Signal

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

API

• uniform_int
• uniform_frac
• normal_approx12
• exp_dist_approx
• bernoulli
• weighted2
• weighted3
• gamma_sample
• poisson_sample
• geometric_sample
• negative_binomial_sample
• dft_real
• dft_imag
• dft_magnitude

---

uniform_int v2.0.0

Return a uniform integer in [lo, hi] inclusive.

Uses LCG RNG. Pass the result as the seed for the next call to chain samples.

fn uniform_int(seed: int, lo: int, hi: int): int

Parameters

Parameter	Description
seed	Any integer seed value
lo	Inclusive lower bound
hi	Inclusive upper bound

Returns: Pseudo-random integer in [lo, hi]

Example

let dmg: int = uniform_int(seed, 5, 15)

---

uniform_frac v2.0.0

Return a uniform fraction in [0, 10000] (×10000 scale).

fn uniform_frac(seed: int): int

Parameters

Parameter	Description
seed	Any integer seed value

Returns: Pseudo-random integer in [0, 10000]

Example

let frac: int = uniform_frac(seed)

---

normal_approx12 v2.0.0

Approximate N(0, 1) variate using the Irwin–Hall method (sum of 12 uniform samples).

Result is in ×10000 scale, range approximately [−60000, 60000]. Each call chains the seed 12 times internally.

fn normal_approx12(seed: int): int

Parameters

Parameter	Description
seed	Any integer seed value

Returns: Approximate N(0,1) sample ×10000

Example

let z: int = normal_approx12(seed)

---

exp_dist_approx v2.0.0

Sample from an exponential distribution with rate lambda_fx.

Method: -ln(U) / λ where U ~ Uniform(0.01, 1). Result is capped at 100000 (= 10.0 × 10000) for Minecraft sanity.

Requires import "stdlib/math" for the ln function.

fn exp_dist_approx(seed: int, lambda_fx: int): int

Parameters

Parameter	Description
seed	Any integer seed value
lambda_fx	Rate parameter ×10000 (e.g. 10000 = rate 1.0)

Returns: Exponential variate ×10000, capped at 100000

Example

let wait: int = exp_dist_approx(seed, 10000)

---

bernoulli v2.0.0

Return 1 with probability p_fx / 10000, otherwise 0.

fn bernoulli(seed: int, p_fx: int): int

Parameters

Parameter	Description
seed	Any integer seed value
p_fx	Probability ×10000 (e.g. 5000 = 50%, 1000 = 10%)

Returns: 1 with the given probability, 0 otherwise

Example

if (bernoulli(seed, 3000) == 1) { /* 30% chance */ }

---

weighted2 v2.0.0

Choose 0 or 1 with the given integer weights.

fn weighted2(seed: int, w0: int, w1: int): int

Parameters

Parameter	Description
seed	Any integer seed value
w0	Weight for outcome 0
w1	Weight for outcome 1

Returns: 0 or 1 sampled proportionally to the weights

Example

let side: int = weighted2(seed, 3, 7)

---

weighted3 v2.0.0

Choose 0, 1, or 2 with the given integer weights.

fn weighted3(seed: int, w0: int, w1: int, w2: int): int

Parameters

Parameter	Description
seed	Any integer seed value
w0	Weight for outcome 0
w1	Weight for outcome 1
w2	Weight for outcome 2

Returns: 0, 1, or 2 sampled proportionally to the weights

Example

let tier: int = weighted3(seed, 50, 30, 20)

---

gamma_sample v2.0.0

Sample from a Gamma(k, θ) distribution via summing k exponential samples.

Handles integer shape k = 1..5 (pass shape_k × 10000). Requires ln from stdlib/math.

fn gamma_sample(shape_k: int, scale_theta: int, seed: int): int

Parameters

Parameter	Description
shape_k	Shape parameter k ×10000 (e.g. 20000 = k=2)
scale_theta	Scale parameter θ ×10000 (e.g. 10000 = θ=1.0)
seed	Any integer seed value

Returns: Gamma variate ×10000

Example

let g: int = gamma_sample(20000, 10000, seed)

---

poisson_sample v2.0.0

Sample from a Poisson(λ) distribution using the Knuth algorithm.

Works well for lambda ≤ 20 (200000 in ×10000). Hard cap at 100 iterations. Requires exp_fx from stdlib/math.

fn poisson_sample(lambda: int, seed: int): int

Parameters

Parameter	Description
lambda	Rate parameter ×10000 (e.g. 30000 = λ=3.0)
seed	Any integer seed value

Returns: Poisson count (plain integer, not ×10000)

Example

let n: int = poisson_sample(30000, seed)

---

geometric_sample v2.0.0

Sample from a Geometric(p) distribution (number of failures before first success).

Method: floor(ln(U) / ln(1 - p)). Requires ln from stdlib/math.

fn geometric_sample(p_success: int, seed: int): int

Parameters

Parameter	Description
p_success	Success probability ×10000 (e.g. 5000 = p=0.5)
seed	Any integer seed value

Returns: Non-negative integer count of failures

Example

let fails: int = geometric_sample(5000, seed)

---

negative_binomial_sample v2.0.0

Sample from a Negative Binomial NegBin(r, p) distribution.

Method: sum r independent Geometric(p) samples.

fn negative_binomial_sample(r: int, p_success: int, seed: int): int

Parameters

Parameter	Description
r	Number of successes (plain integer, e.g. 1, 2, 3)
p_success	Success probability ×10000 (e.g. 5000 = p=0.5)
seed	Any integer seed value

Returns: Total number of failures before r successes

Example

let n: int = negative_binomial_sample(3, 5000, seed)

---

dft_real v2.0.0

Real part of DFT bin k for a real-valued signal with up to 8 samples.

All values use ×10000 scale. Angle convention: multiples of 45°. Unused sample arguments (beyond n) are ignored.

Formula: real[k] = (1/n) Σ samples[j] × cos(2πkj/n)

fn dft_real(s0: int, s1: int, s2: int, s3: int, s4: int, s5: int, s6: int, s7: int, n: int, k: int): int

Parameters

Parameter	Description
s0	Sample 0 ×10000 … @param s7 Sample 7 ×10000
n	Number of samples (1–8)
k	Bin index (0 to n−1)

Returns: Real part of DFT bin k, ×10000

Example

let re0: int = dft_real(10000, 0, -10000, 0, 0, 0, 0, 0, 4, 0)

---

dft_imag v2.0.0

Imaginary part of DFT bin k for a real-valued signal with up to 8 samples.

Uses the negative-sine convention: imag[k] = -(1/n) Σ samples[j] × sin(2πkj/n)

fn dft_imag(s0: int, s1: int, s2: int, s3: int, s4: int, s5: int, s6: int, s7: int, n: int, k: int): int

Parameters

Parameter	Description
s0	Sample 0 ×10000 … @param s7 Sample 7 ×10000
n	Number of samples (1–8)
k	Bin index (0 to n−1)

Returns: Imaginary part of DFT bin k, ×10000

---

dft_magnitude v2.0.0

Magnitude of DFT bin k: sqrt(real² + imag²) in ×10000.

fn dft_magnitude(s0: int, s1: int, s2: int, s3: int, s4: int, s5: int, s6: int, s7: int, n: int, k: int): int

Parameters

Parameter	Description
s0	Sample 0 ×10000 … @param s7 Sample 7 ×10000
n	Number of samples (1–8)
k	Bin index (0 to n−1)

Returns: Magnitude of DFT bin k in ×10000

Example

let mag: int = dft_magnitude(10000, 0, -10000, 0, 0, 0, 0, 0, 4, 1)

---