amuse/lib/Voice.cpp

1329 lines
38 KiB
C++

#include "amuse/Voice.hpp"
#include <cmath>
#include <cstring>
#include "amuse/AudioGroup.hpp"
#include "amuse/Common.hpp"
#include "amuse/DSPCodec.hpp"
#include "amuse/Engine.hpp"
#include "amuse/IBackendVoice.hpp"
#include "amuse/IBackendVoiceAllocator.hpp"
#include "amuse/N64MusyXCodec.hpp"
#include "amuse/Submix.hpp"
#include "amuse/VolumeTable.hpp"
namespace amuse {
float Voice::VolumeCache::getVolume(float vol, bool dls) {
if (vol != m_curVolLUTKey || dls != m_curDLS) {
m_curVolLUTKey = vol;
m_curDLS = dls;
if (dls)
m_curVolLUTVal = LookupDLSVolume(vol);
else
m_curVolLUTVal = LookupVolume(vol);
}
return m_curVolLUTVal;
}
void Voice::_destroy() {
Entity::_destroy();
for (auto& vox : m_childVoices)
vox->_destroy();
m_studio.reset();
m_backendVoice.reset();
m_curSample.reset();
m_sequencer.reset();
}
Voice::~Voice() {
// fprintf(stderr, "DEALLOC %d\n", m_vid);
}
Voice::Voice(Engine& engine, const AudioGroup& group, GroupId groupId, int vid, bool emitter, ObjToken<Studio> studio)
: Entity(engine, group, groupId), m_vid(vid), m_emitter(emitter), m_studio(studio) {
// fprintf(stderr, "ALLOC %d\n", m_vid);
}
Voice::Voice(Engine& engine, const AudioGroup& group, GroupId groupId, ObjectId oid, int vid, bool emitter,
ObjToken<Studio> studio)
: Entity(engine, group, groupId, oid), m_vid(vid), m_emitter(emitter), m_studio(studio) {
// fprintf(stderr, "ALLOC %d\n", m_vid);
}
void Voice::_macroSampleEnd() {
if (m_sampleEndTrap.macroId != 0xffff) {
if (m_sampleEndTrap.macroId == std::get<0>(m_state.m_pc.back())) {
std::get<2>(m_state.m_pc.back()) = std::get<1>(m_state.m_pc.back())->assertPC(m_sampleEndTrap.macroStep);
m_state.m_inWait = false;
} else
loadMacroObject(m_sampleEndTrap.macroId, m_sampleEndTrap.macroStep, m_state.m_ticksPerSec, m_state.m_initKey,
m_state.m_initVel, m_state.m_initMod);
} else
m_state.sampleEndNotify(*this);
}
bool Voice::_checkSamplePos(bool& looped) {
looped = false;
if (!m_curSample)
return true;
if (m_curSamplePos >= m_lastSamplePos) {
if (m_curSample->isLooped()) {
/* Turn over looped sample */
m_curSamplePos = m_curSample->m_loopStartSample;
if (m_curFormat == SampleFormat::DSP) {
m_prev1 = m_curSample->m_ADPCMParms.dsp.m_hist1;
m_prev2 = m_curSample->m_ADPCMParms.dsp.m_hist2;
}
looped = true;
} else {
/* Notify sample end */
_macroSampleEnd();
m_curSample = nullptr;
return true;
}
}
/* Looped samples issue sample end when ADSR envelope complete */
if (m_volAdsr.isComplete(*this)) {
_macroSampleEnd();
m_curSample = nullptr;
return true;
}
return false;
}
void Voice::_doKeyOff() {
m_voxState = VoiceState::KeyOff;
if (m_state.m_inWait && m_state.m_keyoffWait) {
if (m_volAdsr.isAdsrSet() || m_state.m_useAdsrControllers)
m_volAdsr.keyOff(*this);
if (m_pitchAdsr.isAdsrSet())
m_pitchAdsr.keyOff();
} else {
m_volAdsr.keyOff(*this);
m_pitchAdsr.keyOff();
}
m_state.keyoffNotify(*this);
}
void Voice::_setTotalPitch(int32_t cents, bool slew) {
// fprintf(stderr, "PITCH %d %d \n", cents, slew);
const int32_t interval = std::clamp(cents, 0, 12700) - m_curSample->getPitch() * 100;
const double ratio = std::exp2(interval / 1200.0) * m_dopplerRatio;
m_sampleRate = m_curSample->m_sampleRate * ratio;
m_backendVoice->setPitchRatio(ratio, slew);
}
bool Voice::_isRecursivelyDead() {
if (m_voxState != VoiceState::Dead)
return false;
for (auto& vox : m_childVoices)
if (!vox->_isRecursivelyDead())
return false;
return true;
}
void Voice::_bringOutYourDead() {
for (auto it = m_childVoices.begin(); it != m_childVoices.end();) {
Voice* vox = it->get();
vox->_bringOutYourDead();
if (vox->_isRecursivelyDead()) {
it = _destroyVoice(it);
continue;
}
++it;
}
}
ObjToken<Voice> Voice::_findVoice(int vid, ObjToken<Voice> thisPtr) {
if (m_vid == vid)
return thisPtr;
for (ObjToken<Voice>& vox : m_childVoices) {
ObjToken<Voice> ret = vox->_findVoice(vid, vox);
if (ret)
return ret;
}
return {};
}
std::unique_ptr<int8_t[]>& Voice::_ensureCtrlVals() {
if (m_ctrlValsSelf) {
return m_ctrlValsSelf;
}
m_ctrlValsSelf = std::make_unique<int8_t[]>(128);
return m_ctrlValsSelf;
}
std::list<ObjToken<Voice>>::iterator Voice::_allocateVoice(double sampleRate, bool dynamicPitch) {
amuse::ObjToken<Voice> tok =
MakeObj<Voice>(m_engine, m_audioGroup, m_groupId, m_engine.m_nextVid++, m_emitter, m_studio);
auto it = m_childVoices.emplace(m_childVoices.end(), tok);
