You balance the camera on a stack of books and record your best take. On playback, the top edge cuts through your forehead. You try again from farther away, but your face becomes a dot in the middle of the living room. On take three, your eyes point down because the lens is sitting on a table. Your acting has not changed: the frame is making casting search for your eyes.
The short answer: for clear self-tape framing, shoot landscape, put the lens at eye level, and use a chest-up shot that shows your head, shoulders, and upper torso. Leave a small amount of space above your hair, place the reader beside the camera, and record ten seconds with a real gesture. Fix the picture before you perform the whole scene.
Choose a chest-up shot, not a face lost in the room
A chest-up shot keeps your face large enough to read a silence while giving your shoulders and upper torso room. The screen should show the top of your head, both shoulders, and part of your chest. The coffee table, radiator, and half the ceiling do not belong in that frame.
Shoot too wide and a raised eyebrow disappears like a gesture at the end of a corridor. Shoot too tight and one small lean toward your partner cuts off your chin or hair. Start chest-up, then change it only when the casting instructions request another shot. Their brief always comes first.
Turn the camera to landscape. The validated sources support this format and a chest-up composition for a standard self-tape. They do not give one universal actor-to-lens distance because cameras and rooms differ. Put tape under your shoes, check the video preview, and move back a few inches at a time until both shoulders fit without making the living room the main character.
Raise the lens to eye level and keep it still
A camera on a table looks up under your chin. A camera balanced on a high cupboard presses your forehead and shoulders down. Put the center of the lens on a horizontal line with your eyes. A tripod is useful, but a steady shelf and a rigid box can support the camera too, provided nothing shifts during the take.
Sit or stand before you set the height. If you align the lens while standing and then act from a chair, it suddenly looks down at you. Mark one tripod foot with tape and mark your shoes or the chair legs. You can rebuild the same geometry after a break.
Put your reader close to the lens, slightly to one side and level with your gaze. Their eyes become a visible target just off axis. When you rehearse alone, a small paper marker can hold that spot. This is not an absolute eyeline rule: follow the casting direction if they ask for direct address or a different position.
Leave simple headroom above your hair
Keep a little air between your hair and the top edge. Too much empty space makes you sink toward the bottom of the picture. No space turns the frame into a low ceiling pressing on your skull. Two finger widths on the preview can be a starting point, but the sources do not present that measurement as a universal standard.
Check the headroom while moving. Turn toward the reader, glance down at an imaginary envelope, then return to your target. If your hair leaves the frame with the first breath or your shoulders vanish when you lean, widen the shot slightly. The right frame holds the action written in this scene, not every movement you could ever make.
Clean all four edges too. A door handle poking in from the left or a black cable hanging on the right pulls the eye. Run a finger around the preview as if you were inspecting a printed photograph. For the rest of the setup, use the guide to home self-tape lighting.
Approve the frame on video, not on a still image
A still can look tidy while your first line takes you out of shot. Record ten seconds at the energy you plan to use. Breathe, turn to the reader, and perform the biggest action in the scene, such as closing a jacket or handing over a sheet of paper.
Watch three things: do both eyes remain visible, does the top of your head keep its margin, and do your shoulders stay in frame? Change one variable between clips. If you alter height, distance, and your mark together, the three pieces of tape on the floor tell you nothing.
Framing supports the performance. It does not replace it. Once the camera is locked, stop monitoring your image during the take. Watch the reader, receive the cue, and let the frame do its job. To work on that connection, read how to nail your self-tape alone.
Three practical exercises
Exercise 1: the eye line (5 min)
- Goal: find a neutral camera height you can reproduce.
- Steps: stand on a floor mark, then raise or lower the lens until it faces your eyes. Record five seconds, lower the camera by one hand width, and record again. Compare your chin, forehead, and shoulder line.
- Result: you can see which height neither looks down nor up at your face.
Exercise 2: the four-edge check (7 min)
- Goal: build a clean chest-up shot with modest headroom.
- Steps: trace all four screen edges with your eyes. Remove every cut-off object, then record ten seconds while turning your head left and right. Adjust only the distance between you and the camera.
- Result: your head stays whole, your shoulders read clearly, and no object distracts at the edge.
Exercise 3: the widest gesture (10 min)
- Goal: make sure the frame holds the physical action in the scene.
- Steps: choose the widest action in the sides, such as offering an envelope. Perform it three times in the same chest-up shot. If the hand disappears, reduce the gesture or widen the shot slightly, then repeat without moving the reader.
- Result: you know the visible limits and can act without checking the screen.
How to use La Réplique to test self-tape framing
La Réplique runs in your browser with nothing to install. Paste the script or add a PDF, choose your role, and put the computer near the camera.
- Test the shot while playing the scene. Neural voices read the other roles while you turn, breathe, and move for real inside the frame.
- Keep your gaze available. Speech recognition validates your lines in real time, progressive hiding removes the text, and the automatic prompter helps when you get stuck.
- Record in the self-tape studio. You can act to camera while the AI reads the cue, then create a shareable casting link and track opens and viewing progress.
La Réplique is a preparation tool. It does not replace a human scene partner or coach.
Common mistakes
- Putting the camera below your face. The table fills the lower edge and the lens looks under your chin. Raise it to your eye line.
- Leaving a ceiling above your head. Your face floats at the bottom of a large empty rectangle. Tighten the shot and keep only a small margin over your hair.
- Reframing every take. The tripod feet move and continuity disappears. Mark the camera, chair, and actor position before you perform.
In summary
Useful self-tape framing has four visible anchors: landscape, a chest-up shot, an eye-level lens, and modest headroom. Put the reader close to the axis, mark every position, and test the widest gesture for ten seconds. When casting specifies a shot, their instruction overrides this starting setup. Open your scene in La Réplique and lock the frame before the first full take.
FAQ
What is the best framing for a self-tape?
A landscape chest-up shot is a useful starting point because it shows your face, shoulders, and upper torso clearly. Change it when casting requests another composition.
How high should the camera be for a self-tape?
Put the lens at your eye level in the position you will act from. Check the angle on a short recording, not only on a still preview.
How much headroom should a self-tape have?
Leave a small visible margin without creating a large empty area. Two finger widths can be a practical preview marker, but test it with your actual movement.
Should a self-tape be portrait or landscape?
Use landscape for a standard self-tape unless casting says otherwise. Always read the submission instructions before setting the camera.
Where should the reader stand for a self-tape?
Put the reader just beside the lens and level with your gaze. You appear to speak to a person without staring into the camera, unless the scene calls for direct address.