abstract-value-propositions
abstract-value-propositions
ABSTRACT · VALUE PROPOSITIONS — four claims, each with the thing that proves it in the same frame, in the abstract reel's grammar. A value scene is 5.5–6 s: the line is on screen by 1.2 s and stays, and the picture spends the rest of the shot DOING what the line says. The reel's rule: a claim that is only typed is a caption; a claim the set performs is a proposition.
Read the rig's notes
# scene dur transition device the claim, and its proof
01 v01-one-engine 6.0 match-move iphone ONE ENGINE. EVERY SURFACE — one gradient-glass definition on a sphere, a knot and the phone's shell, re-tinting together
02 v02-fast 5.5 match-move iphone LIVE IN UNDER A SECOND — the screen builds itself in 0.8 s while the camera crashes in on a ramp
04 v04-consistent 6.0 match-move iphone+ipad+laptop THE SAME ON EVERY SCREEN — one emissive scan crosses the family and each screen lights as it passes
05 v05-quiet 5.5 — iphone AND THEN, QUIET — the problem file's pile lifts off the phone and leaves; the key comes up
23.0 s of scene + three gaps (2.4 s) ≈ 25.4 s on the wall. Every scene also
stands alone: cut one out, set transition=none, and it is a value beat.
The fifteen-phone SCALE shot that was v03 lives in the scene-fit library as
projects/parts/v03-scale (its own folder, no world copy, hero on a ui3d stage).
THE PROOFS, and what each one costs:
· v01 — the coda's gradient-glass `house` definition, with its colour
tracks on the DEFINITION's fields, so the sphere, the knot and the
phone re-tint on one clock. A per-element track would opt that element
out of the shared field set and break the demonstration.
· v02 — eight screen cards on a 0.1 s stagger, and a camera push sampled
from a 0 · 0.06 · 0.62 · 1 table, so most of the travel is early and
the last 300 mm take 2 s: the screen is done before the camera is.
· v04 — a 6 × 600 mm emissive bar crossing 1.8 m in 1.8 s; each device's
screen keys its cards ON at the second the bar's x passes its own.
· v05 — p01's pile, solved the same way (30 slabs, 40 % over the phone),
run BACKWARDS: each slab lifts 1600 mm on easeInCubic, 0.06 s apart,
farthest first. The phone was there all along.
THE COPY. An eyebrow ("VALUE") in dim and a headline in ink, weight 600,
worldHeight 44–56, on screen by 1.2 s. Placeholders; an assembly replaces them.
MATCH. Every scene's phone carries match=hero (v04's phone); all four carry
hz0/hz1/hz2.
LOOK — carried unchanged from the abstract reel:
ink #e9e6df dim #7c7970 line #23232a
mark #ffb347 cool #7ad7ff warm #ffa25e
POST. bloom → film_grain → fxaa; no dispersion in this file.
shot bloom thresh grain / vignette
01 one engine 0.65 0.82 0.012 / 0.32
02 fast 0.60 0.82 0.012 / 0.32
04 consistent 0.60 0.82 0.011 / 0.30
05 quiet 0.45 0.86 0.012 / 0.32
A LIMIT OF THE CAPTURE TOOL: headless capture does not render transmission
(see abstract-hero-scenes), so v01's gradient-glass reads as its opaque
tint in captures; the re-tint itself is visible there, the glass is not.
NO AUDIO. Primitives reel; add at assembly time.
Capture: nx run tools:capture:webgl -- --project promos/primitives/scenes/abstract/parts/abstract-value-propositions.promo --width 1920 --height 1080 --times "2.0,5.6,8.0,11.6,14.8,18.2,21.0,24.6"
Preview: /packages/runtime/index.html?project=promos/primitives/scenes/abstract/parts/abstract-value-propositions.promo- Topic
- scenes
- Scenes
- 4
- Kind
- reference rig
//
ABSTRACT · VALUE PROPOSITIONS — four claims, each with the thing that
proves it in the same frame, in the abstract reel's grammar. A value scene
is 5.5–6 s: the line is on screen by 1.2 s and stays, and the picture
spends the rest of the shot DOING what the line says. The reel's rule: a
claim that is only typed is a caption; a claim the set performs is a
proposition.
