dsp::biquad_coeffs Class Reference

#include <biquad.h>

Inheritance diagram for dsp::biquad_coeffs:

Public Types
typedef std::complex< double >	cfloat

Public Member Functions
	biquad_coeffs ()
void	set_null ()
void	set_lp_rbj (float fc, float q, float sr, float gain=1.0)
void	set_lp_zoelzer (float fc, float q, float odsr, float gain=1.0)
void	set_hp_rbj (float fc, float q, float esr, float gain=1.0)
void	set_hp_rbj_optimized (float fc, float q, float esr, float gain=1.0)
void	set_bp_rbj (double fc, double q, double esr, double gain=1.0)
void	set_br_rbj (double fc, double q, double esr, double gain=1.0)
void	set_allpass (float freq, float pole_r, float sr)
void	set_bilinear (double aa0, double aa1, double aa2, double ab0, double ab1, double ab2)
	set digital filter parameters based on given analog filter parameters
void	set_bilinear_direct (double aa0, double aa1, double aa2, double ab1, double ab2)
	set digital filter parameters directly
void	set_peakeq_rbj (double freq, double q, double peak, double sr)
void	set_lowshelf_rbj (float freq, float q, float peak, float sr)
void	set_highshelf_rbj (float freq, float q, float peak, float sr)
void	copy_coeffs (const biquad_coeffs &src)
	copy coefficients from another biquad
float	freq_gain (float freq, float sr) const
cfloat	h_z (const cfloat &z) const

Static Public Member Functions
static double	prewarp (float freq, float sr)
	prewarping for bilinear transform, maps given digital frequency to analog counterpart for analog filter design
static double	unwarp (float omega, float sr)
	convert analog angular frequency value to digital
static double	unwarpf (float t, float sr)
	convert analog filter time constant to digital counterpart

Public Attributes
double	a0
double	a1
double	a2
double	b1
double	b2

Detailed Description

Coefficients for two-pole two-zero filter, for floating point values, plus a bunch of functions to set them to typical values.

Coefficient calculation is based on famous Robert Bristow-Johnson's equations, except where it's not. The coefficient calculation is NOT mine, the only exception is the lossy optimization in Zoelzer and rbj HP filter code.

See http://www.musicdsp.org/files/Audio-EQ-Cookbook.txt for reference.

don't use this for integers because it won't work

Member Typedef Documentation

cfloat

typedef std::complex<double> dsp::biquad_coeffs::cfloat

Constructor & Destructor Documentation

biquad_coeffs()

dsp::biquad_coeffs::biquad_coeffs ( )

inline

Member Function Documentation

copy_coeffs()

void dsp::biquad_coeffs::copy_coeffs ( const biquad_coeffs & src )

inline

copy coefficients from another biquad

freq_gain()

float dsp::biquad_coeffs::freq_gain ( float freq, float sr ) const

inline

Return the filter's gain at frequency freq

Parameters

freq	Frequency to look up
sr	Filter sample rate (used to convert frequency to angular frequency)

h_z()

cfloat dsp::biquad_coeffs::h_z ( const cfloat & z ) const

inline

Return H(z) the filter's gain at frequency freq

Parameters

z	Z variable (e^jw)

prewarp()

double dsp::biquad_coeffs::prewarp ( float freq, float sr )

inlinestatic

prewarping for bilinear transform, maps given digital frequency to analog counterpart for analog filter design

set_allpass()

void dsp::biquad_coeffs::set_allpass ( float freq, float pole_r, float sr )

inline

set_bilinear()

void dsp::biquad_coeffs::set_bilinear ( double aa0, double aa1, double aa2, double ab0, double ab1, double ab2 )

inline

set digital filter parameters based on given analog filter parameters

set_bilinear_direct()

void dsp::biquad_coeffs::set_bilinear_direct ( double aa0, double aa1, double aa2, double ab1, double ab2 )

inline

set digital filter parameters directly

set_bp_rbj()

void dsp::biquad_coeffs::set_bp_rbj ( double fc, double q, double esr, double gain = 1.0 )

inline

Bandpass filter based on Robert Bristow-Johnson's equations (normalized to 1.0 at center frequency)

