# Fade a name-and-role lower third in after the cut, hold, and fade it out before the next cut

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 "How European Small Businesses Are Using ChatGPT and Codex | OpenAI Academy" (4Nv--b12hxQ), published 2026-07-22.
Source interval [1012, 1131) on the 25/1 fps master: 119 frames, 4.760000 s. Local frame 0 = source frame 1012.
Create a 1920x1080 progressive timeline at 25/1 fps and a Fusion composition exactly 119 frames long. Keep integer frames; do not round the frame rate.
Background field: #5d6a6d. Export silent H.264 yuv420p to match the intended playback container.

## Observation
Six frames of a cool-toned workshop b-roll shot, then a hard cut (~f6) to a warm ballroom interview. At about f10 a bold white name begins to fade in at the lower-left with a smaller role line beneath it; both are fully opaque by f24. The lower third holds through f104, then both lines fade together to zero by f117, leaving the interview frame clean for the next cut.

## Mechanism
A classic interview lower third: a short delayed fade-in after the cut, a long hold, and a fade-out that clears before the shot ends.

## Phases (local frames)
- 0: B-roll shot
- 6: Hard cut to interview
- 10: Name and role fade in (14 frames)
- 24: Hold
- 104: Fade out (13 frames)
- 117: Clean frame

## Layer inventory (absolute 1920x1080 top-left pixels, back to front)
- broll [group] at (0, 0) size 1920x1080 visible [0, 6)
  - broll-bg [rect] at (0, 0) size 1920x1080 fill #5d6a6d
  - broll-table [rect] at (0, 640) size 1920x440 fill #c9ccc8
  - broll-glass [rect] at (1200, 0) size 720x700 fill #8a9aa0
- interview [group] at (0, 0) size 1920x1080 visible [6, 119)
  - int-bg [rect] at (0, 0) size 1920x1080 fill #bdb2a1
  - int-ceiling [rect] at (0, 0) size 1920x260 fill #d8d2c4
  - int-pillar [rect] at (1380, 180) size 120x900 fill #8c7458
  - int-subject [ellipse] at (640, 120) size 560x1200 fill #2b2622
  - int-flowers [ellipse] at (1080, 560) size 420x360 fill #b9c6d6
- lower-third [group] at (60, 880) size 800x120
    keyframes opacity: f10=0, f24=1 (out), f104=1, f117=0 (in)
  - name [text] at (60, 880) text "Rebecca Lane" 42px weight 700 #ffffff
  - role [text] at (60, 936) text "Founder at The Floral Hire" 22px weight 400 #f0f0f0

## Construction steps
1. Timeline 1920x1080 at 25 fps; Fusion composition 119 frames. Local 0 = master frame 1012; the cut to the interview is at local 6.
2. Place the interview clip from frame 6; the six frames before it are the tail of the previous b-roll shot.
3. Lower third: Text+ name 42px weight 700 white at (60,880); Text+ role 22px at (60,936); parent both to one Merge so a single Blend keyframe drives them.
4. Blend keyframes: 0 at frame 10, 1 at frame 24 (ease out), hold, 1 at frame 104, 0 at frame 117 (ease in).
5. Easy mistake: starting the fade-in on the cut frame; the source waits four frames after the cut, and the fade-out completes two frames before the shot ends so no text crosses the next cut.
6. No slide or scale: this lower third is opacity only.

## Content variants
Default copy:
- name: Rebecca Lane
- role: Founder at The Floral Hire
Adaptation changes:
- name: Tomasz Wiśniewski
- role: Owner, Wiśniewski Bakery

## 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.