pulp::signal::SvfBankT< SampleType, Voices > Class Template Reference

Voices independent SvfT voices advanced together. More...

#include <svf_bank.hpp>

Public Types
using	Filter = SvfT<SampleType>
using	Mode = typename Filter::Mode

Public Member Functions
	SvfBankT ()
void	configure (std::size_t voice, Mode mode, SampleType sample_rate, SampleType frequency, SampleType resonance)
	Install a coherent mode, rate, cutoff and resonance for one voice (SvfT::configure()).
void	set_sample_rate (std::size_t voice, SampleType sample_rate)
void	set_frequency (std::size_t voice, SampleType hz)
void	set_resonance (std::size_t voice, SampleType q)
void	set_mode (std::size_t voice, Mode mode)
Filter::Coefficients	coefficients (std::size_t voice) const
	The voice's coefficients (SvfT::design() for its current settings).
void	process (const SampleType *in, SampleType *out)
	Advance every voice by one sample: out[v] = voice v's output for in[v].
void	process_voices (SampleType *const *buffers, std::size_t num_samples)
	Advance every voice by num_samples samples in place, one buffer per voice (buffers[v] holds voice v's input and receives its output).
void	reset (std::size_t voice)
void	reset ()

Static Public Attributes
static constexpr std::size_t	voice_count = Voices

Detailed Description

template<typename SampleType = float, std::size_t Voices = 16>
class pulp::signal::SvfBankT< SampleType, Voices >

Voices independent SvfT voices advanced together.

Contract:

• Each voice's output is bit-identical to an SvfT<SampleType> with the same mode, sample rate, frequency and resonance, driven with the same inputs through SvfT::process(SampleType) from the same reset state, including its per-sample denormal snap. Setters have SvfT's semantics: coefficients come from SvfT::design() and are recomputed by every rate, frequency or resonance change; a default-constructed voice has all-zero coefficients until one of them is set, as SvfT does.
• Voices share nothing but storage, and each has its own mode.
• process() advances every voice by one sample; process_voices() advances every voice by a block, one buffer per voice.

Why it is faster: the voices' state is held in structure-of-arrays form and every step is a loop over independent voices, which the compiler vectorizes; the per-voice mode becomes a per-lane select instead of a branch.

RT contract: storage is fixed-size; every member function allocates no memory.

Member Typedef Documentation

Filter

template<typename SampleType = float, std::size_t Voices = 16>

using pulp::signal::SvfBankT< SampleType, Voices >::Filter = SvfT<SampleType>

Mode

template<typename SampleType = float, std::size_t Voices = 16>

using pulp::signal::SvfBankT< SampleType, Voices >::Mode = typename Filter::Mode

Constructor & Destructor Documentation

SvfBankT()

template<typename SampleType = float, std::size_t Voices = 16>

pulp::signal::SvfBankT< SampleType, Voices >::SvfBankT ( )

inline

Member Function Documentation

coefficients()

template<typename SampleType = float, std::size_t Voices = 16>

Filter::Coefficients pulp::signal::SvfBankT< SampleType, Voices >::coefficients ( std::size_t voice ) const

inline

The voice's coefficients (SvfT::design() for its current settings).

configure()

template<typename SampleType = float, std::size_t Voices = 16>

void pulp::signal::SvfBankT< SampleType, Voices >::configure ( std::size_t voice, Mode mode, SampleType sample_rate, SampleType frequency, SampleType resonance )

inline

Install a coherent mode, rate, cutoff and resonance for one voice (SvfT::configure()).

process()

template<typename SampleType = float, std::size_t Voices = 16>

void pulp::signal::SvfBankT< SampleType, Voices >::process ( const SampleType * in, SampleType * out )

inline

Advance every voice by one sample: out[v] = voice v's output for in[v].

in and out may be the same array.

process_voices()

template<typename SampleType = float, std::size_t Voices = 16>

void pulp::signal::SvfBankT< SampleType, Voices >::process_voices ( SampleType *const * buffers, std::size_t num_samples )

inline

Advance every voice by num_samples samples in place, one buffer per voice (buffers[v] holds voice v's input and receives its output).

reset() [1/2]

template<typename SampleType = float, std::size_t Voices = 16>

void pulp::signal::SvfBankT< SampleType, Voices >::reset ( )

inline

reset() [2/2]

template<typename SampleType = float, std::size_t Voices = 16>

void pulp::signal::SvfBankT< SampleType, Voices >::reset ( std::size_t voice )

inline

set_frequency()

template<typename SampleType = float, std::size_t Voices = 16>

void pulp::signal::SvfBankT< SampleType, Voices >::set_frequency ( std::size_t voice, SampleType hz )

inline

set_mode()

template<typename SampleType = float, std::size_t Voices = 16>

void pulp::signal::SvfBankT< SampleType, Voices >::set_mode ( std::size_t voice, Mode mode )

inline

set_resonance()

template<typename SampleType = float, std::size_t Voices = 16>

void pulp::signal::SvfBankT< SampleType, Voices >::set_resonance ( std::size_t voice, SampleType q )

inline

set_sample_rate()

template<typename SampleType = float, std::size_t Voices = 16>

void pulp::signal::SvfBankT< SampleType, Voices >::set_sample_rate ( std::size_t voice, SampleType sample_rate )

inline

Member Data Documentation

voice_count

template<typename SampleType = float, std::size_t Voices = 16>

std::size_t pulp::signal::SvfBankT< SampleType, Voices >::voice_count = Voices

staticconstexpr

---

The documentation for this class was generated from the following file:

• /home/runner/work/pulp/pulp/core/signal/include/pulp/signal/svf_bank.hpp