# scene dur transition device the claim, and its proof
01 v01-one-engine 6.0 match-move iphone ONE ENGINE. EVERY SURFACE — one gradient-glass definition on a sphere, a knot and the phone's shell, re-tinting together
02 v02-fast 5.5 match-move iphone LIVE IN UNDER A SECOND — the screen builds itself in 0.8 s while the camera crashes in on a ramp
04 v04-consistent 6.0 match-move iphone+ipad+laptop THE SAME ON EVERY SCREEN — one emissive scan crosses the family and each screen lights as it passes
05 v05-quiet 5.5 — iphone AND THEN, QUIET — the problem file's pile lifts off the phone and leaves; the key comes up
23.0 s of scene + three gaps (2.4 s) ≈ 25.4 s on the wall. Every scene also
stands alone: cut one out, set transition=none, and it is a value beat.
The fifteen-phone SCALE shot that was v03 lives in the scene-fit library as
projects/parts/v03-scale (its own folder, no world copy, hero on a ui3d stage).
THE PROOFS, and what each one costs:
· v01 — the coda's gradient-glass `house` definition, with its colour
tracks on the DEFINITION's fields, so the sphere, the knot and the
phone re-tint on one clock. A per-element track would opt that element
out of the shared field set and break the demonstration.
· v02 — eight screen cards on a 0.1 s stagger, and a camera push sampled
from a 0 · 0.06 · 0.62 · 1 table, so most of the travel is early and
the last 300 mm take 2 s: the screen is done before the camera is.
· v04 — a 6 × 600 mm emissive bar crossing 1.8 m in 1.8 s; each device's
screen keys its cards ON at the second the bar's x passes its own.
· v05 — p01's pile, solved the same way (30 slabs, 40 % over the phone),
run BACKWARDS: each slab lifts 1600 mm on easeInCubic, 0.06 s apart,
farthest first. The phone was there all along.
THE COPY. An eyebrow ("VALUE") in dim and a headline in ink, weight 600,
worldHeight 44–56, on screen by 1.2 s. Placeholders; an assembly replaces them.
MATCH. Every scene's phone carries match=hero (v04's phone); all four carry
hz0/hz1/hz2.
LOOK — carried unchanged from the abstract reel:
ink #e9e6df dim #7c7970 line #23232a
mark #ffb347 cool #7ad7ff warm #ffa25e
POST. bloom → film_grain → fxaa; no dispersion in this file.
shot bloom thresh grain / vignette
01 one engine 0.65 0.82 0.012 / 0.32
02 fast 0.60 0.82 0.012 / 0.32
04 consistent 0.60 0.82 0.011 / 0.30
05 quiet 0.45 0.86 0.012 / 0.32
A LIMIT OF THE CAPTURE TOOL: headless capture does not render transmission
(see abstract-hero-scenes), so v01's gradient-glass reads as its opaque
tint in captures; the re-tint itself is visible there, the glass is not.
NO AUDIO. Primitives reel; add at assembly time.
Capture: nx run tools:capture:webgl -- --project promos/primitives/scenes/abstract/parts/abstract-value-propositions.promo --width 1920 --height 1080 --times "2.0,5.6,8.0,11.6,14.8,18.2,21.0,24.6"
Preview: /packages/runtime/index.html?project=promos/primitives/scenes/abstract/parts/abstract-value-propositions.promo
promo#abstract-value-propositions(width=1920 height=1080 fps=30)
materials
// the abstract reel's six materials, carried whole so this file grades on its own
physical#clearcoat(color=#15161a metalness=0 roughness=0.45 c