and he said ‘In two days there is going to be a Lear jet waiting for you at the San Jose airport.’” Warshaw was expected to get on that plane with a conceptual design of the game.
He knew it couldn’t be a very complex game—not and be finished in five weeks. But he also wanted to do something innovative. So he came up with the idea of a 3-D world, played in a cubic space, with some features randomized, so it would feel fresh with every playing.
On Thursday, 29 July, he got on the Lear jet in San Jose. The jet stopped in Monterey to pick up the Atari CEO, legal counsel, and another engineer who’d been tapped to do an arcade version of the game, and then flew on to Burbank, where Warshaw presented his proposal to legendary Hollywood producer Steven Spielberg. “Spielberg asked me, ‘Can’t you do something more like Pac-Man?’” Warshaw was taken aback—Steven Spielberg was asking him to do a knockoff? He argued for his original approach—and prevailed. He then moved a game development system into his home and essentially locked himself in for five weeks to finish the game. That was an incredibly short period of time—most games of the era took 6 to 10 months to develop.
“If I had had seven weeks I would have had people play it and give feedback, and I would have changed things. If I could have changed only one thing I would have made it impossible for E.T. to fall back into the hole once he climbs out. If you didn’t understand what was happening, you would jump from one screen to another and keep falling in the hole,” Warshaw says.
(Indeed, I gave that game as a gift to my cousins for Christmas 1982 and we quickly sat down to play it—we all got stuck in the falling-in-the-hole trap and quit in frustration.)
E.T. frustrated a lot of people—that's likely what earned it the “worst game of all time” moniker. “It probably wasn’t the worst,” Warshaw says, though he doesn’t mind that people think so. “People expect me to be ashamed of E.T.,” he says, “but I’m not. It was an incredible thing just to get it out and have it playable in five weeks. It just didn’t have a tuning phase, with people giving feedback.”
“In retrospect,” he says, “maybe I should have done that Pac-Man rip-off. But I don’t know if I could have done that in five weeks.”
E.T.’s fame didn’t come just because it was an almost unplayable game (except for a few hard core gamers that muscled through, learned the tricks, and turned into a small E.T. fan club). It was all about timing. The game came out just as the videogames industry began to tank. Atari went from the fastest growing company to date to the fastest falling company, and business journalists who had covered the company on the way up were looking to put a face on the story of its failure. E.T. became that face; in the news stories, it eventually got credit for single-handedly destroying the game industry.
Looking back, Warshaw says, it’s clear that that’s not what happened. In fact, the game industry was nowhere near destroyed. Rather, the falling fortunes of Atari simply represented the end of the first product life cycle of videogame machines. These days, game companies like Sony and Microsoft always have their next-generation game system in the pipeline when one generation is released—Sony, for example, was working on the PlayStation 4 when the PlayStation 3 was first hitting retail shelves. Atari simply kept selling its golden goose, the 2600, long after that technology had become dated.
Warshaw left Atari in September of 1984, after the company had contracted from a peak of 10 000 employees in early 1983 to around 200. It was hard to go. “I was 27 or so at the time, I’d been getting huge money to do whatever I wanted to do with virtually no supervision as long as I kept putting out games," he says. "I’d worked at HP before; I had an idea of what engineering was like in the real world; I knew what I was going to have to go back to.”
Warshaw put off going back for a while. He got a real estate broker’s license and tried that career—he didn’t like it. He eventually went on to work on compilers at Convergent Technologies, spending his free time on a variety of projects. He wrote a few books and then produced a documentary series, “Once Upon Atari,” some of which aired on PBS affiliates. He later took a job at video game hardware startup 3DO and moved up into engineering management. When 3DO collapsed in 2003, he realized that he’d come to be more interested in the behavior of the people he had been managing more than the technology, and went back to college to train as a psychotherapist. Today, he calls himself “the Silicon Valley therapist,” specializing in working with engineers and others involved in technology. “I’m better able to relate to engineers than the average therapist,” he says, “because I am an engineer who is a therapist.”
He’d put the whole Atari experience behind him. “I had this big fail,” he says. “I don’t think it killed the industry, but let’s say it did... That was a very big fail. But I moved on, and I went on to succeed in other things. It taught me a lot about resiliency.”
But last December, he started hearing rumors about a documentary film project, backed by Microsoft, that planned to hunt for the E.T. burial site. A producer on the project contacted him in February, and last month, Warshaw was among those several hundred spectators in the desert.
And, much to Warshaw’s surprise, the cartridges were there. There was his E.T., to be sure, but, to his satisfaction, E.T. wasn’t alone: Joining it were unsold, shrink-wrapped cartridges of Defender, Centipede, Space Invaders, and Missile Command—a gallery of Atari 2600 hits.
“I’m glad they found it,” he says. “It makes the story into a much better movie.”
And, he says, in a crazy way, the dig turns E.T. into what he considers a success. “The whole purpose of doing the game was to entertain people. And so here we were, in a landfill in the middle of the New Mexico desert, and there were hundreds of people lining up to get in. There was a lot of excitement; everything was very intense. So something I did 32 years ago is still generating entertainment and interest. That’s something. Because how many 2600 games do people still play today? Not many. And how many do they make movies about?”
Just, it turns out, one that failed.
For more about the challenges of programming the Atari 2600, see IEEE Spectrum’s “Design case history: the Atari Video Computer System” [pdf]
Tekla S. Perry is a senior editor at IEEE Spectrum. Based in Palo Alto, Calif., she's been covering the people, companies, and technology that make Silicon Valley a special place for more than 40 years. An IEEE member, she holds a bachelor's degree in journalism from Michigan State University.