m_childVoices.back()->m_backendVoice =
m_engine.getBackend().allocateVoice(*m_childVoices.back(), sampleRate, dynamicPitch);
return it;
}
std::list<ObjToken<Voice>>::iterator Voice::_destroyVoice(std::list<ObjToken<Voice>>::iterator it) {
if ((*it)->m_destroyed)
return m_childVoices.begin();
(*it)->_destroy();
return m_childVoices.erase(it);
}
template <typename T>
static T ApplyVolume(float vol, T samp) {
return samp * vol;
}
void Voice::_procSamplePre(int16_t& samp) {
double dt = 0.0;
/* Block linearized will use a larger `dt` for amplitude sampling;
* significantly reducing the processing expense */
switch (m_engine.m_ampMode) {
case AmplitudeMode::PerSample:
m_voiceSamples += 1;
dt = 1.0 / m_sampleRate;
break;
case AmplitudeMode::BlockLinearized: {
uint32_t rem = m_voiceSamples % 160;
m_voiceSamples += 1;
dt = m_sampleRate * 160;
if (rem != 0) {
/* Lerp within 160-sample block */
const float t = rem / 160.f;
const float l = std::clamp(m_lastLevel * (1.f - t) + m_nextLevel * t, 0.f, 1.f);
/* Apply total volume to sample using decibel scale */
samp = ApplyVolume(m_lerpedCache.getVolume(l * m_engine.m_masterVolume, m_dlsVol), samp);
return;
}
dt = 160.0 / m_sampleRate;
break;
}
}
m_voiceTime += dt;
/* Process active envelope */
if (m_envelopeTime >= 0.0) {
m_envelopeTime += dt;
const float start = m_envelopeStart;
const float end = m_envelopeEnd;
float t = std::clamp(float(m_envelopeTime / m_envelopeDur), 0.f, 1.f);
if (m_envelopeCurve && m_envelopeCurve->data.size() >= 128) {
t = m_envelopeCurve->data[t * 127.f] / 127.f;
}
m_envelopeVol = std::clamp((start * (1.0f - t)) + (end * t), 0.f, 1.f);
// printf("%d %f\n", m_vid, m_envelopeVol);
/* Done with envelope */
if (m_envelopeTime > m_envelopeDur)
m_envelopeTime = -1.f;
}
/* Dynamically evaluate per-sample SoundMacro parameters */
/* Process user volume slew */
if (m_engine.m_ampMode == AmplitudeMode::PerSample) {
if (m_targetUserVol != m_curUserVol) {
float samplesPer5Ms = m_sampleRate * 5.f / 1000.f;
if (samplesPer5Ms > 1.f) {
float adjRate = 1.f / samplesPer5Ms;
if (m_targetUserVol < m_curUserVol) {
m_curUserVol -= adjRate;
if (m_targetUserVol > m_curUserVol)
m_curUserVol = m_targetUserVol;
} else {
m_curUserVol += adjRate;
if (m_targetUserVol < m_curUserVol)
m_curUserVol = m_targetUserVol;
}
} else
m_curUserVol = m_targetUserVol;
}
} else
m_curUserVol = m_targetUserVol;
/* Factor in ADSR envelope state */
float adsr = m_volAdsr.advance(dt, *this);
m_lastLevel = m_nextLevel;
m_nextLevel = m_curUserVol * m_curVol * m_envelopeVol * adsr * (m_state.m_curVel / 127.f);
/* Apply tremolo */
if (m_state.m_tremoloSel && (m_tremoloScale || m_tremoloModScale)) {
float t = m_state.m_tremoloSel.evaluate(m_voiceTime, *this, m_state) / 127.f;
if (m_tremoloScale && m_tremoloModScale) {
float fac = (1.0f - t) + (m_tremoloScale * t);
float modT = m_state.m_modWheelSel ? (m_state.m_modWheelSel.evaluate(m_voiceTime, *this, m_state) / 127.f)
: (getCtrlValue(1) / 127.f);
float modFac = (1.0f - modT) + (m_tremoloModScale * modT);
m_nextLevel *= fac * modFac;
} else if (m_tremoloScale) {
float fac = (1.0f - t) + (m_tremoloScale * t);
m_nextLevel *= fac;
} else if (m_tremoloModScale) {
float modT = m_state.m_modWheelSel ? (m_state.m_modWheelSel.evaluate(m_voiceTime, *this, m_state) / 127.f)
: (getCtrlValue(1) / 127.f);
float modFac = (1.0f - modT) + (m_tremoloModScale * modT);
m_nextLevel *= modFac;
}
}
m_nextLevel = std::clamp(m_nextLevel, 0.f, 1.f);
/* Apply total volume to sample using decibel scale */
samp = ApplyVolume(m_nextLevelCache.getVolume(m_nextLevel * m_engine.m_masterVolume, m_dlsVol), samp);
}
template <typename T>
T Voice::_procSampleMaster(double time, T samp) {
const float evalVol = m_state.m_volumeSel ? (m_state.m_volumeSel.evaluate(time, *this, m_state) / 127.f) : 1.f;
return ApplyVolume(m_masterCache.getVolume(std::clamp(evalVol, 0.f, 1.f), m_dlsVol), samp);
}
template <typename T>
T Voice::_procSampleAuxA(double time, T samp) {
float evalVol = m_state.m_volumeSel ? (m_state.m_volumeSel.evaluate(time, *this, m_state) / 127.f) : 1.f;
evalVol *= m_state.m_reverbSel ? (m_state.m_reverbSel.evaluate(time, *this, m_state) / 127.f) : m_curReverbVol;
evalVol += m_state.m_preAuxASel ? (m_state.m_preAuxASel.evaluate(time, *this, m_state) / 127.f) : 0.f;
return ApplyVolume(m_auxACache.getVolume(std::clamp(evalVol, 0.f, 1.f), m_dlsVol), samp);
