Audio: Replace blip_buf usage with a hermite interpolation resampler

Improves sound - blip_buf was altering the audio output quite a bit compared to the DSP's raw output
This commit is contained in:
Sour 2020-02-29 15:58:11 -05:00
parent c6dfcd1900
commit 206214ce2b
15 changed files with 155 additions and 497 deletions

View file

@ -1,6 +1,7 @@
#include "stdafx.h"
#include "PcmReader.h"
#include "../Utilities/VirtualFile.h"
#include "../Utilities/HermiteResampler.h"
PcmReader::PcmReader()
{
@ -10,15 +11,11 @@ PcmReader::PcmReader()
_prevRight = 0;
_loopOffset = 8;
_sampleRate = 0;
_blipLeft = blip_new(10000);
_blipRight = blip_new(10000);
_outputBuffer = new int16_t[20000];
}
PcmReader::~PcmReader()
{
blip_delete(_blipLeft);
blip_delete(_blipRight);
delete[] _outputBuffer;
}
@ -51,8 +48,9 @@ bool PcmReader::Init(string filename, bool loop, uint32_t startOffset)
_fileOffset = startOffset;
_file.seekg(_fileOffset, ios::beg);
blip_clear(_blipLeft);
blip_clear(_blipRight);
_leftoverSampleCount = 0;
_pcmBuffer.clear();
_resampler.Reset();
return true;
} else {
@ -71,10 +69,7 @@ void PcmReader::SetSampleRate(uint32_t sampleRate)
if(sampleRate != _sampleRate) {
_sampleRate = sampleRate;
blip_clear(_blipLeft);
blip_clear(_blipRight);
blip_set_rates(_blipLeft, PcmReader::PcmSampleRate, _sampleRate);
blip_set_rates(_blipRight, PcmReader::PcmSampleRate, _sampleRate);
_resampler.SetSampleRates(PcmReader::PcmSampleRate, _sampleRate);
}
}
@ -104,8 +99,8 @@ void PcmReader::LoadSamples(uint32_t samplesToLoad)
for(uint32_t i = _fileOffset; i < _fileSize && samplesRead < samplesToLoad; i+=4) {
ReadSample(left, right);
blip_add_delta(_blipLeft, samplesRead, left - _prevLeft);
blip_add_delta(_blipRight, samplesRead, right - _prevRight);
_pcmBuffer.push_back(left);
_pcmBuffer.push_back(right);
_prevLeft = left;
_prevRight = right;
@ -123,9 +118,6 @@ void PcmReader::LoadSamples(uint32_t samplesToLoad)
}
}
}
blip_end_frame(_blipLeft, samplesRead);
blip_end_frame(_blipRight, samplesRead);
}
void PcmReader::ApplySamples(int16_t *buffer, size_t sampleCount, uint8_t volume)
@ -134,18 +126,26 @@ void PcmReader::ApplySamples(int16_t *buffer, size_t sampleCount, uint8_t volume
return;
}
int32_t samplesNeeded = (int32_t)sampleCount - blip_samples_avail(_blipLeft);
int32_t samplesNeeded = (int32_t)sampleCount - _leftoverSampleCount;
if(samplesNeeded > 0) {
uint32_t samplesToLoad = samplesNeeded * PcmReader::PcmSampleRate / _sampleRate + 1;
uint32_t samplesToLoad = samplesNeeded * PcmReader::PcmSampleRate / _sampleRate + 2;
LoadSamples(samplesToLoad);
}
int samplesRead = blip_read_samples(_blipLeft, _outputBuffer, (int)sampleCount, 1);
blip_read_samples(_blipRight, _outputBuffer + 1, (int)sampleCount, 1);
uint32_t samplesRead = _resampler.Resample(_pcmBuffer.data(), (uint32_t)_pcmBuffer.size() / 2, _outputBuffer + _leftoverSampleCount*2);
_pcmBuffer.clear();
for(size_t i = 0, len = samplesRead * 2; i < len; i++) {
uint32_t samplesToProcess = std::min<uint32_t>((uint32_t)sampleCount * 2, (samplesRead + _leftoverSampleCount) * 2);
for(uint32_t i = 0; i < samplesToProcess; i++) {
buffer[i] += (int16_t)((int32_t)_outputBuffer[i] * volume / 255);
}
//Calculate count of extra samples that couldn't be mixed with the rest of the audio and copy them to the beginning of the buffer
//These will be mixed on the next call to ApplySamples
_leftoverSampleCount = std::max(0, (int32_t)(samplesRead + _leftoverSampleCount) - (int32_t)sampleCount);
for(uint32_t i = 0; i < _leftoverSampleCount*2; i++) {
_outputBuffer[i] = _outputBuffer[samplesToProcess + i];
}
}
uint32_t PcmReader::GetOffset()

