Skip to content
refs
REF-1578Source onlineBlueprint rev 01

Lapse: a monochrome AI memory film

A 107-second AI art film pairs rapid black-and-white image collage with sparse narration and long wordless passages.

Maker @yzavokuOriginal post ↗
Cut timeline · 60 shots01:47
00:0000:1500:3000:4501:0001:1501:3001:45

Field notes / why it works

The first image changes at 0:00.625, and 14 cuts in the opening ten seconds build curiosity without text or speech. A low narrator enters only at 0:12.9; roughly 15-second and 13-second vocal gaps later let recurring solitary figures, light openings, and eclipse images carry the emotion before a final refrain near 1:40.

01

Plate 01 / The format in one breath

A 107-second, 4:3 black-and-white art film opens with a barrage of eerie images before a low, spare narrator begins at 0:12.9. The voice circles an unsettled feeling while figures, empty places, eyes, light sources, and planetary images recur; the film leaves a question hanging and ends on a restrained title card. Its promise is emotional and visual rather than informational (inferred).

Why it works

  • The first cut lands at 0:00.625; 14 cuts in the first ten seconds create an immediate sensory puzzle before any speech.
  • Visual motifs recur across a wide scale: a lone figure at 0:03.6 and 1:17.7, a doorway/light beam at 0:00.5 and 0:13.3, then an eclipse at 0:07.7 and 1:28.6. This gives an abstract monologue continuity (inferred).
  • Two long vocal gaps, roughly 0:55.9–1:10.9 and 1:27.0–1:40.4, let images carry the story and make the repeated closing admission land harder (inferred).
02

Plate 02 / Format card

Platform · aspect · lengthYouTube · 4:3 · 1:46.9 · measured copy 480×360 at 24 fps
Pace60 shots, 59 cuts, 33.1 cuts/min; 21 changes within shots, 44.9 picture changes/min; mean shot 1.78 s, median 1.45 s
Script70 spoken words; 47 wpm across the full film, 89 wpm while speaking; seven short sentences, no questions or second-person address
VoiceOne low-register, measured, introspective narrator; synthetic narration is credited in the post
On screenAI-generated/collaged monochrome human portraits, solitary silhouettes, empty architecture, nature, cosmic imagery, and a few title graphics; no presenter
CaptionsNone; the end card is the only readable film text visible in the shot sheets
SoundContinuous tonal music-like bed in the measurement; cinematic/ominous feel is inferred; narration begins at 0:12.9; −18.7 LUFS measured integrated
StructureSilent image hook → first-person meditation → long image-only passage → uncertain refrain → cosmic image sequence → title card; no call to action

Pro / 13 plates + recipe

The first plates are open. The full dissection is Pro.

Get the timed shots, type and sound specs, and the runnable recipe.

Open the full blueprint
  1. Plate 03 / The hook, frame by frame (first 3 seconds)Pro
  2. Plate 04 / Structure (beat sheet)Pro
  3. Plate 05 / Shot grammarPro
  4. Plate 06 / Visual style guidePro
  5. Plate 07 / Voice and deliveryPro
  6. Plate 08 / Sound designPro
  7. Plate 09 / PackagingPro
  8. Plate 10 / Script templatePro
  9. Plate 11 / Production planPro
  10. Plate 12 / PromptsPro
  11. Plate 13 / Edit recipePro
  12. Plate 14 / QA checklistPro
  13. Plate 15 / Make it yoursPro
  14. Recipe / runnable promptPro