}
template <typename T>
T Voice::_procSampleAuxB(double time, T samp) {
float evalVol = m_state.m_volumeSel ? (m_state.m_volumeSel.evaluate(time, *this, m_state) / 127.f) : 1.f;
evalVol *= m_state.m_postAuxB ? (m_state.m_postAuxB.evaluate(time, *this, m_state) / 127.f) : m_curAuxBVol;
evalVol += m_state.m_preAuxBSel ? (m_state.m_preAuxBSel.evaluate(time, *this, m_state) / 127.f) : 0.f;
return ApplyVolume(m_auxBCache.getVolume(std::clamp(evalVol, 0.f, 1.f), m_dlsVol), samp);
}
uint32_t Voice::_GetBlockSampleCount(SampleFormat fmt) {
switch (fmt) {
default:
return 1;
case SampleFormat::DSP:
case SampleFormat::DSP_DRUM:
return 14;
case SampleFormat::N64:
return 64;
}
}
static float TriangleWave(float t) {
t = std::fmod(t, 1.f);
if (t < 0.25f)
return t / 0.25f;
if (t >= 0.75f)
return (t - 0.75f) / 0.25f - 1.f;
return (t - 0.25f) / 0.5f * -2.f + 1.f;
}
void Voice::preSupplyAudio(double dt) {
/* Process SoundMacro; bootstrapping sample if needed */
bool dead = m_state.advance(*this, dt);
/* Process per-block evaluators here */
if (m_state.m_pedalSel) {
bool pedal = m_state.m_pedalSel.evaluate(m_voiceTime, *this, m_state) >= 64.f;
if (pedal != m_sustained)
setPedal(pedal);
}
bool panDirty = false;
if (m_state.m_panSel) {
float evalPan = (m_state.m_panSel.evaluate(m_voiceTime, *this, m_state) - 64.f) / 63.f;
evalPan = std::clamp(evalPan, -1.f, 1.f);
if (evalPan != m_curPan) {
m_curPan = evalPan;
panDirty = true;
}
}
if (m_state.m_spanSel) {
float evalSpan = (m_state.m_spanSel.evaluate(m_voiceTime, *this, m_state) - 64.f) / 63.f;
evalSpan = std::clamp(evalSpan, -1.f, 1.f);
if (evalSpan != m_curSpan) {
m_curSpan = evalSpan;
panDirty = true;
}
}
if (panDirty)
_setPan(m_curPan);
if (m_state.m_pitchWheelSel) {
const float evalPWheel = (m_state.m_pitchWheelSel.evaluate(m_voiceTime, *this, m_state) - 64.f) / 63.f;
_setPitchWheel(std::clamp(evalPWheel, -1.f, 1.f));
}
/* Process active pan-sweep */
bool refresh = false;
if (!m_panningQueue.empty()) {
Panning& p = m_panningQueue.front();
p.m_time += dt;
const double start = (p.m_pos - 64) / 64.0;
const double end = (p.m_pos + p.m_width - 64) / 64.0;
const double t = std::clamp(p.m_time / p.m_dur, 0.0, 1.0);
_setPan(float((start * (1.0 - t)) + (end * t)));
refresh = true;
/* Done with panning */
if (p.m_time > p.m_dur)
m_panningQueue.pop();
}
/* Process active span-sweep */
if (!m_spanningQueue.empty()) {
Panning& s = m_spanningQueue.front();
s.m_time += dt;
const double start = (s.m_pos - 64) / 64.0;
const double end = (s.m_pos + s.m_width - 64) / 64.0;
const double t = std::clamp(s.m_time / s.m_dur, 0.0, 1.0);
_setSurroundPan(float((start * (1.0 - t)) + (end * t)));
refresh = true;
/* Done with spanning */
if (s.m_time > s.m_dur)
m_spanningQueue.pop();
}
/* Calculate total pitch */
int32_t newPitch = m_curPitch;
refresh |= m_pitchDirty;
m_pitchDirty = false;
if (m_portamentoTime >= 0.f) {
m_portamentoTime += dt;
const float t = std::clamp(m_portamentoTime / m_state.m_portamentoTime, 0.f, 1.f);
newPitch = (m_curPitch * (1.0f - t)) + (m_portamentoTarget * t);
refresh = true;
/* Done with portamento */
if (m_portamentoTime > m_state.m_portamentoTime) {
m_portamentoTime = -1.f;
m_curPitch = m_portamentoTarget;
}
}
if (m_pitchEnv) {
newPitch += m_pitchAdsr.advance(dt) * m_pitchEnvRange;
refresh = true;
}
/* Process vibrato */
if (m_vibratoTime >= 0.f) {
m_vibratoTime += dt;
float vibrato = TriangleWave(m_vibratoTime / m_vibratoPeriod);
if (m_vibratoModWheel) {
int32_t range = m_vibratoModLevel ? m_vibratoModLevel : m_vibratoLevel;
newPitch += range * vibrato * (m_state.m_curMod / 127.f);
} else
newPitch += m_vibratoLevel * vibrato;
refresh = true;
}
/* Process pitch sweep 1 */
if (m_pitchSweep1It < m_pitchSweep1Times) {
++m_pitchSweep1It;
m_pitchSweep1 = m_pitchSweep1Add * m_pitchSweep1It;
refresh = true;
} else if (m_pitchSweep1Times != 0) {
m_pitchSweep1It = 0;
m_pitchSweep1 = 0;
refresh = true;
}
/* Process pitch sweep 2 */
if (m_pitchSweep2It < m_pitchSweep2Times) {
++m_pitchSweep2It;
m_pitchSweep2 = m_pitchSweep2Add * m_pitchSweep2It;
refresh = true;
} else if (m_pitchSweep2Times != 0) {
m_pitchSweep2It = 0;
m_pitchSweep2 = 0;
refresh = true;
}
if (m_curSample && refresh) {
_setTotalPitch(newPitch + m_pitchSweep1 + m_pitchSweep2 + m_pitchWheelVal, m_needsSlew);
m_needsSlew = true;