View file

@ -1,7 +1,7 @@
#pragma once
#include "stdafx.h"
#include "../Utilities/stb_vorbis.h"
#include "../Utilities/blip_buf.h"
#include "../Utilities/HermiteResampler.h"
class PcmReader
{
@ -22,8 +22,9 @@ private:
bool _loop;
bool _done;
blip_t* _blipLeft;
blip_t* _blipRight;
HermiteResampler _resampler;
vector<int16_t> _pcmBuffer;
uint32_t _leftoverSampleCount = 0;
uint32_t _sampleRate;

View file

@ -9,7 +9,6 @@
#include "Spc.h"
#include "Msu1.h"
#include "../Utilities/Equalizer.h"
#include "../Utilities/blip_buf.h"
SoundMixer::SoundMixer(Console *console)
{

View file

@ -5,19 +5,15 @@
#include "EmuSettings.h"
#include "SoundMixer.h"
#include "VideoRenderer.h"
#include "../Utilities/blip_buf.h"
#include "../Utilities/HermiteResampler.h"
SoundResampler::SoundResampler(Console *console)
{
_console = console;
_blipBufLeft = blip_new(Spc::SpcSampleRate);
_blipBufRight = blip_new(Spc::SpcSampleRate);
}
SoundResampler::~SoundResampler()
{
blip_delete(_blipBufLeft);
blip_delete(_blipBufRight);
}
double SoundResampler::GetRateAdjustment()
@ -76,45 +72,26 @@ double SoundResampler::GetTargetRateAdjustment()
void SoundResampler::UpdateTargetSampleRate(uint32_t sampleRate)
{
uint32_t spcSampleRate = Spc::SpcSampleRate;
double spcSampleRate = Spc::SpcSampleRate;
if(_console->GetSettings()->GetVideoConfig().IntegerFpsMode) {
//Adjust sample rate when running at 60.0 fps instead of 60.1
switch(_console->GetRegion()) {
default:
case ConsoleRegion::Ntsc: spcSampleRate = (uint32_t)(Spc::SpcSampleRate * (60.0 / 60.0988118623484)); break;
case ConsoleRegion::Pal: spcSampleRate = (uint32_t)(Spc::SpcSampleRate * (50.0 / 50.00697796826829)); break;
case ConsoleRegion::Ntsc: spcSampleRate = Spc::SpcSampleRate * (60.0 / 60.0988118623484); break;
case ConsoleRegion::Pal: spcSampleRate = Spc::SpcSampleRate * (50.0 / 50.00697796826829); break;
}
}
double targetRate = sampleRate * GetTargetRateAdjustment();
if(targetRate != _previousTargetRate || spcSampleRate != _prevSpcSampleRate) {
blip_set_rates(_blipBufLeft, spcSampleRate, targetRate);
blip_set_rates(_blipBufRight, spcSampleRate, targetRate);
_previousTargetRate = targetRate;
_prevSpcSampleRate = spcSampleRate;
_resampler.SetSampleRates(spcSampleRate, targetRate);
}
}
uint32_t SoundResampler::Resample(int16_t *inSamples, uint32_t sampleCount, uint32_t sampleRate, int16_t *outSamples)
{
UpdateTargetSampleRate(sampleRate);
blip_add_delta(_blipBufLeft, 0, inSamples[0] - _lastSampleLeft);
blip_add_delta(_blipBufRight, 0, inSamples[1] - _lastSampleRight);
for(uint32_t i = 1; i < sampleCount; i++) {
blip_add_delta(_blipBufLeft, i, inSamples[i * 2] - inSamples[i * 2 - 2]);
blip_add_delta(_blipBufRight, i, inSamples[i * 2 + 1] - inSamples[i * 2 - 1]);
}
_lastSampleLeft = inSamples[(sampleCount - 1) * 2];
_lastSampleRight = inSamples[(sampleCount - 1) * 2 + 1];
blip_end_frame(_blipBufLeft, sampleCount);
blip_end_frame(_blipBufRight, sampleCount);
uint32_t resampledCount = blip_read_samples(_blipBufLeft, outSamples, Spc::SpcSampleRate, true);
blip_read_samples(_blipBufRight, outSamples + 1, Spc::SpcSampleRate, true);
return resampledCount;
return _resampler.Resample(inSamples, sampleCount, outSamples);
}

