Stop-motion animation has a reputation for being slow, and the reputation is earned. One animator sits at one set, moves a puppet a fraction of an inch, takes a frame, moves it again, takes another frame. Twenty-four of those gets you one second of film. A good day might produce three or four seconds. That is why the great stop-motion features take years, and why studios that make them tend to make one at a time and then go and lie down.
The new Shaun the Sheep film was shot in about eight months.
It managed that by refusing to be one production. At its peak, Aardman was running as many as forty-seven separate shooting units at once, spread across nearly fifty handmade sets, with somewhere between thirty and forty animators working in parallel rather than in sequence. Every one of those units was, functionally, its own little film crew, with its own lighting, its own camera, its own animator making his or her own hundred small judgements a day about how a sheep should shift its weight. All of them had to produce footage that cut together as though one person had made it.
The film is called Shaun the Sheep: The Beast of Mossy Bottom, it runs eighty minutes, and it is the third feature built around a character who started life as a supporting player in a Wallace and Gromit short in the mid 1990s. It arrived in cinemas in the United States through GKIDS, with Sky handling the United Kingdom and StudioCanal the rest of the world, and it is directed by two first-time feature directors, Steve Cox and Matthew Walker, both of whom came up through the franchise as animators, story artists and episodic directors.
Shaun’s origin is worth knowing, because it explains the silence. He turned up in 1995 as a one-joke gag in the Wallace and Gromit short A Close Shave, a sheep who gets run through a knitting machine. He was never designed to carry anything. He got a television series more than a decade later, and that series turned into one of the most widely exported pieces of children’s television anyone has made, precisely because it has no dialogue to translate. Nothing needs dubbing. Nothing needs subtitles. A show about a sheep outwitting a farmer plays identically in every country on earth, which is a commercial accident that turned into a creative discipline.
The plot is exactly as daft as it should be. The Farmer wrecks the flock’s pumpkin patch while preparing for Halloween. Shaun, who has never once responded proportionately to anything, decides the solution is to become a mad scientist. The experiment goes the way experiments go. The Farmer vanishes. Something large and hairy starts moving around in the woods outside Mossingham. Nobody says a word for eighty minutes, because nobody in a Shaun the Sheep film ever does.
The film was written by Mark Burton and Giles Pilbrow, produced by Richard Beek, and scored by Natalie Holt, who had not written for an animated feature before. The voice cast, such as it is, is mostly the people who have grunted and bleated their way through the franchise for years: Justin Fletcher as Shaun and as Timmy, John Sparkes as both the Farmer and Bitzer the sheepdog, Kate Harbour as Timmy’s mother. They are credited as a cast, but what they are actually providing is noise with personality in it.
That silence is worth pausing on, because it is the constraint that makes everything else difficult. There is no dialogue. Not a line. Every joke, every scare, every beat of plot has to be carried by a puppet’s posture and the position of two little plastic eyes. When something has to be explained, it has to be shown. When something has to be funny, it has to be funny physically, in the frame, without a single word of help. That is hard when one director is watching one shot. It is a genuinely strange problem when forty-seven crews are shooting simultaneously and nobody can hear what the film sounds like yet.
The way they solved it was structural, and it is the reason there are two directors rather than one. Cox has said plainly that a stop-motion film has to split between two people because the workload is simply too large for one. Walker described the rhythm: each morning the two of them sat down with everything shot the previous day, across every unit, compared notes, and made sure the separate pieces were still adding up to a single coherent film. Then they went back out and did it again. For eight months.
Think about what that meant in practice. On any given morning there were up to forty-seven animators who had each spent the previous day producing a few seconds of footage, alone, at their own set, without seeing what anyone else was doing. A performance choice made at unit nineteen had to match a performance choice made at unit thirty-four, for the same character, possibly for shots that would sit next to each other in the finished film. The only thing holding that together was two people watching dailies and catching drift before it became expensive.
Andy Brown, the art director, put the change in working method in the simplest terms. On earlier films, if you had more time, departments could take turns. Modelling would finish, then set dressing would start, then lighting. Here, everybody was there at once. That sounds like a small scheduling note. It is actually a complete inversion of how a craft-heavy production usually protects itself. Taking turns is how you catch mistakes, because each department inspects what the last one handed over. Working simultaneously removes that safety net, and the only way to replace it is to front-load the scrutiny.
Which is what they did. Brown described the pre-filming stage as the place where the time went, checking, for instance, that there were no bubbles in the stonework of a miniature wall. Bubbles in stonework. On a set a few feet across, photographed by a camera that will be very close to it, a bubble reads as a hole. Once forty-seven units are shooting, there is no time to go back and fix a wall, so the wall has to be right before anyone turns a light on.
The film also gave the studio an excuse to look different, which after years of a television series was apparently a relief. The Shaun series is locked into a single visual key. Brown has described it as permanently the middle of June, the middle of the day, the same light in every episode. A feature set at Halloween unlocks all of it. Oranges. Autumn. Long shadows. A gothic register the franchise has never had access to before, because you cannot do gothic at noon in June.
