The future arrived in 1994 as a Saturday-morning cartoon. It crackled with the excitement of the early internet age: of modems and Windows and CPUs, the animated equivalent of dialing into AOL or Prodigy for the first time. It was the first computer-generated television show ever made — a show about computers, made with computers — and unlike anything else on air.
ReBoot took place in Mainframe, a city inside a computer that was constantly under threat by malevolent digital forces. There were a pair of evil viruses, Megabyte and Hexadecimal, and an ominous purple cube that descended from the sky, droning, “WARNING: INCOMING GAME.” Megabyte and Hexadecimal wanted to take over Mainframe, and infect other computers connected to the then-nascent net. The cube was the physical manifestation of a video game, and it threatened Mainframe with destruction each time it appeared.
A mysterious guardian named Bob protected the city with Dot, owner of Dot’s Diner. Fighting alongside them was Dot’s little brother Enzo and Frisket, his robot dog. In each episode, they kept the viruses and game cubes at bay.
ReBoot aired for four seasons, and by the end of its run, it could be seen in more than 60 countries. It won a Gemini Award for best animated series three years running and was inducted into the Smithsonian in 1998 for its technical achievement (nominated by Bill Gates, no less). For a tiny team on a TV budget, it was no small feat.
For kids, ReBoot was a flashy alternative to traditional animation, an unfamiliar world rendered in bleeding-edge 3D. It was slick with groan-worthy catchphrases (“Alphanumeric!”) and computer-savvy puns. It humanized technology by using technology. In bringing computer animation to the small screen for the first time, it was like prestige TV for kids, the way people hailed The Sopranos as Hollywood-level drama that you could watch in your pyjamas at home.
And for adults there was something too: pop culture references and computer terminology that flew straight over younger viewers’ heads, of course, but also the potential to profit from this strange new world. "If you study motion pictures and animation, obviously Disney is the first and last name you'll hear," ReBoot's producer Christopher Brough told the Vancouver Sun in 1996. His company, Mainframe, was preparing a “unilateral assault” against the animation giant, with “toys, video games, interactive products, all of it at once to create a brand recognition, a kind of marketing beachhead” that would use computer-generated shows like ReBoot to take Disney head-on.
With the techniques and processes that the team behind ReBoot were pioneering, a 3D animated series could be made in less time, and with fewer resources, than the traditional hand-drawn animation that dominated movie theatres and television screens. Computer scripts and render farms would do all of the heavy lifting that rooms of human animators once did. And because everything was digital — mere files on a disk — those assets could be reused effortlessly to make commercials, movies, toys, and games. Think Dot on a lunchbox, flying through the city, or Megabyte’s lair on a lampshade.
Even Pixar, which released Toy Story in 1996 to universal acclaim, would be no match for Mainframe’s efficiency. Where Pixar took years to produce just 81 minutes of footage, the ReBoot team was determined to produce the same amount in a matter of weeks.
The message was clear: Mainframe was going to be the biggest animation company in the world — the Pixar of the North — and ReBoot was just the start.
ReBoot's influence is everywhere if you know where to look. In the early to mid-1990s, while Disney was in the midst of its hand-drawn animated renaissance — with movies such as Beauty and the Beast, Aladdin, and The Lion King enjoying blockbuster success — the first generations of computer artists were getting their start. ReBoot was one of the few productions where forward-looking animators and artists could learn to work with computers outside of school. Star Wars, Harry Potter, Finding Nemo, Shrek — all were made with the help of ReBoot alumni.
ReBoot was the brainchild of Ian Pearson, a talented animator and bullheaded Geordie from Northern England who spent the early 1980s in the UK making corporate logos fly around a screen. Neon-tinted wireframes and hard-edged shapes were what passed for groundbreaking stuff. But Pearson had grander aspirations. "He knew what was possible, even if he didn't necessarily know how to do it,” said Chris Welman, a software developer in ReBoot’s earliest days, and later the company’s vice president and chief technology officer. “He had a good visual eye. He could look at a model and tell the guy who built it what was wrong with it and why it wasn't beautiful."
