=== REFERENCE MAP === [LOCATION REF] → the lakeside park scene: the woman on the gravel path under o
=== REFERENCE MAP === [LOCATION REF] → the lakeside park scene: the woman on the gravel path under overhanging tree branches, turned back over her shoulder, low golden sun burning through the canopy directly behind her and rim-lighting her hair, the calm lake glittering with sun path, the small white sailboat, the wooden pier, the hazy mountain ridge across the water, park-goers walking the path and sitting on the grass. This is the exact first frame of the clip and the ONLY light, colour and camera look permitted for the entire take. [WOMAN REF] → face and identity only. Eyes bare, no glasses. [MAN REF] → face and identity only. Eyes bare, no glasses. === CHARACTER ELEMENT === [MAN] → man, mid to late 20s, olive tan skin, dark brown hair styled back off the forehead, thick dark eyebrows, warm brown eyes, short trimmed beard, strong jawline, easy natural smile. Identity from [MAN REF] only. Wardrobe: heather grey ribbed long sleeve top under an open olive green field jacket, sleeves pushed up, straight leg olive cargo trousers, plain leather strap watch. He does not film and never holds the camera. He comes from the right side of the path and goes after her to the left. PART 1 - SHOT BREAKDOWN (14s, one continuous handheld take) MOMENT (0.0-0.5s) - Frame 0.0 is [LOCATION REF] EFFECT: none, plain handheld, unsteadiness already present in the first frame Frame 0.0 is exactly [LOCATION REF]: same path, same framing, same golden backlight, same crop. The take grows straight out of it with no settle in. She is already mid step, body angled away down the path, head turned back over her right shoulder into the lens, her left hand up holding the bag strap at her shoulder exactly as in the image. On the first step her right hand comes loose from the strap and drops to her side, becoming her talking hand for the rest of the clip. Behind her the park continues unposed: two people walking the gravel path away from camera, small groups seated on the grass along the lake edge, the sailboat drifting slowly right to left on the water. Camera is a viewpoint only, handheld at chest height, held by someone walking who cannot hold it steady. Constant drift, small dips, late corrections. Never still. Ambient: her boots on loose gravel, water lapping at the shore, indistinct distant park chatter, leaves rustling overhead. MOMENT (0.5-4.8s) - Line 1a, walking, drifting left EFFECT: reactive handheld pan (late, to the left) + frame-edge clip She turns forward and walks, glancing back into the lens, and says clearly and evenly, every word distinct and easy to make out, in a quiet unbothered American English at natural conversational speed: "You see videos like this and assume there's some crazy prompt behind them." Delivery: relaxed, a little knowing, not rushed. Diction stays crisp throughout, no mumbling, no words swallowed, the line fully intelligible start to finish even with the natural conversational rhythm. A small tilt of her head on "VIDEOS LIKE THIS." Her right hand lifts and gestures loosely back toward the path and the water behind her on "SOME CRAZY PROMPT," palm open, then drops. Her eyes land on the lens once mid-line, casually, and slide off again toward where she's walking. She blinks naturally once. Her walk drifts further left than the camera expects, her hips carrying the drift before her feet do. The camera swings left a beat late, so for roughly half a second she clips the left edge of frame before the pan catches her. Her hair moves in loose strands, the low sun lighting the flyaways like filament. Ambient: her voice at open-air distance, gravel, a walker passing behind, water, breeze. MOMENT (4.8-5.6s) - Line 1b, "There isn't." EFFECT: none, plain handheld, she stays centred in frame Still walking, she says, clearly and simply, two distinct words with a small confident pause before them: "There isn't." Delivery: flat, certain, almost amused, no drama. A small shake of her head, barely there, on "isn't." She holds eye contact with the lens for this line only, briefly, before looking away again. Ambient: her voice, gravel, water. MOMENT (5.6-6.4s) - She turns back, focus catches her, focus hunt 1 EFFECT: reactive handheld pan (catch up, to the right) + autofocus hunt-and-lock (about half a second) She pivots on the ball of her left foot, turning back to the right, front opening toward the camera. The camera corrects across to find her and the autofocus hunts and catches her briefly, a soft breathing of sharpness across her face and the gravel. Silent, in-lens only. MOMENT (6.4-7.3s) - The movement: a small candid spin-step EFFECT: none extra, plain handheld following her body, camera reacts