PandaResampler::Resampler2 Class Reference

#include <smpandaresampler.hh>

Classes
class	Downsampler2
class	IIRDownsampler2
class	IIRUpsampler2
class	Upsampler2

Public Types
enum	Mode { UP , DOWN }
enum	Precision {   PREC_LINEAR = 1 , PREC_48DB = 8 , PREC_72DB = 12 , PREC_96DB = 16 ,   PREC_120DB = 20 , PREC_144DB = 24 }
enum	Filter { FILTER_IIR , FILTER_FIR }

Public Member Functions
	Resampler2 (Mode mode, uint ratio, Precision precision, bool use_sse_if_available=true, Filter filter=FILTER_FIR)
void	process_block (const float *input, uint n_input_samples, float *output)
uint	order () const
double	delay () const
void	reset ()
bool	sse_enabled () const

Static Public Member Functions
static bool	sse_available ()
static bool	test_filter_impl (bool verbose)
static Precision	find_precision_for_bits (uint bits)
static const char *	precision_name (Precision precision)

Protected Member Functions
template<bool USE_SSE>
Impl *	create_impl (uint stage_ratio)
template<bool USE_SSE>
Impl *	create_impl_iir (uint stage_ratio)
template<class CArray >
Impl *	create_impl_iir_with_coeffs (const CArray &carray, double group_delay)
void	init_stage (std::unique_ptr< Impl > &impl, uint stage_ratio)

Static Protected Member Functions
template<class Filter >
static Impl *	create_impl_with_coeffs (const double *d, uint order, double scaling)

Protected Attributes
Mode	mode_
Precision	precision_
bool	use_sse_if_available_
Filter	filter_

Detailed Description

Interface for factor 2 resampling classes

Constructor & Destructor Documentation

Resampler2()

PANDA_RESAMPLER_FN PandaResampler::Resampler2::Resampler2	(	Mode	mode,
		uint	ratio,
		Precision	precision,
		bool	use_sse_if_available = true,
		Filter	filter = FILTER_FIR
	)

creates a resampler instance fulfilling a given specification

Member Function Documentation

delay()

double PandaResampler::Resampler2::delay ( ) const

inline

Return the delay introduced by the resampler. This delay is guaranteed to be >= 0.0, and for factor 2 resampling always a multiple of 0.5 (1.0 for upsampling).

The return value can also be thought of as index into the output signal, where the first input sample can be found.

Beware of fractional delays, for instance for downsampling, a delay() of 10.5 means that the first input sample would be found by interpolating output[10] and output[11], and the second input sample equates output[11].

find_precision_for_bits()

PANDA_RESAMPLER_FN Resampler2::Precision PandaResampler::Resampler2::find_precision_for_bits ( uint  bits )

static

finds a precision which is appropriate for at least the specified number of bits

order()

uint PandaResampler::Resampler2::order ( ) const

inline

return FIR filter order

precision_name()

PANDA_RESAMPLER_FN const char * PandaResampler::Resampler2::precision_name ( Precision  precision )

static

returns a human-readable name for a given precision

process_block()

void PandaResampler::Resampler2::process_block ( const float *  input, uint  n_input_samples, float *  output  )

inline

resample a data block

reset()

void PandaResampler::Resampler2::reset ( )

inline

clear internal history, reset resampler state to zero values

sse_available()

PANDA_RESAMPLER_FN bool PandaResampler::Resampler2::sse_available ( )

static

returns true if an optimized SSE version of the Resampler is available

sse_enabled()

bool PandaResampler::Resampler2::sse_enabled ( ) const

inline

return whether the resampler is using sse optimized code

test_filter_impl()

PANDA_RESAMPLER_FN bool PandaResampler::Resampler2::test_filter_impl ( bool  verbose )

static

test internal filter implementation

---

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

• lib/smpandaresampler.hh
• lib/smpandaresampler.cc