View file

@ -1,8 +1,8 @@
#pragma once
#include "stdafx.h"
#include "../Utilities/HermiteResampler.h"
class Console;
struct blip_t;
class SoundResampler
{
@ -11,13 +11,10 @@ private:
double _rateAdjustment = 1.0;
double _previousTargetRate = 0;
uint32_t _prevSpcSampleRate = 0;
double _prevSpcSampleRate = 0;
int32_t _underTarget = 0;
blip_t *_blipBufLeft = nullptr;
blip_t *_blipBufRight = nullptr;
int16_t _lastSampleLeft = 0;
int16_t _lastSampleRight = 0;
HermiteResampler _resampler;
double GetTargetRateAdjustment();
void UpdateTargetSampleRate(uint32_t sampleRate);

View file

@ -126,23 +126,23 @@ SOURCES_CXX := $(LIBRETRO_DIR)/libretro.cpp \
$(UTIL_DIR)/AutoResetEvent.cpp \
$(UTIL_DIR)/AviRecorder.cpp \
$(UTIL_DIR)/AviWriter.cpp \
$(UTIL_DIR)/blip_buf.cpp \
$(UTIL_DIR)/BpsPatcher.cpp \
$(UTIL_DIR)/CamstudioCodec.cpp \
$(UTIL_DIR)/CRC32.cpp \
$(UTIL_DIR)/Equalizer.cpp \
$(UTIL_DIR)/FolderUtilities.cpp \
$(UTIL_DIR)/GifRecorder.cpp \
$(UTIL_DIR)/HermiteResampler.cpp \
$(UTIL_DIR)/HexUtilities.cpp \
$(UTIL_DIR)/IpsPatcher.cpp \
$(UTIL_DIR)/md5.cpp \
$(UTIL_DIR)/miniz.cpp \
$(UTIL_DIR)/snes_ntsc.cpp \
$(UTIL_DIR)/PlatformUtilities.cpp \
$(UTIL_DIR)/PNGHelper.cpp \
$(UTIL_DIR)/Serializer.cpp \
$(UTIL_DIR)/sha1.cpp \
$(UTIL_DIR)/SimpleLock.cpp \
$(UTIL_DIR)/snes_ntsc.cpp \
$(UTIL_DIR)/Socket.cpp \
$(UTIL_DIR)/stb_vorbis.cpp \
$(UTIL_DIR)/stdafx.cpp \

View file

@ -7,6 +7,7 @@
#include "LibretroMessageManager.h"
#include "libretro.h"
#include "../Core/Console.h"
#include "../Core/Spc.h"
#include "../Core/BaseCartridge.h"
#include "../Core/MemoryManager.h"
#include "../Core/VideoDecoder.h"
@ -81,7 +82,7 @@ extern "C" {
AudioConfig audioConfig = _console->GetSettings()->GetAudioConfig();
audioConfig.DisableDynamicSampleRate = true;
audioConfig.SampleRate = 32000;
audioConfig.SampleRate = Spc::SpcSampleRate;
_console->GetSettings()->SetAudioConfig(audioConfig);
PreferencesConfig preferences = _console->GetSettings()->GetPreferences();

