quantize()

void MayaFlux::Kinesis::Discrete::quantize ( std::span< double >  data, uint8_t  bits  )

noexcept

Quantize to n-bit resolution in-place Input is assumed to be in [-1, 1]; values outside are clamped first.

Uses lrint bias trick to avoid scalar std::round and improve auto-vectorisation.

Parameters

data	Target span
bits	Bit depth (1–53)

Definition at line 50 of file Transform.cpp.

{
    const double levels = std::pow(2.0, bits) - 1.0;
    const double inv_levels = 1.0 / levels;
    std::ranges::transform(data, data.begin(),
        [levels, inv_levels](double x) {
            const double clamped = std::clamp(x, -1.0, 1.0);
            return static_cast<double>(std::lrint(clamped * levels)) * inv_levels;
        });
}