OpenAI Film Fieldnotes

Download complete editable project · Read review limits

Populate the deck before the selected slide

Draft construction guide. Native Resolve operation and moving-source fidelity are not verified. This is one candidate, with two editable content variants.

Timing and source

Official film identifier: ECkHTfvf2e8. Source interval [2100,2310), 30/1 fps. Create a silent 1920×1080 composition with 210 included frames, local 0..209. Source ordinal=2100+local frame. Local timeline duration=7.000 seconds. Do not substitute 24/25/29.97 fps or append a closing hold. The hash-bound preserved research excerpt is 960×540; native timing comes from its historical source ledger, not a guessed rate or fresh full-master decode.

Inputs and setup

Unpack the ZIP and run python3 verify_package.py before changing anything. project/ is a complete TypeScript/Remotion project. inputs/default.json and inputs/adaptation.json hold the two invented content sets. The bundled Geist font includes its license. Install it for Text+ and calibrate pixel sizes and baselines against the rendered candidate. The chosen palette is original, not an official brand token. No source movie, narration or source screenshot belongs in the editable composition.

node --experimental-strip-types scripts/controls.mjs regenerates docs/controls.csv from the same integer-frame model as Remotion. It contains one row per included frame and only this treatment's animated controls. For Fusion use global frame=composition start+CSV frame. A scalar step is a held key, not a default spline. For a continuous value, either key every supplied frame or recreate the declared smoothstep and compare all rows. Apply each local transform before its parent. Native Resolve import/export and pixel parity remain untested.

Fusion layers and geometry

Work in a 1920×1080 square-pixel domain. Rebuild rectangles, Text+, masks and original SVG paths as editable nodes, not a flattened screenshot.

Group Pixel geometry Ownership
Window x=145, y=55, w=1630, h=970, r=16 Fixed for the whole interval
Toolbar window-local y=48, h=103 Fixed text and controls
Thumbnail rail x=0, y=151, w=267, h=785 Fixed, clipped by Window
Six thumbnail slots x=30, y=175+127×index, w=212, h=118 Selected slot is index 0
Main workspace x=267, y=151, w=1000, h=785 Fixed background
Selected slide x=310, y=278+slideOffsetY, w=900, h=500 Local Y changes at commit
Assistant panel x=1267, y=151, w=363, h=785 Independent history gate
  1. The opening already contains three populated later thumbnails. Keep selected thumbnail 0 and the main slide as placeholders; do not animate those earlier slides being made. At local 34, expose two additional thumbnail bodies.
  2. History gates use stages 0,1,2,3 at local 0,35,68,164. These replace complete text states without synthesizing a cursor or implying real tool latency.
  3. Keep the selected slide blank through 105. At 106 switch its placeholder and the selected thumbnail to the same formatted-slide subtree. Hard switch: no dissolve, scale reveal or staggered paragraph construction.
  4. The selected slide's local Y simultaneously changes by −72 px. The toolbar, thumbnail rail and assistant remain in their fixed coordinate systems.
  5. Rebuild Slide once at 900×500: white field; accent strip(30,34,7,428); kicker (60,38),18 px; title(60,86),518 px box,38 px/1.28; subtitle(60,238),510 px box,21 px; body(60,368),510 px box,17 px. Side panel(625,32,245,436), its heading(652,64) and three rows y=158,238,318. The thumbnail uses the same subtree at scale.232.
  6. Gate the placeholder: dashed box(100,90,700,185), subtitle box(100,275,700,100), title(120,206),42 px, subtitle(120,300),21 px. These belong to the slide group, not the window. Main content remains held after commit while history changes.

The 34/106 boundaries and history gates are authored sample-constrained choices; they are not newly measured exact source event frames. The source's full ribbon, paragraph details, history scrolling and precise transient camera motion differ.

Verification and limits

Compare both actual Remotion exports, not just their manifests. Check the first and final included frames, event brackets and clipping domains. The source movie is a research comparison only. The original artwork and this schematic UI do not demonstrate real product behavior. Mechanical passes do not close the published source gaps. No native .drp, automatic Resolve import or performance claim is supplied. All timing interpolation between sampled anchors is authored.