Chapter 770: 3D Will One Day Change Film

Duke's unconventional shooting methods for Gravity meant its post-production would be vastly different from his previous films. As director, he had realized this during early pre-production.

Setting aside technical and financial issues, the first major hurdle was time.

To secure limited screenings before the new year and qualify for the Oscars, they had only three months for post-production. For a film of this scale, the difficulty was staggering. Even John Schwartzman and Tim Webber had urged Duke to skip this year's Oscars, arguing that waiting for the next wouldn't hurt.

But they didn't know what Duke was thinking. He was rushing the film specifically for this year's Oscars. Unless Black Swan also pushed its North American release to the following year, he wasn't going to change his plans.

That was obviously impossible. Black Swan had already held its global premiere at the Venice Film Festival and was slated to hit North American theaters in October.

After wrapping the shoot, Duke made another trip to Industrial Light & Magic and George Lucas's home. Gravity's production budget had swelled to $230 million, with nearly two-thirds of those funds poured directly into visual effects.

This was a commercial industry. As long as he could pay the price, he could hire enough people.

Industrial Light & Magic mobilized nearly every available hand for Gravity's post-production. Beyond North America, ILM's branches in London, India, and Singapore joined the fray, assembling a massive army of personnel and digital artists dedicated solely to the most time-consuming task: shot rendering.

However, Duke knew that making a truly outstanding film required more than just visual effects; storytelling was equally vital. Gravity was realistic, but it wasn't rigidly bound to absolute reality.

For instance, the orbital altitudes of many space stations and satellites were deliberately compressed in post-production.

Because the film was constructed from numerous long takes—with shots exceeding five minutes commonplace—editing was relatively less tedious than on his past projects. Duke dumped the rough cut entirely on Mike Dawson and stationed himself at the visual effects headquarters at Warner Studios, coordinating a global VFX effort spanning North America, South Asia, Europe, and Southeast Asia.

Visual effects work had begun long before cameras even rolled and continued non-stop throughout the shoot. Otherwise, it would have been impossible to finish a production where rendering alone required months. By spending heavily to gather a massive crew, Duke ensured that many shots could be processed simultaneously.

Of course, the consequence was that exorbitant production budget.

But he had the capital for it now. As long as the budget stayed under $300 million, he was confident the review committee at Warner Brothers wouldn't even blink.

There were countless crucial tasks throughout post-production, but shot rendering was undoubtedly the most tedious.

Per Duke's requirements, the completed footage and CGI shots needed a multitude of added lens effects, including light bleed, flares, and chromatic aberration.

Due to the tight schedule, Industrial Light & Magic's visual effects team decided right from the start to use their newly developed VFX software, the Arnold Renderer. It was incredibly powerful, but had never been practically applied to any previous film.

Since it was their first time using it and they faced an exceptionally massive workload, many people—Duke included—were somewhat worried. Fortunately, the Arnold Renderer boasted superb memory management, allowing it to successfully render many complex scenes.

This was where the Arnold Renderer excelled: rendering vast amounts of complex geometry while simultaneously handling extensive ray tracing. Because the rendering process took so long, and with the immense pressure Duke applied, the ILM VFX team had to constantly verify they were on the right track. If their output failed to elevate the film's overall texture, they would undoubtedly be smashing Industrial Light & Magic's reputation.

Furthermore, Duke had them employ numerous instancing techniques to increase visual complexity.

"Part of the reason is that doing this enhances the audience's sensory experience," Duke explained to Tim Webber, Industrial Light & Magic's VFX supervisor on the crew, helping him understand the necessity of this approach. "Because when you shoot in space, you can't possibly use the best cameras, nor can you swap lenses. There's a lot you just can't do. So, using those lens effects replicates the feel of actual space photography."

Based on Duke's understanding, genuine space photography really did capture some light leaks. Because the sun was so distant and space was pitch black, if the shadowed areas didn't reflect the light bouncing off Earth, the contrast would be incredibly harsh—far starker than anything seen on Earth.

Therefore, the film required many high-contrast images.

Figuring out how to convince the audience of the authenticity of events happening in space was quite a headache. Duke couldn't use the standard tropes typically found in space movies, whether through practical simulation or visual effects. For instance, atmospherics—the pulsed electromagnetic radiation caused by atmospheric discharges. Making distant atmospherics look muffled and weak simply wouldn't meet moviegoers' expectations.

