Skip to content
refs
REF-2174Source onlineBlueprint rev 01

A humanoid robot tackles two everyday tasks

A robotics launch film builds trust through long, inspectable tile and lunch-bag demonstrations after a silent moving title.

Maker @GoogleDeepMindFor Google DeepMindOriginal post ↗
Cut timeline · 17 shots01:22
00:0000:1500:3000:4501:0001:15

Field notes / why it works

Walking feet and a staged white title establish the subject before any speech at 0:12.6. A 19.33-second overhead tile sequence makes the first claim inspectable, while a new bag-and-fruit task resets attention around 0:50; replay peaks at 0:52.5–0:53.3 and 0:59.0–0:59.9. The final URL waits until about 1:15, after both demonstrations.

01

Plate 01 / The format in one breath

A short announcement names a new technology, then lets a physical demonstration carry the claim. A first, slower task shows comprehension and precise action; a second, more familiar task shows transfer to everyday use. A minimal information card appears before the demonstrations, with the web address saved for the end.

Why it works

  • A moving robot foot and a large title establish the subject by 0:03 without needing narration; speech only starts at 0:12.6.
  • The tile challenge opens at 0:21.5, and the long overhead hold from 0:30.6–0:49.9 lets the viewer verify the resulting arrangement rather than taking a claim on trust.
  • The lunch-bag task resets the visual environment at 0:50; replay peaks at 0:52.5–0:53.3 and 0:59.0–0:59.9 coincide with the new objects and manipulation.
02

Plate 02 / Format card

Platform · aspect · lengthYouTube · 16:9 · 81.5 s · source measured at 640×360, 29.97 fps
Pace17 shots, 16 cuts, 11.8 cuts/min; 15 internal picture changes, 22.8 picture changes/min; mean shot 4.79 s, median 3.0 s
Script88 spoken words; 93 wpm over the whole film, about 236 wpm while speech is active; 11 short utterances, five questions
VoiceSeveral conversational archetypes: calm human task giver, clear machine-like responder, and brief second human request; dialogue rather than narration (speaker assignment inferred from images and transcript)
On screenAbout 90% live action demonstration and reaction, 10% title, announcement, and end-card treatment (visual estimate); side views alternate with overhead proof
CaptionsNo running subtitles; large white headline/announcement text only, plus tiny black pill labels during demos
SoundLow continuous tonal/music-like bed, approximately 134 BPM (rough estimate); dialogue and real task sounds carry the middle; no clear cut-synced effects
StructureTitle → partnership claim → tile challenge and completion → bag-packing challenge and completion → walking image → URL card

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