Parameters

fc	center frequency (gain at fc = 1.0)
q	=~ fc/bandwidth (not quite, but close) - 1/Q = 2*sinh(ln(2)/2*BW*w0/sin(w0))
sr	sample rate
gain	amplification (gain at sr/2)

set_br_rbj()

void dsp::biquad_coeffs::set_br_rbj ( double fc, double q, double esr, double gain = 1.0 )

inline

set_highshelf_rbj()

void dsp::biquad_coeffs::set_highshelf_rbj ( float freq, float q, float peak, float sr )

inline

RBJ high shelf EQ - amplitication of 0dB at 0 Hz and of peak at sr/2 Hz

Parameters

freq	corner frequency (gain at freq is sqrt(peak))
q	q (relates bandwidth and peak frequency), the higher q, the louder the resonant peak (situated above fc) is
peak	shelf gain (1.0 means no peak, >1.0 means a peak, less than 1.0 is a dip)

set_hp_rbj()

void dsp::biquad_coeffs::set_hp_rbj ( float fc, float q, float esr, float gain = 1.0 )

inline

Highpass filter based on Robert Bristow-Johnson's equations

Parameters

fc	resonant frequency
q	resonance (gain at fc)
sr	sample rate
gain	amplification (gain at sr/2)

set_hp_rbj_optimized()

void dsp::biquad_coeffs::set_hp_rbj_optimized ( float fc, float q, float esr, float gain = 1.0 )

inline

set_lowshelf_rbj()

void dsp::biquad_coeffs::set_lowshelf_rbj ( float freq, float q, float peak, float sr )

inline

RBJ low shelf EQ - amplitication of 'peak' at 0 Hz and of 1.0 (0dB) at sr/2 Hz

Parameters

freq	corner frequency (gain at freq is sqrt(peak))
q	q (relates bandwidth and peak frequency), the higher q, the louder the resonant peak (situated below fc) is
peak	shelf gain (1.0 means no peak, >1.0 means a peak, less than 1.0 is a dip)

set_lp_rbj()

void dsp::biquad_coeffs::set_lp_rbj ( float fc, float q, float sr, float gain = 1.0 )

inline

Lowpass filter based on Robert Bristow-Johnson's equations Perhaps every synth code that doesn't use SVF uses these equations :)

Parameters

fc	resonant frequency
q	resonance (gain at fc)
sr	sample rate
gain	amplification (gain at 0Hz)

set_lp_zoelzer()

void dsp::biquad_coeffs::set_lp_zoelzer ( float fc, float q, float odsr, float gain = 1.0 )

inline

set_null()

void dsp::biquad_coeffs::set_null ( )

inline

set_peakeq_rbj()

void dsp::biquad_coeffs::set_peakeq_rbj ( double freq, double q, double peak, double sr )

inline

RBJ peaking EQ

Parameters

freq	peak frequency
q	q (correlated to freq/bandwidth,

See also

set_bp_rbj)

Parameters

peak	peak gain (1.0 means no peak, >1.0 means a peak, less than 1.0 is a dip)

unwarp()

double dsp::biquad_coeffs::unwarp ( float omega, float sr )

inlinestatic

convert analog angular frequency value to digital

unwarpf()

double dsp::biquad_coeffs::unwarpf ( float t, float sr )

inlinestatic

convert analog filter time constant to digital counterpart

Member Data Documentation

a0

double dsp::biquad_coeffs::a0

a1

double dsp::biquad_coeffs::a1

a2

double dsp::biquad_coeffs::a2

b1

double dsp::biquad_coeffs::b1

b2

double dsp::biquad_coeffs::b2

---

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

• /home/runner/work/lmms-fork/lmms-fork/plugins/LadspaEffect/calf/veal/src/calf/biquad.h