naturally to her motion On the turn she doesn't just pivot and stop, her momentum carries her half a step further than a simple turn needs: she takes one light skipping step sideways, plants her back foot, and lets the spin swing her fully around to face forward again in one continuous motion, her hair flaring outward and catching the low sun in a bright arc as it follows the turn, the hem of her cardigan lifting with the movement. It reads as unconscious and spontaneous, a happy little burst of motion in her step, not a performance, not held or repeated, immediately absorbed back into her normal walk. A small breath of a laugh escapes her, barely audible, as the spin finishes and she keeps moving. The camera does not anticipate this, it follows a half beat behind, catching the tail of the motion rather than leading it, staying handheld and unsteady throughout. Ambient: the soft swish of fabric and hair in the turn, her boots resetting on the gravel, her small breathy laugh, water, breeze. MOMENT (7.3-11.0s) - Line 2, three distinct gestures, small tilt toward the light EFFECT: small handheld tilt (toward the sun, and straight back), she stays in frame throughout Still moving, settled back into her walk, she says clearly and deliberately, each word distinct: "The real trick is a simple workflow most creators completely skip." Delivery: matter of fact, confident, unhurried, every word fully enunciated, nothing rushed or slurred together despite the natural flow. A small breath after "trick." A slight smile arrives on "completely skip" and stays. Gesture on "THE REAL TRICK": her right hand comes up, thumb and forefinger close together in a small precise pinch in the air, like isolating a tiny detail, then opens and drops. Gesture on "A SIMPLE WORKFLOW": both hands come up and open outward at waist height, palms turning up and out into the park around her, travelling away from her body and never back toward her chest or face. Gesture on "COMPLETELY SKIP": her right hand makes one small flat horizontal sweep to the side, like waving something past her, unhurried, and drops. As she says "workflow," the camera makes one small handheld tilt up toward the sun burning through the canopy and comes straight back to her, she remains in frame the entire time. Her hair is still settling from the spin, strands lifting and falling naturally across her shoulder as she walks. Ambient: her voice, gravel, water, breeze, distant birds. MOMENT (11.0-11.8s) - The camera follows her, then turns right EFFECT: none, plain handheld, unsteadiness continues She keeps moving off to the left along the path edge, head turning forward toward the lake. Without any cut, the camera turns slightly to the right and almost straightens. Plain hand movement at ordinary speed, no whip, no smear. MOMENT (11.8-13.4s) - [MAN], close, already there EFFECT: autofocus hunt-and-lock on [MAN] (about half a second) [MAN] is already in frame the instant the camera lands, standing on the path with the same low golden sun behind his right shoulder, the lake glittering out of focus behind him. FRAMING: close, roughly 1 metre from the lens. Head, shoulders, chest, arms and waist fill the frame, cut at the waist. He has come from the right side of the path, already mid-stride, on his way after her. No pause, no dead air. The autofocus hunts briefly and catches him in about half a second, silently. He looks into the lens and says clearly, simply, unhurried: "Let me show you mine." Delivery: slight Australian accent, warm, easy, each word distinct and fully audible, quiet confidence, the smallest emphasis on "MINE." A small easy smile, eyes crinkling, one small natural hand gesture, palm turning slightly toward himself on "mine." His jacket collar and pushed-up sleeve shift in the breeze. The low sun rims the edge of his hair and jaw. Ambient: his voice through the same thin phone mic at the same modest level as hers, wind, gravel, water, birds. MOMENT (13.4-14.0s) - He's already going, hard out EFFECT: hard cut on motion (out point, no trailing beat) The smile holds a beat, then he is already moving again, continuing after her to the left, and the clip cuts exactly at that moment, mid motion, mid step. No settle, no hold, no fade, no black. PART 2 - EFFECTS LIST Unsteady handheld, continuous, 0.0-14.0s. Never still. Reactive handheld pan (late), 2x (left ~1.0-4.8s; catch up right ~5.6-6.4s). Frame edge clip, 1x, about half a second, during Line 1a. Candid spin-step movement, 1x (6.4-7.3s), unforced, camera reacting half a beat behind, never anticipated or held. Small handheld tilt toward the sun and back, 1x (on "workflow"). She stays in frame throughout. Autofocus hunt-and-lock on her, 1x, about half a second, at the turn. Autofocus hunt-and-lock