When a Free Creative Suite Meets a Real Workflow
A fast, satirical tech explainer tests a sweeping replacement-app claim with screen proof, a reversal, and a second experiment.
Field notes / why it works
Speech starts at 0:00, the first cut arrives at 0:01, and the initial claim becomes specific by 0:13. A test reverses the hype at 1:26–1:31, with editor UI proof through 2:34; a new question at 2:34 then carries viewers into the procedure. The 57 shots contain 69 additional picture changes, keeping even long screen demonstrations active.
Plate 01 / The format in one breath
A fast, low-register narrator opens with an apparently dramatic software upheaval, pivots to what a viewer can do with a related tool, then tests the claim on screen. The promised revolution gets undercut by a concrete failure; a second experiment restores forward motion before a sponsor example and quick sign-off.
Why it works
- The first word starts at 0:00, the first cut lands at 0:01, and the first 13 seconds turn a vague industry grievance into a specific “seven apps for free” claim.
- At 1:26–1:31, the narrator reverses the opening promise after actually trying the product; the UI test at 1:47–2:30 shows why, so the criticism has visible proof.
- The new “what if I made my own?” question at 2:34 creates a second plot. Setup, read-only analysis, pseudocode, and a proposed implementation carry it through 4:13.
Plate 02 / Format card
| Platform · aspect · length | YouTube · 16:9 · 5:55 · measured file 640×360 at 23.98 fps |
| Pace | 57 shots, 56 definite cuts; 9.5–10.5 cuts/min including uncertain changes; 21.1 picture changes/min including internal graphic steps; mean shot 6.23 s, median 4.67 s |
| Script | 1,250 spoken words; 212 wpm overall, 226 while talking; 45 sentences averaging 27.8 words; one question; 1.4 “you” per 100 words |
| Voice | One close, quick, dry, low-register male narrator; measured median pitch 113 Hz |
| On screen | Mostly full-frame screen capture, web pages, editor timelines, code, stock/live-action reaction clips, diagrams, memes, and brand cards; no sustained talking head |
| Captions | No continuous subtitles; occasional oversized all-caps headline or joke label, plus source/UI text |
| Sound | Continuous tonal bed inferred from measurements; fast narration; edit accents on many cuts; −21.1 LUFS integrated, −3.0 dBFS true peak |
| Structure | Claim → tool promise → jargon escalation → show open → test and reversal → technical explanation → second experiment → sponsor demonstration → sign-off |
Toolkit for this film: screen recordings of the product; Remotion for motion graphics; ElevenLabs or a recorded voice; Remotion or any video editor. Build the motion and the edit in code; the parts that need other tools come from them. Where the steps below name a single tool, use the toolkit instead.
Create a new 16:9, 1920×1080, 24 fps, 5:55 narrated tech explainer about {{TOPIC}} for {{AUDIENCE}} using Remotion for graphics and ffmpeg to assemble licensed live-action inserts, original screen recordings, narration, and music. Write an original 1,200–1,280-word script with a dry, brisk, low-register narrator archetype; never imitate an identifiable presenter. Open at 0:00 with a recent, sourced {{SURPRISING_CLAIM}} on a black cream-text card; cut to a comic conflict image at 0:01, then show the real {{PRODUCT}} by 0:13. At 0:29 explain why viewers care; by 1:03 promise a test. At 1:14 show the product page, at 1:26 announce the catch, and from 1:47–2:34 show two actual tasks and a fair verdict. Re-hook at 2:34 with {{ALTERNATE_APPROACH}}; demonstrate setup, output, and limits from 3:21–4:13. Use a second application until 4:58, then a clearly disclosed {{SPONSOR_OR_CONCLUSION}} and concise CTA. Target about 57 shots, 9–10 cuts/minute, with cursor actions and crop changes within long UI shots. Use charcoal, cream, red conflict accents, green verified-result accents, and occasional Anton headline cards; skip continuous subtitles. Add a quiet licensed electronic bed and sparse impacts at major pivots. Mark jokes, hypothetical results, and source-backed claims distinctly. Render H.264 MP4 with clear narration and checked loudness.