Module truing_dsf 

Mach-keyed drop-scale-factor (DSF) truing table (MBA-1357).

Applied Ballistics’ published two-stage truing workflow calibrates muzzle velocity (MV) first — that fixes the supersonic drag curve against a chronograph/observed-drop comparison at Mach >= 1.2. Below that, as the bullet moves through the transonic region and into the subsonic regime, no single MV correction can fix drop discrepancies that grow with range: the residual is a slowly-varying function of Mach, not a constant offset. AB’s second stage records a handful of observed drop / predicted drop ratios at specific (subsonic-or-transonic) Mach numbers and uses them to scale predicted drop at nearby Mach numbers on later solves.

This module is a cleanroom reimplementation of that workflow’s shape, not a bit-for-bit copy of AB’s unpublished interpolation — Kestrel/AB do not publish their exact curve. The design decision unique to this implementation is the anchor: the table’s Mach domain is (0, 1.2) (points at/above Mach 1.2 belong to MV truing, not here — DsfTable::from_points rejects them), and every table implicitly continues with a DSF of 1.0 at Mach 1.2 — the exact boundary where MV calibration takes over. That implicit anchor point (1.2, 1.0) is never stored in DsfTable::points; it exists only inside DsfTable::factor_at’s interpolation so the transition from “supersonic, MV-trued, unscaled” to “transonic/subsonic, DSF-scaled” is continuous — a shot solved at Mach 1.1999 and one solved at Mach 1.2001 get (to floating-point precision) the same drop. This is a functional-equivalence choice made for this engine, not a replication of AB’s internal method.

Below the lowest recorded point, DsfTable::factor_at flat-clamps to that point’s DSF — there is no data past it, and AB’s guidance is that further subsonic drop continues to track the last-calibrated regime rather than drift back toward identity.

apply_dsf is a drop-only post-processing step over an already-solved crate::TrajectoryResult: it rescales each point’s vertical position relative to the line of sight by the DSF at that point’s Mach, and touches nothing else — velocity, kinetic energy, time, and downrange/windage position are byte-identical before and after. Per-point Mach is computed the same way the solver’s own diagnostics compute it (see apply_dsf’s doc comment for the exact fields), NOT from a re-derived per-altitude local speed of sound the engine does not store per point.

No feature gate: this module must compile for wasm32-unknown-unknown. It is fs-free (profile persistence of a table’s points is the caller’s job, e.g. main.rs’s saved-profile handling in a later task).

Structs

DsfPoint

One observed drop-scale-factor keyed to the Mach number it was recorded at.

DsfTable

A validated, Mach-sorted table of up to DSF_MAX_POINTS DsfPoints.

Enums

UpsertOutcome

What DsfTable::upsert did with the incoming point.

Constants

DSF_ANCHOR_VALUE

DSF value of the implicit anchor at DSF_MACH_CEILING — identity, matching the MV-trued supersonic regime this table hands off from.

DSF_MACH_CEILING

Upper bound (exclusive) of the Mach domain a DsfPoint may describe. Observations at or above this Mach belong to muzzle-velocity truing, not the DSF table; it doubles as the implicit anchor’s Mach coordinate ((DSF_MACH_CEILING, 1.0)) in DsfTable::factor_at.

DSF_MAX

Exclusive upper bound a point’s dsf must clear.

DSF_MAX_POINTS

Maximum number of distinct points a DsfTable may hold.

DSF_MIN

Exclusive lower bound a point’s dsf must clear.

DSF_SUPERSEDE_TOLERANCE_MACH

A new point within this many Mach units of an existing one supersedes it in DsfTable::upsert instead of being appended.

Functions

apply_dsf

Apply a DSF table to an already-solved trajectory, IN PLACE, scaling only each point’s drop below the line of sight — in BOTH result.points and, when present, result.sampled_points.