}
if (dead && (!m_curSample || m_voxState == VoiceState::KeyOff) && m_sampleEndTrap.macroId == 0xffff &&
m_messageTrap.macroId == 0xffff && (!m_curSample || (m_curSample && m_volAdsr.isComplete(*this)))) {
m_voxState = VoiceState::Dead;
m_backendVoice->stop();
}
}
size_t Voice::supplyAudio(size_t samples, int16_t* data) {
uint32_t samplesRem = samples;
if (m_curSample) {
uint32_t blockSampleCount = _GetBlockSampleCount(m_curFormat);
uint32_t block;
bool looped = true;
while (looped && samplesRem) {
block = m_curSamplePos / blockSampleCount;
uint32_t rem = m_curSamplePos % blockSampleCount;
if (rem) {
uint32_t remCount = std::min(samplesRem, m_lastSamplePos - block * blockSampleCount);
uint32_t decSamples;
switch (m_curFormat) {
case SampleFormat::DSP: {
decSamples =
DSPDecompressFrameRanged(data, m_curSampleData + 8 * block, m_curSample->m_ADPCMParms.dsp.m_coefs,
&m_prev1, &m_prev2, rem, remCount);
break;
}
case SampleFormat::N64: {
decSamples = N64MusyXDecompressFrameRanged(data, m_curSampleData + 256 + 40 * block,
m_curSample->m_ADPCMParms.vadpcm.m_coefs, rem, remCount);
break;
}
case SampleFormat::PCM: {
const int16_t* pcm = reinterpret_cast<const int16_t*>(m_curSampleData);
remCount = std::min(samplesRem, m_lastSamplePos - m_curSamplePos);
for (uint32_t i = 0; i < remCount; ++i)
data[i] = SBig(pcm[m_curSamplePos + i]);
decSamples = remCount;
break;
}
case SampleFormat::PCM_PC: {
const int16_t* pcm = reinterpret_cast<const int16_t*>(m_curSampleData);
remCount = std::min(samplesRem, m_lastSamplePos - m_curSamplePos);
memmove(data, pcm + m_curSamplePos, remCount * sizeof(int16_t));
decSamples = remCount;
break;
}
default:
memset(data, 0, sizeof(int16_t) * samples);
return samples;
}
/* Per-sample processing */
for (uint32_t i = 0; i < decSamples; ++i) {
++m_curSamplePos;
_procSamplePre(data[i]);
}
samplesRem -= decSamples;
data += decSamples;
}
if (_checkSamplePos(looped)) {
if (samplesRem)
memset(data, 0, sizeof(int16_t) * samplesRem);
return samples;
}
if (looped)
continue;
while (samplesRem) {
block = m_curSamplePos / blockSampleCount;
uint32_t remCount = std::min(samplesRem, m_lastSamplePos - block * blockSampleCount);
uint32_t decSamples;
switch (m_curFormat) {
case SampleFormat::DSP: {
decSamples = DSPDecompressFrame(data, m_curSampleData + 8 * block, m_curSample->m_ADPCMParms.dsp.m_coefs,
&m_prev1, &m_prev2, remCount);
break;
}
case SampleFormat::N64: {
decSamples = N64MusyXDecompressFrame(data, m_curSampleData + 256 + 40 * block,
m_curSample->m_ADPCMParms.vadpcm.m_coefs, remCount);
break;
}
case SampleFormat::PCM: {
const int16_t* pcm = reinterpret_cast<const int16_t*>(m_curSampleData);
remCount = std::min(samplesRem, m_lastSamplePos - m_curSamplePos);
for (uint32_t i = 0; i < remCount; ++i)
data[i] = SBig(pcm[m_curSamplePos + i]);
decSamples = remCount;
break;
}
case SampleFormat::PCM_PC: {
const int16_t* pcm = reinterpret_cast<const int16_t*>(m_curSampleData);
remCount = std::min(samplesRem, m_lastSamplePos - m_curSamplePos);
memmove(data, pcm + m_curSamplePos, remCount * sizeof(int16_t));
decSamples = remCount;
break;
}
default:
memset(data, 0, sizeof(int16_t) * samples);
return samples;
}
/* Per-sample processing */
for (uint32_t i = 0; i < decSamples; ++i) {
++m_curSamplePos;
_procSamplePre(data[i]);
}
samplesRem -= decSamples;
data += decSamples;
if (_checkSamplePos(looped)) {
if (samplesRem)
memset(data, 0, sizeof(int16_t) * samplesRem);
return samples;
}
if (looped)
break;
}
}
} else {
_macroSampleEnd();
memset(data, 0, sizeof(int16_t) * samples);
}
if (m_voxState == VoiceState::Dead)
m_curSample.reset();
return samples;
}
void Voice::routeAudio(size_t frames, double dt, int busId, int16_t* in, int16_t* out) {
dt /= double(frames);
switch (busId) {
case 0:
default: {
for (uint32_t i = 0; i < frames; ++i)
out[i] = _procSampleMaster(dt * i + m_voiceTime, in[i]);
break;
}
case 1: {
for (uint32_t i = 0; i < frames; ++i)
out[i] = _procSampleAuxA(dt * i + m_voiceTime, in[i]);
break;
}
case 2: {
for (uint32_t i = 0; i < frames; ++i)
out[i] = _procSampleAuxB(dt * i + m_voiceTime, in[i]);
break;
}
}
}
void Voice::routeAudio(size_t frames, double dt, int busId, int32_t* in, int32_t* out) {
dt /= double(frames);
switch (busId) {
case 0:
default: {
for (uint32_t i = 0; i < frames; ++i)
out[i] = _procSampleMaster(dt * i + m_voiceTime, in[i]);
break;
}