on [MAN], 1x, about half a second. Handheld turn right and near straighten, 1x (11.0-11.8s). Plain hand movement, no sound. Hard cut on motion, 1x (14.0s). The only cut in the clip, the out point. NOT USED ANYWHERE: auto-exposure adaptation of any kind, re-metering, contrast shifts, brightness ramps, any change to light or grade, cuts other than the out point, whips, smears, slow motion, speed ramps, digital zoom, punch ins, stabilisation, colour grading, transition effects, whoosh or swish sounds, shutter or click sounds, music, on screen text, palm over lens, camera turning around, any second line from [MAN], any repeat or held pose of the spin-step. PART 3 - EFFECT DENSITY BY TIME 0.0-4.8s = MEDIUM (late pan left, frame edge clip, one gesture, ambient park life) 4.8-6.4s = LOW (short second line, catch-up pan, focus hunt) 6.4-7.3s = HIGH (the spin-step, hair flare, breathy laugh, full audio shift, the single most physical beat in the clip) 7.3-11.0s = MEDIUM (three gestures, small tilt and return) 11.0-14.0s = LOW (turn right, focus catch on him, his line, held smile, hard cut on motion) PART 4 - MOTION FLOW Opening: real time, already moving, half a step behind someone who isn't waiting for you. Build: her drift left and the late pan pull the frame apart, the short punch line lands flat and certain, she turns back and the focus finds her, and right on the turn her momentum spills into one small unconscious spin, hair catching the light in a bright arc, a breath of a laugh, then straight back into her walk without missing a beat. The three gestures carry her second line to the smile on "completely skip," the small tilt lets the sun into the corner without losing her. Resolution: the camera turns right and he is simply already there, caught in passing on his way after her. The smile holds a beat, then he is gone in the same direction she went and the frame cuts on his first step. === HARD RULES === Total duration is exactly 14 seconds. One single continuous handheld take from first frame to last. No cuts, no scene changes, no transitions, no time skips, no slow motion. The only cut is the out point at 14.0s. Frame 0.0 is exactly [LOCATION REF]. The entire take happens in that one location, on that one gravel lakeside path. The camera never leaves this park. THE LIGHT NEVER CHANGES AND THE GRADE NEVER CHANGES. The warm golden backlight from frame 0.0 holds identically across all 14 seconds: low sun stays burning through the canopy behind them, hair stays rim-lit with a warm halo, skin stays warmly lit, the lake stays glittering, the grass stays warm gold-green. No auto-exposure reaction, no re-metering, no highlight recovery, no contrast tightening, no brightness shift of any kind at any moment. The last frame is graded identically to the first. Both lines of dialogue and his line must be spoken clearly and distinctly throughout: every word fully enunciated and intelligible, natural conversational rhythm but never mumbled, never slurred together, never rushed past comprehension. Volume and clarity stay consistent from first word to last. The spin-step movement is physically grounded and controlled: her feet stay in contact with reasonable footing throughout, this is a light skip and turn, not a leap, not airborne, not a choreographed dance move, it reads as a spontaneous candid burst of movement a real person might do without thinking, not a performance for the camera. It happens once, is never repeated, and is never held as a pose. Skin is never smooth. Both her skin and his keep visible pores, fine texture, small unevenness, faint natural freckling across her nose and cheeks, and a natural sheen where the sun catches. No smoothing, no retouching, no waxy or plastic skin, no beauty filter. Visible film grain across the whole frame at all times, warm and fine, in shadow and bright areas alike. The handheld must be obvious throughout. He is walking while filming, one hand, constant drift, small dips, late corrections, horizon never level for long. Every movement has a visible beginning, middle and end. Nothing is already done, nothing appears mid motion, nobody teleports. The tilt on "workflow" is a small handheld movement toward the sun and straight back with her in frame the entire time, not a swing off her. No unnatural camera work anywhere: no zoom, no punch ins, no whips, no smears, no snap movements, no digital effects, no stabilisation, no speed changes. No flash fires, no strobe, no white frame, no photograph is taken, no shutter sound, no camera click. Exactly one autofocus hunt on her and one on him. No other focus changes. The camera filming this is never seen. The person holding it never speaks and never appears. Her left hand holds the bag strap at her shoulder for