
When a set is three walls and a ceiling, the view outside the window has to carry much of the film. For SNTIENT 2.0, the Fashion Innovation Agency’s film for London College of Fashion, that view is a bleak, abandoned unreal engine city. It exists only inside Unreal Engine.
The story imagines a future where resources have run out. Humans have retreated into a simulation, and the robots left to tend the physical world have begun to find beauty in it. “That gave us our visual starting point. A bleak, oppressive sci-fi environment with the potential for transformation,” says Dickon Knowles, director of content at Dandelion + Burdock.
The finished film explores that transformation less than the team planned. The city still sets the tone for everything else.
Changing the unreal engine city in real time
Speed was the main reason for choosing Unreal Engine 5. Because the engine renders in real time, the team could adjust space, lighting and textures and see the result immediately. They did not have to wait for a render to find out it was wrong. “It made the whole process much more fluid,” Knowles says.
The team tested other routes first. “We tried a couple of generative approaches to creating the city plate for the wall, but resolutions were an issue and the visual fidelity wasn’t there,” Knowles says. They went back to their existing Unreal Engine version.
That unreal engine city still needed work. The original environment was foggy and slightly desaturated, and it did not read the way the camera needed. Knowles replaced the broad haze across the scene with more detailed VDB fog, which is volumetric fog that can be shaped with precision. He reduced its presence and boosted saturation, especially in the city windows.
The change mattered on the LED wall. The fog had been flattening the blacks and reducing contrast. The wall needed a more vivid backdrop to hold up on camera.
Most of the unreal engine city was built before the shoot. On set, the team focused on visual refinements, grading and small tweaks. They were not building a digital world while the cameras rolled. They could see each change on the wall and through the lens, then adjust without waiting for a new render.
Lighting the actor from the window
The LED wall also gave the performer something real to react to. She looked at a lit unreal engine city rather than a green screen she had to imagine.
Ben Saffer, the cinematographer, handled the virtual production through his company, Supervision VP. He says the main benefit was capturing everything at once. Virtual and physical lighting could be controlled together, so the glow on screen matched the light on set. “You’re doing it all in camera and capturing everything together at once, and you’re not having to guess at anything,” he says.
The alternatives were a green screen or a pre-rendered backdrop. Knowles says cost mattered less than image quality. The unreal engine city content already existed from an earlier project. “The city lights reflect on the performer’s face and on the surfaces of the props and set,” he says. “There is a level of visual immersion that is hard to capture in a studio without the LED wall to illuminate the space.”
The wall did more than decorate the view. It lit the performer and the physical set at the same time. The approach resembles how Magnopus used Unreal Engine for Fallout. The aim here was to keep post-production to a minimum and capture footage close to the finished shot.
A shutter made of pixels
Unreal also solved a budget problem. The film needed a shutter that rises to reveal the city. A mechanical shutter built into the set was out of reach for the budget. Adding one in visual effects would have removed the in-camera moment.
So the LED wall became the shutter. It rises to show the digital city behind it. For reverse shots, a second wall on wheels served as a fourth wall. Because the content ran in real time, the team rendered and animated new assets for those angles quickly. “It was nice to get that in camera,” Saffer says.
Not everything was virtual. The pod is a physical build, because the team wanted it to look gritty and lived-in. The LED wall and Unreal handled only the distant environment. The opening dream sequence, which shows the character floating in water before she wakes, was made separately with generative tools.

Two minutes, one room, one lens choice
The film has few story beats. The character wakes, the window opens, and she looks out. Saffer says the real challenge was conveying the world and her place in it within two minutes.
The crew shot on a Blackmagic URSA Cine 17K 65 with prototype Atlas Lens Co. anamorphic lenses. The larger sensor let Saffer capture more of the small set without a extreme wide-angle lens. That kind of lens would have bent the perspective.
The shoot also left room for hands-on tricks. Near the end, as the character turns toward the window, Saffer held a Polly Pocket toy in front of the lens. He shone his phone torch through it from the side to create an optical effect. He also tried cling film and other materials in front of the lens.
The studio’s own account of how SNTIENT 2.0 was made on Dandelion + Burdock goes deeper into the production. The lesson from this shoot is simple. When the view outside the window is lit by the same wall that lights the actor, the unreal engine city stops being a backdrop and becomes part of the scene.

H/T creativebloq.com
Frequently asked questions
Why did the SNTIENT 2.0 team choose Unreal Engine 5 for the film?
Speed was the main reason, because Unreal Engine renders in real time. The team could adjust space, lighting, and textures and see the results immediately instead of waiting for renders.
Why did the SNTIENT 2.0 film use an LED wall as its shutter?
A mechanical shutter built into the set was beyond the film’s budget, and adding one in visual effects would have removed the in-camera moment. The LED wall was used instead, rising to reveal the digital city behind it.
What camera and lenses were used to shoot SNTIENT 2.0?
The crew shot on a Blackmagic URSA Cine 17K 65 with prototype Atlas Lens Co. anamorphic lenses. The larger sensor let the cinematographer capture more of the small set without an extreme wide-angle lens.
Which parts of SNTIENT 2.0 were physical and which were virtual?
The pod was a physical build, made to look gritty and lived-in, while the LED wall and Unreal Engine handled only the distant environment. The opening dream sequence was made separately using generative tools.
More in
Cinema
All Cinema



More like this

Spacejetters Creator Mikeron Turns Vintage Sci-Fi Covers Into AI Worlds

Concept Artist Brady Builds Dune Anthology Showing Giedi Prime's Black-Sun Megaliths

Guillem H. Pongiluppi Blends History and Imagination in Digital Fantasies

Maciej Rebisz Crafts Intricate Space Art with Real Models and Sci-Fi Inspiration

Maxim Goudin’s 3D Art Redefines Reality with Surreal Symbolism and Science Fiction Flair
