Expand Up
@@ -27,20 +27,23 @@
#include " EffectChain.h"
#include " FxMixer.h"
#include " engine.h"
#include " MixHelpers.h"
#include " BufferManager.h"
#include " ValueBuffer.h"
#include " panning.h"
AudioPort::AudioPort ( const QString & _name, bool _has_effect_chain ) :
m_bufferUsage( NoUsage ),
m_firstBuffer( new sampleFrame[engine::mixer()->framesPerPeriod()] ),
m_secondBuffer( new sampleFrame[
engine::mixer ()->framesPerPeriod()] ),
AudioPort::AudioPort ( const QString & _name, bool _has_effect_chain,
FloatModel * volumeModel, FloatModel * panningModel ) :
m_bufferUsage( false ),
m_portBuffer( NULL ),
m_extOutputEnabled( false ),
m_nextFxChannel( 0 ),
m_name( " unnamed port"
m_effects( _has_effect_chain ? new EffectChain( NULL ) : NULL )
m_effects( _has_effect_chain ? new EffectChain( NULL ) : NULL ),
m_volumeModel( volumeModel ),
m_panningModel( panningModel )
{
engine::mixer ()->clearAudioBuffer ( m_firstBuffer, engine::mixer ()->framesPerPeriod () );
engine::mixer ()->clearAudioBuffer ( m_secondBuffer, engine::mixer ()->framesPerPeriod () );
engine::mixer ()->addAudioPort ( this );
setExtOutputEnabled ( true );
}
Expand All
@@ -52,31 +55,12 @@ AudioPort::~AudioPort()
{
setExtOutputEnabled ( false );
engine::mixer ()->removeAudioPort ( this );
delete[] m_firstBuffer;
delete[] m_secondBuffer;
delete m_effects;
}
void AudioPort::nextPeriod ()
{
m_firstBufferLock.lock ();
engine::mixer ()->clearAudioBuffer ( m_firstBuffer, engine::mixer ()->framesPerPeriod () );
qSwap ( m_firstBuffer, m_secondBuffer );
// this is how we decrease state of buffer-usage ;-)
m_bufferUsage = ( m_bufferUsage != NoUsage ) ?
( ( m_bufferUsage == FirstBuffer ) ?
NoUsage : FirstBuffer ) : NoUsage;
m_firstBufferLock.unlock ();
}
void AudioPort::setExtOutputEnabled ( bool _enabled )
{
if ( _enabled != m_extOutputEnabled )
Expand Down
Expand Up
@@ -109,23 +93,152 @@ bool AudioPort::processEffects()
{
if ( m_effects )
{
lockFirstBuffer ();
bool hasInputNoise = m_bufferUsage != NoUsage;
bool more = m_effects->processAudioBuffer ( m_firstBuffer, engine::mixer ()->framesPerPeriod (), hasInputNoise );
unlockFirstBuffer ();
bool more = m_effects->processAudioBuffer ( m_portBuffer, engine::mixer ()->framesPerPeriod (), m_bufferUsage );
return more;
}
return false ;
}
void AudioPort::doProcessing ( sampleFrame * )
void AudioPort::doProcessing ()
{
const fpp_t fpp = engine::mixer ()->framesPerPeriod ();
m_portBuffer = BufferManager::acquire (); // get buffer for processing
engine::mixer ()->clearAudioBuffer ( m_portBuffer, fpp ); // clear the audioport buffer so we can use it
// qDebug( "Playhandles: %d", m_playHandles.size() );
foreach ( PlayHandle * ph, m_playHandles ) // now we mix all playhandle buffers into the audioport buffer
{
if ( ph->buffer () )
{
if ( ph->usesBuffer () )
{
m_bufferUsage = true ;
MixHelpers::add ( m_portBuffer, ph->buffer (), fpp );
}
ph->releaseBuffer (); // gets rid of playhandle's buffer and sets