the entire clip except during the spin, where it may lift naturally with the motion and settle back after; it never swings free or is set down. Her right hand is her talking hand. The hands never swap roles. [MAN] never films and never holds the camera. He is not in frame until the camera turns right at 11.0s. He never touches the camera, never covers the lens. [MAN] is framed close, roughly 1 metre from the lens: head and upper body cut at the waist. His legs are never fully in frame. She goes left. He comes from the right and follows her left. Wind: constant light breeze throughout, lifting loose strands of her hair the whole clip, intensifying naturally during the spin. Background park life continues naturally and unposed throughout: walkers on the path, seated groups on the grass, the sailboat drifting slowly on the lake. Nobody looks at the camera, nobody reacts to her spin, nobody freezes. She never performs for the camera: the camera follows her, she does not adjust for it, her body keeps its own rhythm. Only her eyes come to the lens, landing casually and drifting off, holding briefly only on "There isn't." She blinks naturally, takes small natural breaths, keeps her mouth and brows alive between words. She is never completely still in any frame. She speaks at natural conversational tempo and never drags: all three lines run as flowing, fully intelligible phrases with only small breaths in them, no long gaps between items, finishing a little early rather than late. Her gestures are all clearly different from each other, all candid, never staged, and none of them point at herself. At no point does she gesture toward or indicate her own body, chest or torso. She drifts off centre and is never locked to the middle of the frame. Dialogue is spoken exactly once, exactly as written, in this order: "You see videos like this and assume there's some crazy prompt behind them." / "There isn't." / "The real trick is a simple workflow most creators completely skip." / "Let me show you mine." No added words, no repeats, no fourth line from her, no second line from him. Her lines are hers, his line is his. [MAN] wears no sunglasses, eyes bare and visible. The clip ends the instant [MAN] starts moving after her, mid motion, mid step. No beat after. === AUDIO === Everything is external phone mic sound as captured, raw and unprocessed: thin, a little boxy, light auto gain pumping, handling noise, wind hitting the microphone directly once or twice, no denoising, no polish. No music, no sound effects, no narrator, no on screen text. No sound is attached to any camera movement, the tilt and the turn are silent, the focus hunts are silent. Her voice sits at natural outdoor distance, a few metres from the mic, open and airy with no room reverb, clear and intelligible throughout, never close mic, never dry, never studio. It gets slightly quieter whenever she turns her head away and clearer whenever her face comes back toward the lens. His voice is picked up by the same phone in the breeze, the same thin mic and modest level as hers, clear and fully intelligible, never close mic and never louder or sharper than she was, with wind gusting across it. Sound palette: water lapping gently at the lake edge, a constant light breeze on the mic, distant birds, her boots on loose gravel, the soft swish of fabric and hair during the spin, her small breathy laugh, other walkers' footsteps passing behind, his steps on gravel, indistinct distant park chatter, and, faint and easy to miss somewhere in the middle of the clip, a bicycle bell ringing further off along the path. === VOICE === Both performances must sound fully human and unscripted, never synthesized, and both must be delivered with clear, unmistakable diction: natural pitch variation across each phrase, small breaths audible between clauses, contractions used naturally ("there's," "let me"), a little vocal fry at the ends of phrases, an occasional mouth click, slightly uneven pacing rather than flat even TTS cadence, but never so casual that a word becomes unintelligible. No robotic evenness, no over-articulation, no studio-clean delivery, no mumbling, no swallowed syllables. Her: quiet, silky American accent, relaxed and unbothered, talking to whoever is filming her, not to an audience. Natural conversational speed, each line a flowing, fully clear phrase, finishing a little early rather than late. Line 1a is knowing and easy. "There isn't" lands flat and certain. Line 2 is confident inventory that turns warm and pleased on the last four words. Him: slight Australian accent, natural and friendly, easy grin, quiet certainty, every word distinct. "Let me show you mine" lands unhurried with the smallest weight on "mine," carried by the smile.
Reference Images