There’s a scene near the start of Avatar: Fire and Ash that sums up the premise of the franchise, and its approach to making movies: Jake Sully, a colonialist Marine reborn as a blue-skinned freedom fighter, is trying to persuade his wife (also an alien) to accept the human weapons he’s found at the bottom of the ocean. As a proud Pandoran, she won’t touch the cursed technologies of the “sky people.” So instead he starts to strap grenades onto her wooden arrows, Rambo-style. This can be their compromise, he says: the traditions that she loves, but optimized for kicking ass.
Over the past decade and a half, James Cameron’s three Avatar movies, all shot in three dimensions (and the latter two at a high frame rate), have meted out an argument for going big in film—for strapping on the most explosive new technologies in cinema and using them to blow our minds. That project worked, at first. And then it didn’t.
The original Avatar, released the week before Christmas in 2009, made $750 million in domestic ticket sales, plus another $2 billion around the world. It was the largest total ever netted by a single film, and enough to bend reality toward Cameron’s vision of the future. The industry rearranged itself to accommodate 3-D. New cameras were invented. New theater screens and televisions were ordered and installed.
In 2011, 3-D screenings accounted for nearly one-fifth of all ticket revenue in the U.S. and Canada—a couple of billion dollars in brand-new, rocket-on-an-arrow money. Suddenly, the most famous and successful directors in the world were working in 3-D: Tim Burton, Steven Spielberg, Alfonso Cuarón, Ang Lee, Martin Scorsese. For three years in a row, from 2011 to 2013, the Academy Award for Best Cinematography went to 3-D movies. Art-house auteurs were trying out the format too: Werner Herzog made a 3-D documentary; Gaspar Noé made a 3-D porn film; Jean-Luc Godard put out a deranged 3-D provocation. Wim Wenders swore he’d never make another movie flat.
For people in the business who were learning—and effectively inventing—how to shoot these films, those first few years were chaotic and exciting. Demetri Portelli, a camera operator in Toronto at the time Avatar came out, got the chance to work the double-lensed 3-D cameras for Resident Evil: Afterlife. He had to figure out what it took to shoot in stereo, he told me—how to modify the depth effect by pulling the lenses apart, and how to shape a 3-D space by controlling how he angled them together. As 3-D took off, so did his career. He found employment doing stereography at the Olympics, and then at basketball games airing on the briefly active cable channel ESPN 3-D. He got invited to fly out to England and serve as chief stereographer for Scorsese’s Hugo. (He told me that when he packed his gear, it still was smeared with fake blood from the Resident Evil shoot.) On his website and on social media, Portelli took to calling himself “3DDemetri.”
Then the bubble popped. The next big-budget 3-D film Portelli worked on, 47 Ronin, was a major flop when it came out at the end of 2013. “There’s nothing pretty or exciting about this movie,” the critic Wesley Morris wrote, noting in a parenthetical, “Inexplicably, it’s in 3-D.” By this point, the shock wave of excitement that Avatar had set off was subsiding. In 2014, 3-D screenings accounted for 14 percent of domestic box-office revenue, according to reports from the Motion Picture Association of America, down from 21 percent a few years earlier.
More important, the industry had all but given up on shooting in 3-D. Now conversion was the norm: A movie would be shot the normal way, with a single-lensed camera, then shipped off to a giant team of rotoscopers who would remake it as a 3-D film by going through and splitting up each image, piece by piece and frame by frame. Portelli had been traveling around the world, working with some of the greatest filmmakers, and now he found himself back home in Toronto and struggling to convince his studio contacts that using 3-D cameras was ever worth the time. “I felt like a vacuum salesman,” he said. “It was heartbreaking.”
I covered the 3-D boom from the start, and even early on one could see that the golden goose was cooked. It was clear that the marginal returns on 3-D screenings were rapidly diminishing. (Theaters that showed the 3-D version of a film were making less money, on average, than theaters that showed the 2-D versions.) I wondered what was going wrong. Had the theater chains nudged the cost of movie tickets just a bit too high? Had the practice of 3-D-ifying films in post ruined the experience? Or maybe the problem had to do with quality: Were 3-D movies simply getting worse?
Portelli brought up two more problems that may have short-circuited the boom. First, people weren’t seeing 3-D movies how they should be seen. In the early 2010s, he said, a lot of theaters weren’t set up to project them at the proper brightness. (Hugo came out in 2011, but Portelli said that even he never got a chance to see it properly until 2012. “I said, ‘Oh my God, I can see the whites of the eyes of the actors in front of me again!’”) The other, bigger issue, he said, was that too many 3-D filmmakers were trying to play it safe. Before he set out to England, some colleagues had instructed him to “be careful and make sure that you protect Martin Scorsese.” They didn’t want him to take a chance on overdoing the 3-D. It turned out that Scorsese didn’t want to be protected; according to Portelli, he often pushed to make the 3-D bigger and more fun. But that sense of fear—of not wanting to be seen as gimmicky—became a trap. Some directors leaned so far toward subtlety, and the alleged virtue of immersion, that people in the theater barely noticed the effect.
By the late 2010s, just a tiny handful of directors were still experimenting with the format. Most, like Werner Herzog, never shot 3-D again. But even those like Wenders, who’d sworn that he would only work in 3-D forevermore, have now gone back to making 2-D films. Ang Lee, who won a Best Director Oscar for his 3-D movie Life of Pi, tried to make the format more appealing by shooting at 120 frames per second. With Portelli’s help, he made the 3-D, high-frame-rate movies Billy Lynn’s Long Halftime Walk, which came out in 2016, and Gemini Man starring Will Smith in 2019. Neither found an audience. Last year, in an interview with IndieWire, Lee appeared to turn his back on innovation. “The 3D is just too hard,” he said. “I’ll go back to the regular way, the old way of making movies.”
According to box-office data compiled by Comscore, a new, more modest baseline for the medium has now taken hold: These days, roughly 2 or 3 percent of new releases have a 3-D version, and they account for some 4 percent of all domestic ticket sales. The total numbers aren’t small—3-D screenings still bring in more than $300 million every year in the U.S. and Canada. But at this point, almost every single one of these 3-D movies has been converted to the format as an afterthought, in post.
For his part, Portelli is still experimenting with 3-D—a stereoscopic-video art installation that he helped create is now on display in Munich’s Haus Der Kunst—but he’s also put in work to rebuild his career in normal, 2-D cinema. He has deleted the nickname “3DDemetri” from his website.
The odd thing about all this is that the Avatar series, which almost single-handledly established the market for 3-D cinema, keeps rolling along. The second film, which came out in 2022, was colossally successful, making another $2.3 billion around the world. Perhaps the new one will be another megahit, even as the technology it champions has all but disappeared. James Cameron was among the first to use 3-D in this modern era. Now he is among the very last.