In 1984, Pearson worked with music director Steve Barron on the visual effects for Dire Straits' “Money for Nothing,” the first music video made with computer graphics. Instead of blocky logos, it featured blocky humans, but the result was good enough that Pearson figured there was potential to do a whole show. Given the state of most computer graphics at the time — primitive, flatly shaded, crude — he came up with a clever conceit. The show would take place in a computer, thereby eliminating the expectation that it had to look remotely realistic.
Pearson’s assistant on “Money for Nothing” was a fellow animator, a jolly, denim-clad man in round glasses named Gavin Blair. Over the span of nearly a decade, the two tossed the concept for their computer-generated show back and forth, and started to assemble a team. Blair roped in John Grace, his former animation lecturer, who made a stop-motion animated television show called Portland Bill in 1983. They brought on Phil Mitchell, an ex-classmate of Blair's and a gifted animator in his own right who had grown tired of working on commercials. He was a mountain of a man who listened to death metal at his desk while he worked. Collectively, they were known as The Hub.
Barron, the music video director, introduced the team to Chris Brough, who had been a producer at Hanna-Barbera and DIC Entertainment. Brough had a production company in Vancouver, BC, and was interested in helping The Hub turn ReBoot into a show. So Pearson, Blair, and Mitchell left the UK — Grace elected to stay behind — and joined Brough in Vancouver in the spring of 1993. (Naturally, tax breaks played a part.) They had no staff, only a handful of computers, and spent the first few months working out of hotels, wondering if they could even pull the whole thing off.
"I thought, ‘Great, I'm going to learn how TV shows are made with computers,’" said Welman, who was hired by The Hub when he was then just a recent computer science grad. "Nobody really knew what they were doing at the time."
By today's technical standards, ReBoot does not hold up well. The animation was stiff and unnatural, especially in its first season, the world harshly lit, and the models primitive in their design. In the years that followed, when ReBoot animators later went for job interviews at animation studios in the United States, potential bosses thought the footage in their portfolios was a draft.
Yet, there was a charm to the city of Mainframe in spite of its stilted visuals. It was well-written — and funny — if a bit heavy-handed at times. Mainframe was primarily inhabited by binomes, people shaped liked ones and zeros. Characters spoke to each other through video windows, at a time when mice and cursors were still novelties, shiny and new. The city’s main strip was called “Baudway” and Dot’s full name was Dot Matrix. Episodes parodied Power Rangers, Mad Max, and James Bond. In the second season, X-Files star Gillian Anderson appeared not as Scully but as a character named Data Nully (her partner: Fax Modem). There were puns galore, and slapstick gags were instrumental to the show.
There was the grating sendup of mail-order infomercials, a literal television called Mike the TV. In one episode, software pirates arrive from the net and capture Bob, while the memorable "Talent Night" simply ends with the virus Megabyte treating the city of Mainframe to an extended guitar metal solo.
Those who worked on ReBoot did a little bit of everything. The person who set up the camera on a particular scene might also handle the lighting. A director might notice a scene was missing a building or tree and have to model it themselves. There wasn’t the sort of specialization that exists in the industry now. Everyone was a generalist, and it was technical, time-consuming work. "It was not even reinventing a wheel,” said Scott Speirs, a director at Mainframe from 1993 to 2001, who is now head of production at the Montreal office of visual effects company MPC. “It was trying to invent a wheel while the car was going down the hill. Or build the car as you're driving it — but you're not just driving it, you're in a race, and you have to get to the finish line before anybody else."
Work on the pilot began in earnest towards the end of 1993 and took nine months to complete. Chaotic at the best of times, it set the tone for the rest of Season 1. Blair and Mitchell both remember staying until the wee hours of the morning, working on scripts and waiting for footage to render. People slept under desks, on cots and couches. Andrew Grant, one of the company’s early hires, recalls his wife bringing him fresh clothes and food for those long nights at the office. The wives of some employees referred to themselves as members of the ReBoot Widows Club.