case 1: {
for (uint32_t i = 0; i < frames; ++i)
out[i] = _procSampleAuxA(dt * i + m_voiceTime, in[i]);
break;
}
case 2: {
for (uint32_t i = 0; i < frames; ++i)
out[i] = _procSampleAuxB(dt * i + m_voiceTime, in[i]);
break;
}
}
}
void Voice::routeAudio(size_t frames, double dt, int busId, float* in, float* out) {
dt /= double(frames);
switch (busId) {
case 0:
default: {
for (uint32_t i = 0; i < frames; ++i)
out[i] = _procSampleMaster(dt * i + m_voiceTime, in[i]);
break;
}
case 1: {
for (uint32_t i = 0; i < frames; ++i)
out[i] = _procSampleAuxA(dt * i + m_voiceTime, in[i]);
break;
}
case 2: {
for (uint32_t i = 0; i < frames; ++i)
out[i] = _procSampleAuxB(dt * i + m_voiceTime, in[i]);
break;
}
}
}
int Voice::maxVid() const {
int maxVid = m_vid;
for (const ObjToken<Voice>& vox : m_childVoices)
maxVid = std::max(maxVid, vox->maxVid());
return maxVid;
}
ObjToken<Voice> Voice::_startChildMacro(ObjectId macroId, int macroStep, double ticksPerSec, uint8_t midiKey,
uint8_t midiVel, uint8_t midiMod, bool pushPc) {
std::list<ObjToken<Voice>>::iterator vox = _allocateVoice(NativeSampleRate, true);
if (!(*vox)->loadMacroObject(macroId, macroStep, ticksPerSec, midiKey, midiVel, midiMod, pushPc)) {
_destroyVoice(vox);
return {};
}
(*vox)->setVolume(m_targetUserVol);
(*vox)->setPan(m_curPan);
(*vox)->setSurroundPan(m_curSpan);
if (m_extCtrlVals)
(*vox)->installCtrlValues(m_extCtrlVals);
return *vox;
}
ObjToken<Voice> Voice::startChildMacro(int8_t addNote, ObjectId macroId, int macroStep) {
return _startChildMacro(macroId, macroStep, 1000.0, m_state.m_initKey + addNote, m_state.m_initVel,
m_state.m_initMod);
}
bool Voice::_loadSoundMacro(SoundMacroId id, const SoundMacro* macroData, int macroStep, double ticksPerSec,
uint8_t midiKey, uint8_t midiVel, uint8_t midiMod, bool pushPc) {
m_objectId = id;
if (m_state.m_pc.empty())
m_state.initialize(id, macroData, macroStep, ticksPerSec, midiKey, midiVel, midiMod);
else {
if (!pushPc)
m_state.m_pc.clear();
m_state.m_pc.emplace_back(id, macroData, macroData->assertPC(macroStep));
}
m_voxState = VoiceState::Playing;
m_backendVoice->start();
return true;
}
bool Voice::_loadKeymap(const Keymap* keymap, double ticksPerSec, uint8_t midiKey, uint8_t midiVel, uint8_t midiMod,
bool pushPc) {
const Keymap& km = keymap[midiKey];
midiKey += km.transpose;
bool ret = loadMacroObject(km.macro.id, 0, ticksPerSec, midiKey, midiVel, midiMod, pushPc);
m_curVol = 1.f;
if (km.pan == -128) {
_setSurroundPan(1.f);
} else {
_setPan((km.pan - 64) / 64.f);
_setSurroundPan(-1.f);
}
return ret;
}
bool Voice::_loadLayer(const std::vector<LayerMapping>& layer, double ticksPerSec, uint8_t midiKey, uint8_t midiVel,
uint8_t midiMod, bool pushPc) {
bool ret = false;
for (const LayerMapping& mapping : layer) {
if (midiKey >= mapping.keyLo && midiKey <= mapping.keyHi) {
uint8_t mappingKey = midiKey + mapping.transpose;
if (m_voxState != VoiceState::Playing) {
ret |= loadMacroObject(mapping.macro.id, 0, ticksPerSec, mappingKey, midiVel, midiMod, pushPc);
m_curUserVol = m_targetUserVol = mapping.volume / 127.f;
_setPan((mapping.pan - 64) / 64.f);
_setSurroundPan((mapping.span - 64) / 64.f);
} else {
ObjToken<Voice> vox = _startChildMacro(mapping.macro.id, 0, ticksPerSec, mappingKey, midiVel, midiMod, pushPc);
if (vox) {
vox->m_curUserVol = vox->m_targetUserVol = mapping.volume / 127.f;
vox->_setPan((mapping.pan - 64) / 64.f);
vox->_setSurroundPan((mapping.span - 64) / 64.f);
ret = true;
}
}
}
}
return ret;
}
bool Voice::loadMacroObject(SoundMacroId macroId, int macroStep, double ticksPerSec, uint8_t midiKey, uint8_t midiVel,
uint8_t midiMod, bool pushPc) {
if (m_destroyed)
return false;
const SoundMacro* macroData = m_audioGroup.getPool().soundMacro(macroId);
if (macroData)
return _loadSoundMacro(macroId, macroData, macroStep, ticksPerSec, midiKey, midiVel, midiMod, pushPc);
return false;
}
bool Voice::loadMacroObject(const SoundMacro* macro, int macroStep, double ticksPerSec, uint8_t midiKey,
uint8_t midiVel, uint8_t midiMod, bool pushPc) {
if (m_destroyed)
return false;
if (macro)
return _loadSoundMacro({}, macro, macroStep, ticksPerSec, midiKey, midiVel, midiMod, pushPc);
return false;
}
bool Voice::loadPageObject(ObjectId objectId, double ticksPerSec, uint8_t midiKey, uint8_t midiVel, uint8_t midiMod) {
if (m_destroyed)
return false;
if (objectId.id & 0x8000) {
