Twenty-Four Wishes per Second

SHORT FILM DIRECTED BY ARANTXA BARCIA
MADRID · 2026

Generated shot by shot, graded and sewn by hand

Format:

DIGITAL / 16:9

Type:

SHORT FILM

Production:

MITO Films

Platform:

MITO AI

CRAFT

HISTORY, ZOETROPE TO AI
with Adriana Arratia, script and edit

×9 STAGES

GENERATION, WORST SHOT
Seedance, Grok Imagine, Nano Banana

×300 TAKES

GRADE
DaVinci Resolve, hand-built LUTs

×3 LOOKS

CONTINUITY AND RETOUCH
Photoshop, After Effects

×11 FIXES

THE ROOM
posters and bookshelf, Josefa Toro

×35 TITLES

RESOLUTION, 720P TO 1080P

×2 PASSES

FINAL RUNTIME

9'12"

www.arantxabarcia.com

A DIRECTOR WITH NOTHING TO SAY

falls into a zoetrope and spends

A HUNDRED YEARS OF CINEMA

looking for the reason she started. Nine minutes,

ZOETROPE TO AI,

generated shot by shot and sewn together by hand.

TWENTY-FOUR FRAMES A SECOND,

twenty-four wishes.

“Forget wishes, tell your story.” In the script from the first draft.

TWENTY-FOUR WISHES A SECOND

TWENTY-FOUR FRAMES.

TWENTY-FOUR WISHES.

Cinema runs at twenty-four frames a

second. The frames were never the

hard part. The hard part is wanting

to tell something.

Halfway through, the Tin Man tells

her: forget wishes, tell your story.

The line was in the script from the

first draft and nobody paid it any

attention. When we went looking for

a title, we found it was already

written inside the film. The thesis

was there before we knew it was the

thesis.

TWENTY-FOUR WISHES A SECOND

TWENTY-FOUR FRAMES. TWENTY-FOUR WISHES.

Cinema runs at twenty-four frames a second. The frames were never the hard part. The hard part is wanting to tell something.

Halfway through, the Tin Man tells her: forget wishes, tell your story. The line was in the script from the first draft and nobody paid it any attention. When we went looking for a title, we found it was already written inside the film. The thesis was there before we knew it was the thesis.

1834

2026

EVERY STAGE IN THIS CHAIN WAS, 
IN ITS DAY, THE NEW TOOL 
THAT WAS GOING TO KILL CINEMA.

The zoetrope is from 1834. It is the first machine that made a drawing move. Starting there and ending in AI is not decoration, it is the argument.

Zoetrope

A sequence of drawings, a turning drum. Movement begins in the gaps between still images.

WHAT WAS DONE BY HAND, THAT YOU WILL ASSUME THE MACHINE DID

MODELS MAKE SHOTS. 
THEY DON'T MAKE FILMS. 
EVERYTHING IN BETWEEN 
IS THE WORK.

Nothing in this film was found. Every object in every frame was designed, named and approved before a single shot was generated, and most of them were redesigned two or three times after that.

Eighteen clay puppet characters from the film laid out on a terracotta ground, Aaliyah first and the moon last — Design

Extreme close-up of Aaliyah against a dusk sky, fingerprints and tool marks visible in the clay — In film

We designed Aaliyah before anything else existed, because nine minutes only hold together if she stays the same person. Braided pigtails, coral puffer jacket, the gap in her front teeth: written the same way in every prompt, never paraphrased. Everything is clay and has to look like clay, so when a shot came back smooth, or the gap closed, we generated it again.

The bookshelf in Aaliyah’s room: the zoetrope surrounded by thirty-five invented film and technology titles — Design

A room becomes somebody's room through what they keep in it. We chose the posters one by one, seventy percent of them films directed by women, and rebuilt the bookshelf as a real cinema bibliography, spine by spine.

Nobody will pause the film to read the shelves. But that is the difference between a room that belongs to someone and a set that was assembled.

A model does not understand “orbit around her”. It understands what has to leave the frame, and in what order, so we built the move in 3D first and handed it over as reference.

The title was animated by hand, letter by letter. Hierarchy, spacing and timing are not decisions a tool makes for you.

A MODEL WILL GIVE YOU A SHELF FULL OF BOOKS.

Deciding which books they are

IS STILL DIRECTION.

The space

A model can give you the shot.
It cannot give you the space.

In this outtake, faces distort and a character appears twice. We mapped the Red Room so characters and cameras stayed in the right place.

What worked was cutting the sequence first, as if the space were right, and then generating the shots that cut asked for. The other way round doesn't work.

PROMPT

PERFORMANCE

THERE IS NOBODY TO GIVE THE NOTE TO. 
YOU GIVE IT TO THE TEXT, AND THE TEXT 
CARRIES EVERYTHING A DIRECTION WOULD.