The pace was relentless, no more so than when deadlines loomed. For ReBoot’s third episode, “The Medusa Bug,” a rogue program turned nearly everyone in the city of Mainframe to stone. "Of course,” Blair recalled, “if everybody gets turned to stone, there’s no animation." In reality, this was just a clever cost-saving measure disguised as a plot twist.
The hardware and software was being pushed to its limits too. They bought bleeding-edge computer equipment from Silicon Graphics, the industry leader in 3D imaging at the time, for millions of dollars with little assurance the hardware would work the way they needed. They used the same software, SoftImage, used to create 1993’s summer blockbuster Jurassic Park, but in ways its Montreal-based developers had never intended. Welman recalls one memorable bug in which characters would explode “and their eyes would sink into their face.” When off-the-shelf software was unavailable, or didn’t work the way they wanted, Welman would write his own — such as Grin, which synced recorded dialogue to the characters’ facial animations.
The goal was to create a pipeline that could eventually take ReBoot episodes from script to TV in a matter of weeks. At the time, computer animation was mostly relegated to the big screen, done by companies with deeper pockets and bigger teams — and even then, mostly used for special effects, a few seconds at a time. Rendering, the process of flattening three-dimensional data into a completed visual image, could take hours per frame, and that was assuming you could find talented programmers and artists to do the work, which very few schools taught.
Then came The Hub, who decided they were going to take this cutting-edge technology, primarily used for Hollywood films, and teach a handful of inexperienced students to make serialized computer-generated television on the cheap. By 1996, Mainframe was producing almost 17 minutes' worth of fully rendered footage each week — 24,000 frames — and had 110 computer workstations from Silicon Graphics. Episodes no longer took months to make, but six to eight weeks. They had more than 200 employees. It was audacious, and exactly what investors in this exciting new medium liked. What they didn’t like was what ReBoot’s creators tried to do next.
At the end of ReBoot's first season, another 12 episodes were ordered, and a second season premiered in the fall of 1995. At the same time, they struck a deal with toymaker Hasbro to make Beast Wars, a computer-animated series based on the Transformers line of toys. Though ReBoot was a hit, it was expensive to make, far outstripping what they made from the show. They needed steady cash flow to keep the lights on — not to mention training new animators and investing in new technology. So they signed additional deals to develop shows for toy companies including Trendmasters to Mattel.
Unlike ReBoot, it was service work. Mainframe did what the toy companies wanted, which wasn’t always as fulfilling as working on a concept they owned. But the goal with these projects was to make money, not art. Mainframe needed cash to subsidize its founders’ real ambition: to make Hollywood films.
Just as computer-generated television was in its infancy when ReBoot hit the small screen, so too were the Hollywood films being created with computer graphics. There was Pixar, of course, but most animation studios were still focused on superimposing computer graphics — think explosions and spaceships — onto live-action scenes. Mainframe saw an opportunity for its creativity to scale. The idea was that once the television side was humming along, they could finally get a feature film off the ground, something with a modest budget of $30–$40 million — low for an animated picture at that time, but enough to produce a film nonetheless.
The company's partnership with IMAX in 1996 was a natural first step, a way to prove they could scale their technology and process to the literal big screen. Mainframe would work with IMAX on something called a Ridefilm — moving seats that, when combined with the immersion of a massive IMAX screen, could instil feelings of dramatic motion.
In theory, Mainframe would do a handful of Ridefilms and graduate to doing IMAX films, which would lead to animating feature-length scripts. It would show prospective partners that the studio had what it took to make not only a Hollywood film, but one that could go toe-to-toe with some of the best in the business — upstarts like DreamWorks and stalwarts like Disney.
There was certainly no shortage of ideas. Former employees recall the amazing film tests the company did for an IMAX adaptation of Gulliver's Travels. Announced in September 1998, it would be rendered in stereoscopic 3D — requiring audiences wear those plastic throwaway glasses — and Mitchell was going to direct. They bought the rights to Graeme Base's children's book Sign of the Seahorse. There was a script written by Steve Oedekerk of Ace Ventura fame, a comedic sendup of Disney cartoons and happily-ever-after style stories. They poured millions into research and development.