const std::vector<LayerMapping>* layer = m_audioGroup.getPool().layer(objectId);
if (layer)
return _loadLayer(*layer, ticksPerSec, midiKey, midiVel, midiMod);
} else if (objectId.id & 0x4000) {
const Keymap* keymap = m_audioGroup.getPool().keymap(objectId);
if (keymap)
return _loadKeymap(keymap, ticksPerSec, midiKey, midiVel, midiMod);
} else {
const SoundMacro* sm = m_audioGroup.getPool().soundMacro(objectId);
if (sm)
return _loadSoundMacro(objectId, sm, 0, ticksPerSec, midiKey, midiVel, midiMod);
}
return false;
}
void Voice::_macroKeyOff() {
if (m_voxState == VoiceState::Playing) {
if (m_sustained)
m_sustainKeyOff = true;
else
_doKeyOff();
}
}
void Voice::keyOff() {
if (m_destroyed)
return;
if (m_keyoffTrap.macroId != 0xffff) {
if (m_keyoffTrap.macroId == std::get<0>(m_state.m_pc.back())) {
std::get<2>(m_state.m_pc.back()) = std::get<1>(m_state.m_pc.back())->assertPC(m_keyoffTrap.macroStep);
m_state.m_inWait = false;
} else
loadMacroObject(m_keyoffTrap.macroId, m_keyoffTrap.macroStep, m_state.m_ticksPerSec, m_state.m_initKey,
m_state.m_initVel, m_state.m_initMod);
} else
_macroKeyOff();
for (const ObjToken<Voice>& vox : m_childVoices)
vox->keyOff();
}
void Voice::message(int32_t val) {
if (m_destroyed)
return;
m_latestMessage = val;
if (m_messageTrap.macroId != 0xffff) {
if (m_messageTrap.macroId == std::get<0>(m_state.m_pc.back()))
std::get<2>(m_state.m_pc.back()) = std::get<1>(m_state.m_pc.back())->assertPC(m_messageTrap.macroStep);
else
loadMacroObject(m_messageTrap.macroId, m_messageTrap.macroStep, m_state.m_ticksPerSec, m_state.m_initKey,
m_state.m_initVel, m_state.m_initMod);
}
}
void Voice::startSample(SampleId sampId, int32_t offset) {
if (m_destroyed)
return;
if (const SampleEntry* sample = m_audioGroup.getSample(sampId)) {
std::tie(m_curSample, m_curSampleData) = m_audioGroup.getSampleData(sampId, sample);
m_state.m_sampleEnd = false;
m_sampleRate = m_curSample->m_sampleRate;
m_curPitch = m_curSample->getPitch();
m_pitchDirty = true;
_setPitchWheel(m_curPitchWheel);
m_backendVoice->resetSampleRate(m_curSample->m_sampleRate);
m_needsSlew = false;
const int32_t numSamples = m_curSample->getNumSamples();
if (offset) {
if (m_curSample->isLooped()) {
if (offset > int32_t(m_curSample->m_loopStartSample))
offset = ((offset - m_curSample->m_loopStartSample) % m_curSample->m_loopLengthSamples) +
m_curSample->m_loopStartSample;
} else {
offset = std::clamp(offset, 0, numSamples);
}
}
m_curSamplePos = offset;
m_prev1 = 0;
m_prev2 = 0;
m_curFormat = m_curSample->getSampleFormat();
if (m_curFormat == SampleFormat::DSP_DRUM)
m_curFormat = SampleFormat::DSP;
m_lastSamplePos =
m_curSample->isLooped() ? (m_curSample->m_loopStartSample + m_curSample->m_loopLengthSamples) : numSamples;
if (m_lastSamplePos)
--m_lastSamplePos;
bool looped;
_checkSamplePos(looped);
/* Seek DSPADPCM state if needed */
if (m_curSample && m_curSamplePos && m_curFormat == SampleFormat::DSP) {
uint32_t block = m_curSamplePos / 14;
uint32_t rem = m_curSamplePos % 14;
for (uint32_t b = 0; b < block; ++b)
DSPDecompressFrameStateOnly(m_curSampleData + 8 * b, m_curSample->m_ADPCMParms.dsp.m_coefs, &m_prev1, &m_prev2,
14);
if (rem)
DSPDecompressFrameStateOnly(m_curSampleData + 8 * block, m_curSample->m_ADPCMParms.dsp.m_coefs, &m_prev1,
&m_prev2, rem);
}
}
}
void Voice::stopSample() { m_curSample.reset(); }
void Voice::setVolume(float vol) {
if (m_destroyed)
return;
m_targetUserVol = std::clamp(vol, 0.f, 1.f);
for (ObjToken<Voice>& vox : m_childVoices)
vox->setVolume(vol);
}
std::array<float, 8> Voice::_panLaw(float frontPan, float backPan, float totalSpan) const {
std::array<float, 8> coefs{};
/* -3dB panning law for various channel configs */
switch (m_engine.m_channelSet) {
case AudioChannelSet::Stereo:
default:
/* Left */
coefs[0] = std::sqrt(-frontPan * 0.5f + 0.5f);
/* Right */
coefs[1] = std::sqrt(frontPan * 0.5f + 0.5f);
break;
case AudioChannelSet::Quad:
/* Left */
coefs[0] = -frontPan * 0.5f + 0.5f;
coefs[0] *= -totalSpan * 0.5f + 0.5f;
coefs[0] = std::sqrt(coefs[0]);
/* Right */
coefs[1] = frontPan * 0.5f + 0.5f;
coefs[1] *= -totalSpan * 0.5f + 0.5f;
coefs[1] = std::sqrt(coefs[1]);
/* Back Left */
coefs[2] = -backPan * 0.5f + 0.5f;
coefs[2] *= totalSpan * 0.5f + 0.5f;
coefs[2] = std::sqrt(coefs[2]);
/* Back Right */
coefs[3] = backPan * 0.5f + 0.5f;
coefs[3] *= totalSpan * 0.5f + 0.5f;