The trick was making it gothic without making it frightening, and that tension ran through the whole production. Walker has said they wanted some tension and some scariness but not too much, and that early versions of several sequences went too far and had to be pulled back. Their fix was structural rather than cosmetic: any time the film built real tension, they undercut it with a joke, so the audience could laugh and feel safe again. That is a very old piece of craft, and it is the difference between a children’s film with a monster in it and a children’s film that produces a phone call from a parent.
The photography is where the practical obsession really shows. Charles Copping, the director of photography, has described the studio’s standing rule: it has to feel in-camera and believable. The ambition, in his framing, is that the audience should feel they could walk into the laboratory set and find it looking exactly as it does on screen. That is a peculiar thing to be insisting on in 2026, when a convincing mad scientist’s lab can be generated by a computer in an afternoon and nobody would be able to tell. Aardman built one, lit it, and shot it.
The glowing liquids in that lab are a good example of how much effort that costs. Rather than paint a glow on afterwards, the crew added ultraviolet-reactive material to the liquids themselves and shot a separate pass under pure ultraviolet light, so the actual luminescence could be captured and then used by the visual effects team. The lab was also dotted with tiny working lights, including Christmas tree bulbs, placed to sit out of focus in the background and produce soft highlights. That is not nostalgia. It is a real optical effect that a physical light source produces and a simulated one has to imitate.
Copping made one other point that cuts against the whole narrative of stop-motion as a slow and agonised art. Squeezing the schedule, he suggested, is actually part of what gives Shaun its look. The compression produces a feeling of immediacy, something organic and slightly rough that a more comfortable timeline would sand off. Shooting a Shaun feature, in his description, is intuitive. You would expect a director of photography to complain about eight months. He argued the eight months were the point.
It is worth remembering what the puppets themselves involve, because the number of shooting units only means something if you know what is being shot. Aardman’s characters are sculpted by hand, built around armatures, and finished in a modelling material that holds a thumbprint if you are careless. Faces come apart so that mouths and brows can be swapped between frames. A single character in heavy use needs multiples, because one puppet cannot be in two units at once and forty-seven units were running at once. Every one of those copies has to be identical enough that a cut between them does not register. That is an entire department’s worth of work that never appears in a single frame of the finished film and would collapse the whole schedule if it were even slightly wrong.
For scale, the previous Shaun feature, Farmageddon, was in production from late 2017 through the middle of 2019, roughly nineteen months. The new film did it in a bit over a third of that time. It is genuinely faster than nearly everything comparable, and the compression came almost entirely from parallelism rather than from cutting corners on the puppets, which are still built and dressed and moved by hand exactly as they always were.
So this is a film that solved a very hard production problem in a very clever way, and made something that critics loved. It landed at ninety-eight per cent on Rotten Tomatoes across its first wave of reviews, sitting just under the first film’s ninety-nine and comfortably above Farmageddon’s ninety-six. Both of the previous Shaun features were nominated for the Academy Award for Best Animated Feature. On the evidence of the reviews, this one has a case.
And then the thing that actually happened, happened.
It opened in North America to about 1.9 million dollars, across 2,183 cinemas. That is roughly 872 dollars per screen. It is, by the trade’s own accounting, the weakest North American opening for a new Aardman feature released in more than two thousand cinemas. The first Shaun film opened to around four million in 2015 from a similar number of screens, which makes this debut a little over half of it. Aardman’s previous low point was Early Man in 2018, at 3.2 million. Even Missing Link, the Laika film that became a byword for a theatrical disappointment, opened to 5.9 million, which is more than three times what Mossy Bottom took.
There is no comfortable way to write that next to everything above it. Forty-seven units. Nearly fifty sets. Ultraviolet passes and Christmas tree bulbs and a man checking miniature stonework for bubbles. Two directors watching dailies every morning for eight months so that forty-seven separate films would turn out to be one. Ninety-eight per cent. And then a per-screen average that would embarrass a re-release.
It is also not the whole picture, and it would be lazy to pretend it is. Shaun has never been primarily an American property. Farmageddon never had a proper American cinema run at all; it was redirected to streaming, and still finished with around 43 million dollars worldwide off a 25 million dollar budget, with its strongest business in the United Kingdom, Germany and France. Family animation now does the bulk of its work over months rather than a weekend, and a franchise built on a character with no dialogue travels further and lasts longer than an opening weekend number suggests. This film has the rest of the world, a Halloween setting with an annual second life, and a streaming afterlife that will almost certainly outperform the cinema run by a wide margin.
But the gap between the craft and the opening is real, and it is the honest ending to this story rather than a tidy one. The people who made this film went to extraordinary lengths to do something the hard way on purpose, in a year when doing it the easy way has never been easier or cheaper. They compressed a multi-year process into eight months without giving up a single frame of physical reality. They built the lab instead of rendering it, because they wanted you to believe you could walk into it.
Most of the audience will meet that work on a television screen, at home, on a weekday, without ever knowing that the flickering light behind the mad scientist’s bench was a real bulb the size of a grain of rice, placed there by hand by somebody who cared whether it was in focus. The work is in the film whether or not anyone counts the crews. Forty-seven of them, all shooting at once, so that a sheep who never speaks could be scared of something in the woods.