Describing an emotion gets you a neutral face. What works is subtext: where each character is coming from, what they want, what they are hiding.

Aaliyah at her desk with both hands pulling at her hair, screen light on her face — Design

The film opens on someone who has nothing to say and a client who wants it by tomorrow, and that had to land in about four seconds without a line explaining it. Describing an emotion doesn't work. "Frustrated" gets you a neutral face. What works is giving her something physical to do that only a frustrated person would do: hammering the keyboard, the cigarette already lit next to the cereal, the two-finger gesture at a screen that has already hung up. The other half of it is what she is not doing. No sigh, no head in hands, no looking at camera for sympathy. She is annoyed the way people are actually annoyed at work, which is quietly and while doing something else.

Judy alone among the poppies, eyes shut, cigarette in hand — Design

Judy is the one who will not let go. She hugs Aaliyah goodbye and lifts the lighter from her pocket while she does it, smiles about it, and ends up flat on the yellow brick road lighting up with it. None of that is in the dialogue: it is in what her hands are doing.

So the note goes in before generating, and it is not “warm” or “sly”. It is where each of them is coming from, what they want from the other and what they are keeping to themselves. An emotion on its own is too broad; a relationship gives the movement something to respond to.

Aaliyah face to face with the frozen studio boss, both mid-gesture — Design

Aaliyah walks into the animation studio and the camera finds the floor first: rows of stools, and a chain on every ankle. She does not say anything about it. The next shot is her face working out what she is looking at.

Anger written as “angry” comes back as a scowl. What works is giving her something to receive: the shot before decides the performance, and the cut decides how long she has to let it land.

Nobody can be given a note. There is no actor to talk to between takes, no one to pull aside and tell what the scene is really about. So the note goes into the text.

Where each character is coming from, what they want, what they are hiding. Written down, before anything is generated, in the same words every time.

Do that and the performance shows up on its own: how close they stand, who leans, who breaks the hug first. Skip it and you get very good-looking people feeling nothing.

Sewn together

Each shot is generated on its own. We corrected the details by hand to keep the film consistent from one frame to the next.

BeforeComparison 01After

BeforeComparison 02After

BeforeComparison 03After

BeforeComparison 04After

THE CUT

THE FILM WAS FOUND

IN THE EDIT, NOT IN THE SHOTS.

We replaced the opening wide shot

with details of her room: the fan,

posters, window, cereal and ashtray.

The boss's voice arrives before his

face, and hers comes last.

Sequences that added nothing were

cut, keeping the film under ten

minutes.

Three looks were graded by hand in

DaVinci. Skin saturation at 0.15 made

the Red Room feel lit, not filtered.

THE CUT

THE FILM WAS FOUND IN THE EDIT, NOT IN THE SHOTS.

We replaced the opening wide shot with details of her room: the fan, posters, window, cereal and ashtray. The boss's voice arrives before his face, and hers comes last.

Sequences that added nothing were cut, keeping the film under ten minutes.

Three looks were graded by hand in DaVinci. Skin saturation at 0.15 made the Red Room feel lit, not filtered.

Olmo designed the sound. The edit decides when it enters: room tone and keys first, music only when needed. Credits and grain were finished by hand.

We also cut a line explaining the trick. It made the machine sound like the filmmaker. Leaving it unsaid gave the ending more room.

NOTHING HERE

was decided by the tool.

THE FILM IS THE CUT.

9:12

The longest thing we've made

300

Bad versions of one shot before the good one

720

1080

Shots redone when the resolution arrived

The time didn't go into generating. Generating is fast; that was never the problem. It went to three places, none of them the one people imagine.

Continuity. The bag, the coffee maker, the clips, the reel. Sewing together a film generated in pieces that don't know each other.

Resolution. A good part of the film was generated at 720p, because that was what existed when we started. Then 1080 arrived and shots had to be redone. Upscaling saves nothing: when a frame holds a lot of information, a wide shot of a room full of things, the upscaler invents and comes out worse than the original.

Iteration. On one shot, three hundred bad versions before the good one. That is not a figure of speech. Multiply it by every shot in nine minutes and there is the time.

THESE TOOLS ARE VERY GOOD

at doing what you ask them badly.

A BEAUTIFUL SHOT, IMMEDIATELY.

What they don't give you is

THE JUDGEMENT:

whether that shot goes before or after,

WHETHER IT'S SURPLUS,

whether she was carrying the bag,

WHERE THE LIGHT FALLS.

All of that is still

DIRECTING, THE SAME JOB

it always was.

What has changed is where the bottleneck is. It used to be budget and access. Now it's knowing what you want to say, and nobody has made that cheaper.

References

What each world borrows