coefs[3] = std::sqrt(coefs[3]);
break;
case AudioChannelSet::Surround51:
/* Left */
coefs[0] = (frontPan <= 0.f) ? -frontPan : 0.f;
coefs[0] *= -totalSpan * 0.5f + 0.5f;
coefs[0] = std::sqrt(coefs[0]);
/* Right */
coefs[1] = (frontPan >= 0.f) ? frontPan : 0.f;
coefs[1] *= -totalSpan * 0.5f + 0.5f;
coefs[1] = std::sqrt(coefs[1]);
/* Back Left */
coefs[2] = -backPan * 0.5f + 0.5f;
coefs[2] *= totalSpan * 0.5f + 0.5f;
coefs[2] = std::sqrt(coefs[2]);
/* Back Right */
coefs[3] = backPan * 0.5f + 0.5f;
coefs[3] *= totalSpan * 0.5f + 0.5f;
coefs[3] = std::sqrt(coefs[3]);
/* Center */
coefs[4] = 1.f - std::fabs(frontPan);
coefs[4] *= -totalSpan * 0.5f + 0.5f;
coefs[4] = std::sqrt(coefs[4]);
/* LFE */
coefs[5] = 0.25f;
break;
case AudioChannelSet::Surround71:
/* Left */
coefs[0] = (frontPan <= 0.f) ? -frontPan : 0.f;
coefs[0] *= (totalSpan <= 0.f) ? -totalSpan : 0.f;
coefs[0] = std::sqrt(coefs[0]);
/* Right */
coefs[1] = (frontPan >= 0.f) ? frontPan : 0.f;
coefs[1] *= (totalSpan <= 0.f) ? -totalSpan : 0.f;
coefs[1] = std::sqrt(coefs[1]);
/* Back Left */
coefs[2] = -backPan * 0.5f + 0.5f;
coefs[2] *= (totalSpan >= 0.f) ? totalSpan : 0.f;
coefs[2] = std::sqrt(coefs[2]);
/* Back Right */
coefs[3] = backPan * 0.5f + 0.5f;
coefs[3] *= (totalSpan >= 0.f) ? totalSpan : 0.f;
coefs[3] = std::sqrt(coefs[3]);
/* Center */
coefs[4] = 1.f - std::fabs(frontPan);
coefs[4] *= (totalSpan <= 0.f) ? -totalSpan : 0.f;
coefs[4] = std::sqrt(coefs[4]);
/* LFE */
coefs[5] = 0.25f;
/* Side Left */
coefs[6] = -backPan * 0.5f + 0.5f;
coefs[6] *= 1.f - std::fabs(totalSpan);
coefs[6] = std::sqrt(coefs[6]);
/* Side Right */
coefs[7] = backPan * 0.5f + 0.5f;
coefs[7] *= 1.f - std::fabs(totalSpan);
coefs[7] = std::sqrt(coefs[7]);
break;
}
return coefs;
}
void Voice::_setPan(float pan) {
if (m_destroyed || m_emitter) {
return;
}
m_curPan = std::clamp(pan, -1.f, 1.f);
const float totalPan = std::clamp(m_curPan, -1.f, 1.f);
const float totalSpan = std::clamp(m_curSpan, -1.f, 1.f);
const auto coefs = _panLaw(totalPan, totalPan, totalSpan);
_setChannelCoefs(coefs);
}
void Voice::setPan(float pan) {
if (m_destroyed)
return;
_setPan(pan);
for (ObjToken<Voice>& vox : m_childVoices)
vox->setPan(pan);
}
void Voice::_setSurroundPan(float span) {
m_curSpan = std::clamp(span, -1.f, 1.f);
_setPan(m_curPan);
}
void Voice::setSurroundPan(float span) {
if (m_destroyed)
return;
_setSurroundPan(span);
for (ObjToken<Voice>& vox : m_childVoices)
vox->setSurroundPan(span);
}
void Voice::_setChannelCoefs(const std::array<float, 8>& coefs) {
m_backendVoice->setChannelLevels(m_studio->getMaster().m_backendSubmix.get(), coefs, true);
m_backendVoice->setChannelLevels(m_studio->getAuxA().m_backendSubmix.get(), coefs, true);
m_backendVoice->setChannelLevels(m_studio->getAuxB().m_backendSubmix.get(), coefs, true);
}
void Voice::setChannelCoefs(const std::array<float, 8>& coefs) {
if (m_destroyed)
return;
_setChannelCoefs(coefs);
for (ObjToken<Voice>& vox : m_childVoices)
vox->setChannelCoefs(coefs);
}
void Voice::startEnvelope(double dur, float vol, const Curve* envCurve) {
m_envelopeTime = 0.f;
m_envelopeDur = dur;
m_envelopeStart = std::clamp(m_envelopeVol, 0.f, 1.f);
m_envelopeEnd = std::clamp(vol, 0.f, 1.f);
m_envelopeCurve = envCurve;
}
void Voice::startFadeIn(double dur, float vol, const Curve* envCurve) {
m_envelopeTime = 0.f;
m_envelopeDur = dur;
m_envelopeStart = 0.f;
m_envelopeEnd = std::clamp(vol, 0.f, 1.f);
m_envelopeCurve = envCurve;
}
void Voice::startPanning(double dur, uint8_t panPos, int8_t panWidth) {
m_panningQueue.push({0.f, dur, panPos, panWidth});
}
void Voice::startSpanning(double dur, uint8_t spanPos, int8_t spanWidth) {
m_spanningQueue.push({0.f, dur, spanPos, spanWidth});
}
void Voice::setPitchKey(int32_t cents) {
m_curPitch = cents;
m_pitchDirty = true;
}
void Voice::setPedal(bool pedal) {
if (m_destroyed)
return;
if (m_sustained && !pedal && m_sustainKeyOff) {
m_sustainKeyOff = false;
_doKeyOff();
}
m_sustained = pedal;
for (ObjToken<Voice>& vox : m_childVoices)
vox->setPedal(pedal);
}
void Voice::setDoppler(float) {}
void Voice::setVibrato(int32_t level, bool modScale, float period) {
m_vibratoTime = std::fabs(period) < FLT_EPSILON ? -1.f : 0.f;
m_vibratoLevel = level;
m_vibratoModWheel = modScale;
m_vibratoPeriod = period;
}
void Voice::setMod2VibratoRange(int32_t modLevel) { m_vibratoModLevel = modLevel; }