// pointer to null, so if it doesn't get re-acquired we know to skip it next time
}
}
if ( m_bufferUsage )
{
// handle volume and panning
// has both vol and pan models
if ( m_volumeModel && m_panningModel )
{
ValueBuffer * volBuf = m_volumeModel->valueBuffer ();
ValueBuffer * panBuf = m_panningModel->valueBuffer ();
// both vol and pan have s.ex.data:
if ( volBuf && panBuf )
{
for ( f_cnt_t f = 0 ; f < fpp; ++f )
{
float v = volBuf->values ()[ f ] * 0 .01f ;
float p = panBuf->values ()[ f ] * 0 .01f ;
m_portBuffer[f][0 ] *= ( p <= 0 ? 1 .0f : 1 .0f - p ) * v;
m_portBuffer[f][1 ] *= ( p >= 0 ? 1 .0f : 1 .0f + p ) * v;
}
}
// only vol has s.ex.data:
else if ( volBuf )
{
float p = m_panningModel->value () * 0 .01f ;
float l = ( p <= 0 ? 1 .0f : 1 .0f - p );
float r = ( p >= 0 ? 1 .0f : 1 .0f + p );
for ( f_cnt_t f = 0 ; f < fpp; ++f )
{
float v = volBuf->values ()[ f ] * 0 .01f ;
m_portBuffer[f][0 ] *= v * l;
m_portBuffer[f][1 ] *= v * r;
}
}
// only pan has s.ex.data:
else if ( panBuf )
{
float v = m_volumeModel->value () * 0 .01f ;
for ( f_cnt_t f = 0 ; f < fpp; ++f )
{
float p = panBuf->values ()[ f ] * 0 .01f ;
m_portBuffer[f][0 ] *= ( p <= 0 ? 1 .0f : 1 .0f - p ) * v;
m_portBuffer[f][1 ] *= ( p >= 0 ? 1 .0f : 1 .0f + p ) * v;
}
}
// neither has s.ex.data:
else
{
float p = m_panningModel->value () * 0 .01f ;
float v = m_volumeModel->value () * 0 .01f ;
for ( f_cnt_t f = 0 ; f < fpp; ++f )
{
m_portBuffer[f][0 ] *= ( p <= 0 ? 1 .0f : 1 .0f - p ) * v;
m_portBuffer[f][1 ] *= ( p >= 0 ? 1 .0f : 1 .0f + p ) * v;
}
}
}
// has vol model only
else if ( m_volumeModel )
{
ValueBuffer * volBuf = m_volumeModel->valueBuffer ();
if ( volBuf )
{
for ( f_cnt_t f = 0 ; f < fpp; ++f )
{
float v = volBuf->values ()[ f ] * 0 .01f ;
m_portBuffer[f][0 ] *= v;
m_portBuffer[f][1 ] *= v;
}
}
else
{
float v = m_volumeModel->value () * 0 .01f ;
for ( f_cnt_t f = 0 ; f < fpp; ++f )
{
m_portBuffer[f][0 ] *= v;
m_portBuffer[f][1 ] *= v;
}
}
}
}
// as of now there's no situation where we only have panning model but no volume model
// if we have neither, we don't have to do anything here - just pass the audio as is
// handle effects
const bool me = processEffects ();
if ( me || m_bufferUsage != NoUsage )
if ( me || m_bufferUsage )
{
engine::fxMixer ()->mixToChannel ( firstBuffer (), nextFxChannel () );
nextPeriod ();
engine::fxMixer ()->mixToChannel ( m_portBuffer, m_nextFxChannel ); // send output to fx mixer
// TODO: improve the flow here - convert to pull model
m_bufferUsage = false ;
}
BufferManager::release ( m_portBuffer ); // release buffer, we don't need it anymore
}
void AudioPort::addPlayHandle ( PlayHandle * handle )
{
m_playHandleLock.lock ();
m_playHandles.append ( handle );
m_playHandleLock.unlock ();
}
void AudioPort::removePlayHandle ( PlayHandle * handle )
{
m_playHandleLock.lock ();
PlayHandleList::Iterator it = qFind ( m_playHandles.begin (), m_playHandles.end (), handle );
if ( it != m_playHandles.end () )
{
m_playHandles.erase ( it );
}
m_playHandleLock.unlock ();
}