Download complete editable project · Read review limits

Reveal two pip-anchored identities on staggered gates

Manual construction instructions, not a supplied or executed Resolve project. Native Resolve validation and source-quality approval are unverified. The two variants are one example. Use the original SVG geometry and editable text; the attributed source clip is research only, never a Loader input.

Clock and canvas

Create a1920×1080 Fusion composition with109 frames, local0..108 inclusive. Use24000/1001 fps, commonly labeled23.976, not24. Source mapping is sourceFrame = localFrame + 562. Preserve1001/24000 seconds per frame. If an existing timeline uses another cadence, make a matching timeline rather than changing this shot's speed. Export silent video; do not add narration.

docs/controls.csv contains every native-frame control. Times below are local integer frames. Pip colors and text substitutions are stepped; do not spline through them. Only the small settling translations and context drift interpolate.

Editable branches

1. Build an opaque room from public/art/studio-room.svg or garden-room.svg. Preserve its vector groups, paths and gradient as editable shapes.
2. Add the matching tall figure at pixel(960,334), size620×746, then the small figure at(462,551), size340×400. Their original silhouettes replace the source person and pet. Do not use source screenshots or flattened film frames.
3. Add two32px circle pips: right center(1172,277), left center(631,475).
4. Add four independent Text+ branches above the figures and room. Use the bundled Geist font, white text and the selected input JSON. Keep each text branch independent from its pip so their off-gates can differ.
5. Merge all branches over the room. No camera cut, fade or added outro occurs in this109-frame example. The last two frames retain only the original scene.

Importing an SVG as a flattened image is a comparison aid, not an editable equivalent. Rebuild/import paths into native shape groups and preserve the generator. SVG gradient and Text+ metric parity require application calibration.

Text boxes

Positions use top-left pixel coordinates. Convert a pixel point(x,y) to Fusion Center(x/1920,1-y/1080). Convert rectangle width/height by dividing by1920/1080. The listed text boxes include full width, not the current revealed prefix width.

Use line height1.15; authored tracking is−0.7px for names,−0.2px for roles. Native Text+ tracking is not assumed to use the same units; calibrate rendered widths against the shipped frame. Do not auto-shrink text to fit. The Remotion renderer checks actual bundled-font widths and rejects oversize content.

Pip steps

Right: hidden0; yellow1..3; blue4..6; orange7..9; violet10..12; pink13..15; green16..18; white19..92; hidden93..108. Left repeats the same schedule eight frames later: first yellow9, white27..100, hidden101..108.

The authored palette is yellow#f5da62, blue#64afe8, orange#f5a26a, violet#b2a0ea, pink#eba6c0, green#8cc7a1, white#ffffff. These are adaptation tokens, not measured official production colors. No opacity or hue interpolation.

Name substitutions

Right name: hidden0..18; first word19..23; complete24..94; surname95..98; hidden99..108. Split at the first space; additional surname words stay together. Keep the removed first word's horizontal advance invisible so the surname stays at its full-name position. Do not reflow the surname to the left box edge. Left name: hidden0..26; first three characters27..31; complete32..104; hidden105..108. The source's left name was five characters; these original four-character names preserve the observed prefix/complete gates, not its copy.

Role reveal and exit

Right role starts at23 with0.4 of the full string and holds through28. On frames 29..39 use ratios14/30,16/30,18/30,21/30,22/30,24/30,26/30,27/30,28/30, 29/30,1 respectively. At99 hide the leading category and & but preserve its horizontal advance, retaining the suffix at its prior position; at101 hide the role.

Left role starts31 with0.6 of its string and holds through36. Ratios change at 37→13/20,38→14/20,40→15/20,41→16/20,42→18/20,44→19/20,46→1. At105 remove the first word and its following space, retaining the suffix; at107 hide the role. Both role branches show the prefix containing max(1,floor(fullLength*ratio+0.00000001)) characters when the ratio is positive. The role ratios scale original copy to its field; source glyph shapes are not reproduced. Use stepped text states or explicit masks, not an even typewriter.

Independent continuous controls

Room x is−10frame/108, with rendered width1940 and height1080. Tall figure x is960+0.6roomX; small figure x is462+0.4roomX. This shallow drift is authored. Right text x offset is6max(0,1−max(0,frame−19)/5); left text offset is −6*max(0,1−max(0,frame−27)/5). Pips remain fixed. These small linear settling paths approximate the observed relationship, not recovered source trajectories.

Inspect the result

Check the right/left color stagger, role-prefix plateaus and independent exits. At93 the right pip is absent but its full name remains;95 has only the surname. At99 the right name is gone but its role suffix remains. At101 the right side is empty and the left pip is gone; the left name/role still remain. At105 only the left role suffix remains. Frames107 and108 have no pips or labels.

Use the actual rendered variants and manifest checks as mechanical comparisons. They are not proof of native Fusion behavior, independent moving review or source fidelity. The original setting, figures, font metrics and content are deliberate adaptation differences.