But despite negotiations with multiple studios, nothing made it into production. For Gulliver’s Travels, IMAX and Mainframe couldn’t agree on a script. Sign of the Seahorse had great potential, but then Pixar's Finding Nemo came out and no one wanted another fish-based movie. Mainframe even had a shot at Shrek when the film still starred Chris Farley and was a mix of live action and CG, but they turned it down in favour of their own scripts.
Attempts to make a movie dragged on for years — the exact opposite of Mainframe’s pitch to investors that it could make computer animation faster than anyone else. “We were burning millions of dollars a year in this quest to make films, and we weren't making anything out of it,” Brough recalls.
Mainframe was the quintessential dot-com startup. There were catered meals and an office hockey team, where everyone wore jerseys stylized to look like ReBoot’s villain Megabyte. Employees watched the same television shows and drank together during happy hour. Some remember a Nerf gun fight in which a stray foam bullet landed on a keyboard and, as Blair recalls, interrupted a render in progress. (Nerf guns were consequently banned in the office.) Rendering even a few seconds of footage could take hours, so animators spent a lot of time waiting, often at the local bar, The Cecil. In truth, it was a strip club, but it was also the closest thing to an English-style pub ReBoot's expat creators had. (In the show, they even named the waiter at Dot’s Diner after The Cecil, which is why he hung from a brass rail.)
The Cecil became an extension of the office; they even did job interviews there. "The theory was, if you could get through an interview focused at Cecil's then you could probably get through life at Mainframe, because there was a lot of distraction, a lot of drama,” said Paula Fellbaum, the studio’s first recruiter and head of HR. (She did, eventually, convince them to stop conducting business at a strip club.)
And in typical startup fashion, the studio had big ambitions. Mainframe debuted on the Toronto Stock Exchange on June 17, 1997, with an initial offering of 3 million shares at $9.75 CAD apiece — nearly $30 million. The proceeds would go towards new equipment, new shows, and the development of a movie. It was supposed to take the company to another level. As far as their stock price was concerned, they assumed that level was up, not down.
Part of Mainframe’s pitch to investors was its library of proprietary content. They had a pipeline that would take the studio’s intellectual property and find ways to generate revenue not just from the sale and syndication of its programming, but from toys and merchandise. The strategy worked well for ReBoot, which Mainframe owned, but less so for properties like Beast Wars, which saw a significant amount of the royalties go to the toy companies. Before long, making shows for toy companies was pretty much all Mainframe did.
By the end of the decade, Mainframe had five computer-animated television shows simultaneously in production, all based on toys. There were at least as many under development, based on ideas from Star Trek creator Gene Roddenberry, the Alien franchise, and PlayBoy. It was all very promising — except that while Mainframe received production advances and government funding for many of its shows, the bulk of its revenue came upon delivery. And the number of episodes it was actually delivering to TV networks was relatively flat year over year.
At the same time, the company had grown to more than 400 employees, and there simply wasn’t enough work, nor revenue from the work they had, to justify paying such a sizable staff. Management alone were earning upwards of $1.7 million a year by 1998 — a tenth of what the company was making—with nearly $500,000 apiece going to salaries for Pearson and Brough in 1998.
It was clear that, after ReBoot, the company was struggling to find stable revenue and was plagued by massive overhead costs.
Everything was compounded by the fact that the industry around Mainframe was changing too. It was only a matter of time before other studios tried to do what Mainframe was doing. Some employees also left Mainframe to start animation studios of their own. Hardware and software became cheaper and more accessible, lowering the barrier to entry for the competition. At the same time, Mainframe still had to sink millions in research and development costs to keep up. Eventually, deep-pocketed US giants such as Disney and DreamWorks developed their own computer animation capabilities, meaning they could now do themselves what they once might have hired Mainframe for.
