Technique / TitleCut

Type a premise, then cut to the product

Separate the claim from its visual proof.

Run the original

5 s · 30 fps · silent

Original neutral assets · 5 seconds · phase values are chosen defaults, not source timings.

npm run render -- TitleCut out/TitleCut.mp4 --codec=h264 --muted

Download the complete pinned project · Inputs, phase meanings & full source code

Official source / inspected interval

Introducing ChatGPT Atlas

Watch the official OpenAI upload ↗ · Published 2025-10-21T10:22:07-07:00

Inspected 16.5–20.5 s. Consecutive subset: 19.75–20.05 s, 9 frames. Dates come from the official watch-page publication field.

Title construction replaces the page at source 17.233000 seconds — 17.233000 s · Title construction replaces the page

Complete phrase visible in this sample at source 19.033000 seconds — 19.033000 s · Complete phrase visible in this sample

Last title frame before the cut at source 19.867000 seconds — 19.867000 s · Last title frame before the cut

First browser frame; later scale-up is separate at source 19.900000 seconds — 19.900000 s · First browser frame; later scale-up is separate

Source: OpenAI. Bounded excerpt and annotated stills for the analysis below; not included in the project download.

What changes across the edit

Shot boundary: Adjacent source frames 596 (19.867 s) and 597 (19.900 s) change from title to browser. This is a direct cut, not a dissolve.
Framing: Centered two-line title gives way to a wide browser inside a colored field. The browser continues enlarging after the cut.
Layer ownership: Title on white; then browser, page, cursor and background replace it together.
Motion: Characters accumulate and the text recenters. A formed browser arrives at the cut; subsequent camera scale changes are separate.
Typography: Black, heavy, centered type. Exact font and metrics are not recovered.
Action order: Page context → title construction → complete title → browser proof.
Pacing & readable holds: The complete phrase is present in the 19.033 s sample and through 19.867 s: at least 0.834 s of sampled complete-title visibility, not its exact full hold.

Recreation versus observation

Matches

Growing centered text, complete-title hold and one-frame title-to-window replacement.

Differs

Original wording, neutral fixture and chosen type metrics. The recreation keeps the arriving window fixed instead of continuing the source scale-up.

Measured

Cut pair: 19.867 / 19.900 s, source frames 596 / 597.

Inferred

Reading begins when the complete phrase is available; no comprehension duration was measured.

Chosen defaults

All Remotion phase frames, cubic Bézier (.16,1,.3,1), geometry, colors, font choice and reading holds are chosen defaults. None is a recovered OpenAI specification.

{
  "title": "Make room for the idea.",
  "prompt": "Arrange a small exhibition.",
  "result": "Three studies, ready to compare.",
  "label": "EXHIBITION / 03",
  "accent": "#40685b",
  "action": 0,
  "change": 42,
  "settle": 84,
  "hold": 66,
  "focusX": 0.66,
  "focusY": 0.64,
  "zoom": 1.55
}

Increase hold to extend only the final quiet state. Change the phase frames to change choreography; preserve their ordering. Content limits and the exact meaning of each phase are in the handoff.

Exact original geometry, layer ownership and asset contracts · Schedule against an actual recording