# Hard cut to a close-up, then a dot, a mono name and a bracketed role type on beside the speaker

Manual DaVinci Resolve / Fusion construction guide generated from the recipe. It is not a tested .drp.
The accompanying Remotion project (recipe.json) is the executable definition; the numbers below are copied from it.
Neither construction establishes source fidelity: this is an original vector approximation of an inspected editing moment.

## Source and clock
Reference: OpenAI's "We're giving 100,000 academic researchers free access to our frontier models" (MLehRytu9Zo), published 2026-07-29.
Source interval [512, 572) on the 24000/1001 fps master: 60 frames, 2.502500 s. Local frame 0 = source frame 512.
Create a 1920x1080 progressive timeline at 24000/1001 fps and a Fusion composition exactly 60 frames long. Keep integer frames; do not round the frame rate.
Background field: #b9b7ad. Export silent H.264 yuv420p to match the intended playback container.

## Observation
A three-person wide shot holds for five frames, then hard-cuts to a single-speaker close-up (~f5). Two frames after the cut a small white dot appears mid-frame to the right of the speaker (~f7). A monospaced name types on to the right of the dot (~f9–f13), then a two-line bracketed role types on beneath it, indented under the name (~f13–f23). The block holds for the rest of the window.

## Mechanism
Mid-frame identifier build: cut, anchor dot, monospaced name type-on, then a bracketed two-line role type-on indented under the name.

## Phases (local frames)
- 0: Wide shot
- 5: Hard cut to close-up
- 7: Anchor dot appears
- 9: Name types on
- 13: Bracketed role types on
- 23: Hold

## Layer inventory (absolute 1920x1080 top-left pixels, back to front)
- wide-bg [rect] at (0, 0) size 1920x1080 fill #b9b7ad visible [0, 5)
- wide-window [rect] at (560, 0) size 800x520 fill #e9ebe4 visible [0, 5)
- wide-table [rect] at (0, 700) size 1920x380 fill #8f7b62 visible [0, 5)
- close-bg [rect] at (0, 0) size 1920x1080 fill #a99d88 visible [5, 60)
- close-board [rect] at (0, 0) size 640x720 fill #2e3d3c visible [5, 60)
- close-counter [rect] at (0, 760) size 1920x320 fill #c9b89c visible [5, 60)
- close-mug [ellipse] at (520, 780) size 150x190 fill #dfe4de visible [5, 60)
- dot [ellipse] at (1162, 634) size 12x12 fill #ffffff visible [7, 60)
- name [text] at (1186, 626) text "Mara Quint" 24px weight 700 #ffffff reveal chars frames 9-13 visible [9, 60)
- role [text] at (1216, 660) text "[Computational / Biologist]" 22px weight 400 #ffffff reveal chars frames 13-23 visible [13, 60)

## Construction steps
1. Timeline 1920x1080 at 23.976 fps; Fusion composition 60 frames. Local 0 = master frame 512.
2. Edit: wide shot for frames 0–4, hard cut to the close-up at frame 5 (no transition).
3. Dot: 12px white ellipse at (1162,634), visible from frame 7. In Fusion use a Background + Ellipse mask; in the Edit page a shape generator.
4. Name: Text+ monospaced 24px bold white at (1186,626); Write On 0→1 over frames 9–13 in character mode.
5. Role: Text+ monospaced 22px regular, two lines '[Computational' / 'Biologist]' with 28px line spacing, indented to x=1216; Write On over frames 13–23 starting only after the name has finished.
6. Easy mistake: the identifier sits mid-frame beside the speaker, not in the lower-third band, and the role is indented one glyph further than the name.

## Content variants
Default copy:
- name: Mara Quint
- role: [Computational / Biologist]
Adaptation changes:
- name: Idris Vale
- role: [Marine / Geologist]

## Limits
Easing names map to Fusion spline shapes: linear = linear, in/out/inOut = ease curves on the same frames, hold = step, spring/overshoot = slight overshoot past the target before settling.
Text reveal frames give the first and last unit; distribute characters/words evenly between them. Cursor click rings last 14 frames.
Vector illustration is an approximation of layout and motion, not the product UI or recovered production assets.