The irony is that computers actually helped rather than hindered the more traditional forms of animation that Mainframe sought to supplant. “While CGI was storming ahead, so too was Flash, Claymation, old standard 2D, and even paper cut-outs programs,” said Dale Taylor, a programming executive at the Canadian television channel YTV who helped put ReBoot on air. Computers, too, made these mediums “cheaper, expendable, and far more efficient to get to market.”
And on top of it all, there was all the time, money, and effort that the team had poured into a movie with nothing to show for it. “[Pearson] was constantly arguing back and forth with the board about the direction of the company, and we got frustrated,” Blair said.
Two years after that initial IPO, its stock was trading for less than a dollar. Investors had lost faith in the direction of the company, and the way it was being managed. Its CEO, Christopher Brough, resigned. And the movie? “We could see that it was never going to happen,” Blair said.
For a certain generation, too young to have grown up with ReBoot or even Beast Wars, Mainframe is better known as the studio behind Mattel’s series of computer-animated Barbie films. These were released annually for over a decade, starting with Barbie in the Nutcracker in 2001. They were the elusive cash cow that, just barely, kept Mainframe in business each year. It was that same year, in 2001, that ReBoot’s fourth season premiered.
In the nearly four-year hiatus since ReBoot’s third season, Mainframe pursued other shows and feature films. In search of stable revenue, it shed a couple hundred employees. At the same time, it became clear that the world into which ReBoot was first introduced had also undergone a change. Where in its first season computer graphics were a novelty, now they were commonplace. Pixar had already put out a handful of films. There were other animation companies making CGI for TV — and the quality had gotten better, too.
Ultimately, ReBoot’s fourth and final season was a love letter from The Hub — Ian Pearson, Gavin Blair, and Phil Mitchell — who had realized the company they had founded almost a decade prior was slipping from their control. What better way to go than with the show that started it all?
After the turn of the century, control of the company changed hands multiple times as investors sought a leader who could steady Mainframe’s financials. Naturally, that meant pursuing more work for hire and fewer creative risks. Pearson and Blair left the company in the months following the premiere of ReBoot’s final season, while Mitchell was out by 2005. Mainframe was eventually sold, and subsequently renamed Rainmaker in 2007. (Pearson declined to comment for this piece through Blair, one of the few former employees with whom he still speaks.)
In a roundabout way, Mainframe (now Rainmaker) finally did make its movie, Escape From Planet Earth, in 2013, grossing $75 million USD worldwide, a modest success for an animated film. (Comparatively, Monsters University, which came out that same year, grossed just under $745 million USD.)
Riding high on the popularity of ReBoot, and a successful IPO, there was a time when Mainframe thought it could be the greatest animation company in the world. "Maybe it sounded a bit cocky, but I felt, had we been given a little broader bandwidth, I think we might have pulled it off,” Brough recalls.
Pioneering as Mainframe was, only Disney, Pixar, and DreamWorks remain household names. Yet it’s hard to overstate how important a show ReBoot was to the burgeoning world of computer animation. In Canada, certainly, it was a hallmark of children’s television. Globally, it was a technical marvel. But behind the scenes, there were a couple hundred people figuring out what computer animation might become.
"We called it ‘Mainframe University,’" Mitchell said. It wasn’t long into ReBoot’s development before early employees started to leave for Pixar, DreamWorks, and Industrial Light & Magic. Others started their own studios, such as Nerd Corps, which is now owned by DHX Media. Young animators would cut their teeth at Mainframe, working on one of its myriad toy-based shows — but what they were really doing was getting a crash course in animation. Some of those employees still work at what’s known as Rainmaker today.
If only for a brief period of time, Bob, Dot, and Enzo — the city of Mainframe itself — were the future of animation, a potential threat to the primacy of traditional animation, and to giants like Disney. Then the ominous purple cube descended on Mainframe, too, and there was no way for them to win.
"Lightning in a bottle," said Blair, recalling a phrase that one of Mainframe’s former executives once said. “It was the right people, in the right place, at the right time."