Every instrument before this one narrated the wheel. This one asks you to compute with it. Three derivations, each carried from the turn-counting to a number, each number set beside the measurement that already exists. You supply the inputs. The calculus supplies the outputs. The laboratory record supplies the verdict.
Feynman's arrows earned their permanence partly because a motivated reader could actually compute with them — the metaphor was load-bearing. If the wheel can be made load-bearing the same way — a page where a reader uses the turn-counting to derive a fringe spacing or a lensing angle themselves, numbers in, numbers out — then it stops being an unusually complete narration and becomes a method.
The calculus has five operations. Each is established physics — measured, not proposed. What T Theory adds is the reading that threads them: every t replaced by the count. Below, the three benches use nothing but these five lines.
The Double Slit showed the two counts agreeing and disagreeing down the wall. Now carry it to the number. Two doors a distance d apart, a wall a distance L beyond, a figure whose count repeats every λ. At height y on the wall the two paths differ by d·y∕L, so the counts differ by ΔN = d·y∕(λL) turns — whole turns bright, half turns dark — and the bright bands land every Δy = λL∕d. That spacing is not a metaphor. It has been measured, three ways, below.
Where the count runs slow, light takes longer — so a ray passing a mass at distance b leans toward the thin count, as through glass, and comes out bent by α ≈ 2rs∕b. Set the mass. Set the passage. Read the angle — and note the trap laid open on the right: a count thinned in time alone gives exactly half the lean. The measurement is what closes that door.
The gentlest of the derivations, and the most mercilessly verified. Up a tower the count runs faster by gh∕c² — Pound and Rebka heard the difference over 22.5 metres in 1959; by 2022 optical lattice clocks resolved it across one millimetre. And every satellite is a clock aloft: gravity thins the count below it, speed thins its own. Two thinnings, opposite signs, one net drift — engineered into every GPS fix you have ever taken.
Every operation of the calculus as physics: the Compton clock; λ = h∕p from electrons (Davisson–Germer 1927) to 25,000-amu molecules (Fein 2019); fringe spacings measured for electrons (Jönsson 1961), neutrons (Zeilinger 1988), and fullerenes (Arndt 1999); light deflection at the Sun to 1 part in 10⁴ (VLBI) and γ = 1 ± 2.3 × 10⁻⁵ (Cassini 2003); gravitational rate shift from 22.5 m (1959) down to 1 mm (NIST 2022); the GPS correction, running in production for four decades.
The thread alone: mass as tempo, the wavelength as the count laid on the chart, interference as the agreement of two counts, gravity as the thinning of the count, lensing as light leaning toward the thin count. No equation on this page is altered by the reading — that is the point, and the price. The wheel carries the same load by a different grip.
The critique asked for a bridge from metaphor to calculation. This page is the crossing: every number above was reached by counting turns, and every one lands on the measured value.
Δy = λL∕d is the small-angle, ideal-door form; real apparatus adds envelopes, biprisms, magnification — the cited records fold those in. α = 2rs∕b is weak-field only; the full lean is Einstein's equation, which no one line abbreviates. The half-count trap in § 03 is real: time-thinning alone gives rs∕b, and only the full index n(r) — space thinned with time — doubles it. The reading must carry both thinnings or it fails the 1919 number. It is carried here; it must be carried everywhere.
The count on the chart. Find one massive figure whose measured fringe spacing tracks anything other than λ = h∕p — one interferometer, any mass from an electron to a virus, where Δy ≠ λL∕d beyond the apparatus budget. One verified case, and R·02 falls, and § 02 with it.
The lean. A weak-field deflection departing from 2rs∕b — equivalently γ ≠ 1 beyond the Cassini bound of 2.3 × 10⁻⁵. One verified case, and R·05 falls, and § 03 with it.
The thinned count. A clock whose rate shift departs from the lapse √(1−rs∕r) — at any height a lattice clock can resolve. One verified case, and R·04 falls, and § 04 with it.
The standing flag — re-derivation is not discovery. Every output of this calculus equals the standard prediction, by construction. The bridge is crossed: the reading computes. It becomes a theory only on the day the counting demands a number the standard account does not — and the measurement takes the counting's side. No such case is known. Until one exists, this page claims exactly what it shows: a method that carries the measured world, not a rival to it. This flag stays raised, in the open, with the same prominence as the claims it tests.