Then there were explosions. If an explosion happened in the cosmos, hardly anyone actually knew what it would look like.

Therefore, post-production wouldn't strictly adhere to the data provided by astronauts. Sometimes, they had to make exceptions. For example, it usually took an astronaut fifty minutes just to exit a space station. They obviously couldn't copy that verbatim; if the audience was enjoying the film and suddenly had to sit through something like that, the consequences would be disastrous.

Of course, regarding movement, Duke still aimed to base it on reality first, though there would be exceptions. But he would always ensure those movements felt purposeful.

There was another task that would directly determine the film's success or failure, and that was 3D.

Driven by the craze sparked by Avatar, many Hollywood studios were now preparing to convert their classic films into 3D for theatrical re-releases, hoping to cash in on the trend and make a fortune off recycled hits.

Without a doubt, Duke was ahead of all of Hollywood. Earlier that year, the 3D version of The Lord of the Rings trilogy had already ridden Avatar's coattails to gross over $300 million at the global box office.

As for Gravity, Duke decided to make it a 3D film more to serve its story. Most of its scenes involved floating in outer space, and using 3D would undoubtedly yield spectacular results.

Early on, Duke had confirmed that Gravity would be a blend of CGI and live-action. But prior to rolling cameras, he hadn't decided whether to shoot directly in 3D or convert it in post-production.

Duke and John Schwartzman had tested shooting in 3D, but in practice, squeezing bulky 3D camera rigs into the cramped Light Box was incredibly cumbersome.

So, when Duke visited Industrial Light & Magic and brought back the 3D conversion tests, the crew realized something. Given the specific environment of space and their unique shooting methods, the post-converted live-action footage looked no different from native 3D. Plus, shooting in 2D was vastly more efficient.

Consequently, Duke abandoned the idea of shooting in native 3D without hesitation.

But Gravity was absolutely not going to be a fake 3D film. Its shooting and production methods were extremely unique, completely different from traditional 3D conversions.

The traditional method was to convert the movie into 3D during post-production. Gravity was different.

As early as pre-production, ILM's 3D conversion work had already begun. Duke's 3D director, Chris Parks, had been working closely with the relevant teams at Industrial Light & Magic, explaining Duke's meticulously crafted concepts to them in detail.

For example, the stark contrast between the vast emptiness of the cosmos and the claustrophobically cramped space capsules.

From the very beginning, this team faced a massive challenge: they had to ensure the 3D-converted live-action footage would seamlessly integrate with the natively 3D CGI visuals.

The live-action shots needed to be converted to 3D, while the CGI was natively rendered in 3D. Both were 3D, but one was a conversion and the other was built through visual effects.

Furthermore, while doing all this, they couldn't even guarantee these would be the final shots Duke wanted. The team had to maintain as much flexibility and creativity as possible throughout the conversion process.

Moreover, when they started working, Gravity wasn't even a finished film. They worked alongside the crew, following the shooting schedule. This team didn't wait until post-production to begin converting; they maintained the exact same pace as Duke's production crew.

The entire post-production pipeline was tightly interconnected—a textbook team effort where Duke was simply the commander-in-chief. For many shots, processing solutions were drafted in Los Angeles and San Francisco, transmitted to India and Singapore for the most tedious rotoscoping work, forwarded to London for further processing, and finally bounced back to Los Angeles for Duke and VFX supervisor Tim Webber to review.

Duke had to ensure that all visuals matched their environments perfectly, making those environments even more beautiful whenever possible. Naturally, achieving this required the united effort of the entire post-production team.

In the final edit, the strict technical demand was to make the visual transitions absolutely seamless, hiding the cuts entirely from the audience. Duke often considered where to cut from a 3D perspective, and the entire editing team did the same for the 3D process—finding that subtle cut point that made the footage look entirely unedited.

Without a doubt, Gravity was the most complex and detail-rich project Industrial Light & Magic had ever completed.

From the length of the takes to the high demands for detail during the conversion, Duke had pushed everything to the absolute limit of modern capabilities. Even many of those who had seen his completed fragments believed that 3D would one day change film.

Time quickly marched toward Thanksgiving. Just before the holiday arrived, Duke temporarily set aside his work, preparing to attend the premiere of the final chapter in the Harry Potter series.