Technique / ObjectContext

Carry identity across a container cut

Retain the same request when moving from isolated input to device context.

Run the original

5.8 s · 30 fps · silent

Original neutral assets · 5.8 seconds · phase values are chosen defaults, not source timings.

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

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

Official source / inspected interval

Find the perfect gift by shopping with ChatGPT

Watch the official OpenAI upload ↗ · Published 2026-03-27T13:55:13-07:00

Inspected 7.5–11.5 s. Consecutive subset: 8–8.7 s, 42 frames. Dates come from the official watch-page publication field.

Last isolated composer frame at source 8.116667 seconds — 8.116667 s · Last isolated composer frame

First phone frame: no continuous morph at source 8.133333 seconds — 8.133333 s · First phone frame: no continuous morph

Request begins returning within the phone at source 8.216667 seconds — 8.216667 s · Request begins returning within the phone

Same request now readable in its new context at source 8.300000 seconds — 8.300000 s · Same request now readable in its new context

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

What changes across the edit

Shot boundary: Source frame 487 at 8.116667 s shows isolated input; frame 488 at 8.133333 s shows an almost-empty phone. No geometric morph connects them.
Framing: Readable isolated composer gives way to a smaller request inside a tall phone on a colored field.
Layer ownership: White field and composer are replaced by phone and background; the request is introduced within the phone after the cut.
Motion: Phone establishes with changing scale/position, then the request grows into view. The same sentence supplies semantic continuity.
Typography: The same text changes scale and line breaks when it returns; identity does not imply fixed coordinates.
Action order: Submitted-looking input → cut → briefly empty phone → same request → later response.
Pacing & readable holds: The native frames show request content returning from about 8.216667 s and becoming complete by roughly 8.300 s. No exact product latency is implied.

Recreation versus observation

Matches

Hard container cut, a brief empty-device interval, then the same request returns.

Differs

Original artwork is added as a second continuity cue. Four output frames of empty device and the 30-frame settle interval are chosen, slower than the source.

Measured

Cut pair 8.116667 / 8.133333 s, frames 487 / 488.

Inferred

Semantic handoff rather than pixel-matched object motion.

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": 30,
  "change": 54,
  "settle": 84,
  "hold": 90,
  "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