Math_hp

Auto-generated from src/stdlib/math_hp.mcrs — do not edit manually.

API

• double_mul_fixed
• double_mul
• ln_hp
• ln_5term
• double_sqrt
• pow_fast

---

double_mul_fixed v1.3.0

Multiply a double-precision value d by a fixed-point integer f (× 10000) using a macro trick to keep all arithmetic in IEEE 754 double precision.

fn double_mul_fixed(d: double, f: int): double

Parameters

Parameter	Description
d	Double operand (stored in NBT as IEEE 754 double)
f	Fixed-point multiplier × 10000 (e.g. 15000 = 1.5)

Returns: d × (f / 10000) as a double

Example

let r: double = double_mul_fixed(3.14d, 20000); // 6.28  (3.14 × 2.0)

---

double_mul v1.3.0

Multiply two doubles via scoreboard integer approximation (~4 decimal digits). ⚠ Overflows for |a| × |b| > ~21474. Use double_mul_fixed for higher precision.

fn double_mul(a: double, b: double): double

Parameters

Parameter	Description
a	First factor (double)
b	Second factor (double)

Returns: a × b (scoreboard approximation, ~4 sig. figs.)

---

ln_hp v1.3.0

High-precision natural logarithm via one Newton refinement of the 5-term atanh series. Precision ~8–9 significant digits. Valid for x ∈ [100, 1 000 000] (× 10000 scale).

fn ln_hp(x: int): int

Parameters

Parameter	Description
x	Input value × 10000 (e.g. 10000 = 1.0, 27183 ≈ e)

Returns: ln(x / 10000) × 10000

Example

let r: int = ln_hp(27183); // ≈ 10000  (ln(e) ≈ 1.0)

---

ln_5term v1.3.0

Natural logarithm using a 5-term atanh series (more accurate than the 3-term ln()). Max error ≈ 0.000002. Valid for x ∈ [100, 1 000 000] (× 10000 scale).

fn ln_5term(x: int): int

Parameters

Parameter	Description
x	Input value × 10000

Returns: ln(x / 10000) × 10000

---

double_sqrt v1.3.0

Square root of a double-precision value, returned as a double. Internally converts to × 10000 fixed-point, applies isqrt, then converts back.

fn double_sqrt(x: double): double

Parameters

Parameter	Description
x	Input double (must be ≥ 0)

Returns: √x as a double

---

pow_fast v1.3.0

Double base raised to an integer exponent using binary exponentiation (O(log n)). Handles negative exponents via double_div.

fn pow_fast(base: double, exp_val: int): double

Parameters

Parameter	Description
base	Base value (double)
exp_val	Integer exponent (may be negative)

Returns: base^exp_val as a double

Example

let r: double = pow_fast(2.0d, 10); // 1024.0

---