void Voice::setTremolo(float tremoloScale, float tremoloModScale) {
m_tremoloScale = tremoloScale;
m_tremoloModScale = tremoloModScale;
}
void Voice::setPitchSweep1(uint8_t times, int16_t add) {
m_pitchSweep1 = 0;
m_pitchSweep1It = 0;
m_pitchSweep1Times = times;
m_pitchSweep1Add = add;
}
void Voice::setPitchSweep2(uint8_t times, int16_t add) {
m_pitchSweep2 = 0;
m_pitchSweep2It = 0;
m_pitchSweep2Times = times;
m_pitchSweep2Add = add;
}
void Voice::setReverbVol(float rvol) {
if (m_destroyed) {
return;
}
m_curReverbVol = std::clamp(rvol, 0.f, 1.f);
for (ObjToken<Voice>& vox : m_childVoices) {
vox->setReverbVol(rvol);
}
}
void Voice::setAuxBVol(float bvol) {
if (m_destroyed) {
return;
}
m_curAuxBVol = std::clamp(bvol, 0.f, 1.f);
for (ObjToken<Voice>& vox : m_childVoices) {
vox->setAuxBVol(bvol);
}
}
void Voice::setAdsr(ObjectId adsrId, bool dls) {
if (m_destroyed)
return;
if (dls) {
const ADSRDLS* adsr = m_audioGroup.getPool().tableAsAdsrDLS(adsrId);
if (adsr) {
m_volAdsr.reset(adsr, m_state.m_initKey, m_state.m_initVel);
if (m_voxState == VoiceState::KeyOff)
m_volAdsr.keyOff(*this);
}
} else {
const ADSR* adsr = m_audioGroup.getPool().tableAsAdsr(adsrId);
if (adsr) {
m_volAdsr.reset(adsr);
if (m_voxState == VoiceState::KeyOff)
m_volAdsr.keyOff(*this);
}
}
}
void Voice::setPitchFrequency(uint32_t hz, uint16_t fine) {
if (m_destroyed)
return;
m_sampleRate = hz + fine / 65536.0;
m_backendVoice->setPitchRatio(1.0, false);
m_backendVoice->resetSampleRate(m_sampleRate);
}
void Voice::setPitchAdsr(ObjectId adsrId, int32_t cents) {
if (m_destroyed)
return;
const ADSRDLS* adsr = m_audioGroup.getPool().tableAsAdsrDLS(adsrId);
if (adsr) {
m_pitchAdsr.reset(adsr, m_state.m_initKey, m_state.m_initVel);
m_pitchEnvRange = cents;
m_pitchEnv = true;
}
}
void Voice::_setPitchWheel(float pitchWheel) {
if (m_destroyed)
return;
if (pitchWheel > 0.f)
m_pitchWheelVal = m_pitchWheelUp * m_curPitchWheel;
else if (pitchWheel < 0.f)
m_pitchWheelVal = m_pitchWheelDown * m_curPitchWheel;
else
m_pitchWheelVal = 0;
m_pitchDirty = true;
}
void Voice::setPitchWheel(float pitchWheel) {
if (m_destroyed) {
return;
}
m_curPitchWheel = std::clamp(pitchWheel, -1.f, 1.f);
_setPitchWheel(m_curPitchWheel);
for (ObjToken<Voice>& vox : m_childVoices) {
vox->setPitchWheel(pitchWheel);
}
}
void Voice::setPitchWheelRange(int8_t up, int8_t down) {
if (m_destroyed)
return;
m_pitchWheelUp = up * 100;
m_pitchWheelDown = down * 100;
_setPitchWheel(m_curPitchWheel);
}
void Voice::setAftertouch(uint8_t aftertouch) {
if (m_destroyed)
return;
m_curAftertouch = aftertouch;
for (ObjToken<Voice>& vox : m_childVoices)
vox->setAftertouch(aftertouch);
}
bool Voice::doPortamento(uint8_t newNote) {
bool pState;
switch (m_state.m_portamentoMode) {
case SoundMacro::CmdPortamento::PortState::Disable:
default:
pState = false;
break;
case SoundMacro::CmdPortamento::PortState::Enable:
pState = true;
break;
case SoundMacro::CmdPortamento::PortState::MIDIControlled:
pState = m_state.m_portamentoSel ? (m_state.m_portamentoSel.evaluate(m_voiceTime, *this, m_state) >= 64.f)
: (getCtrlValue(65) >= 64);
break;
}
if (!pState)
return false;
m_portamentoTime = 0.f;
m_portamentoTarget = newNote * 100;
m_state.m_initKey = newNote;
return true;
}
void Voice::_notifyCtrlChange(uint8_t ctrl, int8_t val) {
if (ctrl == 0x40) {
setPedal(val >= 0x40);
} else if (ctrl == 0x5b) {
setReverbVol(val / 127.f);
} else if (ctrl == 0x5d) {
setAuxBVol(val / 127.f);
} else if (ctrl == 0x1) {
m_state.m_curMod = uint8_t(val);
} else if (ctrl == 0x64) {
// RPN LSB
m_rpn &= ~0x7f;
m_rpn |= val;
} else if (ctrl == 0x65) {
// RPN MSB
m_rpn &= ~0x3f80;
m_rpn |= val << 7;
} else if (ctrl == 0x6) {
if (m_rpn == 0)
m_pitchWheelUp = m_pitchWheelDown = val * 100;
} else if (ctrl == 0x60) {
if (m_rpn == 0) {
m_pitchWheelUp += 100;
m_pitchWheelDown += 100;
}
} else if (ctrl == 0x61) {
if (m_rpn == 0) {
m_pitchWheelUp -= 100;
m_pitchWheelDown -= 100;
}
}
for (ObjToken<Voice>& vox : m_childVoices)
vox->_notifyCtrlChange(ctrl, val);
}
size_t Voice::getTotalVoices() const {
size_t ret = 1;
for (const ObjToken<Voice>& vox : m_childVoices)
ret += vox->getTotalVoices();
return ret;
}
void Voice::kill() {
if (m_destroyed)
return;
m_voxState = VoiceState::Dead;
m_backendVoice->stop();
for (const ObjToken<Voice>& vox : m_childVoices)
vox->kill();
}
} // namespace amuse