View file

@ -0,0 +1,76 @@
#include "stdafx.h"
#include "HermiteResampler.h"
//Adapted from http://paulbourke.net/miscellaneous/interpolation/
//Original author: Paul Bourke ("Any source code found here may be freely used provided credits are given to the author.")
int16_t HermiteResampler::HermiteInterpolate(double values[4], double mu)
{
constexpr double tension = 0; //Tension: 1 is high, 0 normal, -1 is low
constexpr double bias = 0; //Bias: 0 is even, positive is towards first segment, negative towards the other
double m0, m1, mu2, mu3;
double a0, a1, a2, a3;
mu2 = mu * mu;
mu3 = mu2 * mu;
m0 = (values[1] - values[0]) * (1 + bias) * (1 - tension) / 2;
m0 += (values[2] - values[1]) * (1 - bias) * (1 - tension) / 2;
m1 = (values[2] - values[1]) * (1 + bias) * (1 - tension) / 2;
m1 += (values[3] - values[2]) * (1 - bias) * (1 - tension) / 2;
a0 = 2 * mu3 - 3 * mu2 + 1;
a1 = mu3 - 2 * mu2 + mu;
a2 = mu3 - mu2;
a3 = -2 * mu3 + 3 * mu2;
double output = a0 * values[1] + a1 * m0 + a2 * m1 + a3 * values[2];
return (int16_t)std::max(std::min(output, 32767.0), -32768.0);
}
void HermiteResampler::PushSample(double prevValues[4], int16_t sample)
{
prevValues[0] = prevValues[1];
prevValues[1] = prevValues[2];
prevValues[2] = prevValues[3];
prevValues[3] = (double)sample;
}
void HermiteResampler::Reset()
{
for(int i = 0; i < 4; i++) {
_prevLeft[i] = 0.0;
_prevRight[i] = 0.0;
}
_fraction = 0.0;
}
void HermiteResampler::SetSampleRates(double srcRate, double dstRate)
{
_rateRatio = srcRate / dstRate;
}
uint32_t HermiteResampler::Resample(int16_t* in, uint32_t inSampleCount, int16_t* out)
{
if(_rateRatio == 1.0) {
memcpy(out, in, inSampleCount * 2 * sizeof(int16_t));
return inSampleCount;
}
uint32_t outPos = 0;
for(uint32_t i = 0; i < inSampleCount * 2; i += 2) {
while(_fraction <= 1.0) {
//Generate interpolated samples until we have enough samples for the current source sample
out[outPos] = HermiteInterpolate(_prevLeft, _fraction);
out[outPos + 1] = HermiteInterpolate(_prevRight, _fraction);
outPos += 2;
_fraction += _rateRatio;
}
//Move to the next source sample
PushSample(_prevLeft, in[i]);
PushSample(_prevRight, in[i + 1]);
_fraction -= 1.0;
}
return outPos / 2;
}

View file

@ -0,0 +1,20 @@
#pragma once
#include "stdafx.h"
class HermiteResampler
{
private:
double _prevLeft[4] = {};
double _prevRight[4] = {};
double _rateRatio = 1.0;
double _fraction = 0.0;
__forceinline int16_t HermiteInterpolate(double values[4], double mu);
__forceinline void PushSample(double prevValues[4], int16_t sample);
public:
void Reset();
void SetSampleRates(double srcRate, double dstRate);
uint32_t Resample(int16_t* in, uint32_t inSampleCount, int16_t* out);
};

View file

@ -423,7 +423,6 @@
<ClInclude Include="AviRecorder.h" />
<ClInclude Include="AviWriter.h" />
<ClInclude Include="Base64.h" />
<ClInclude Include="blip_buf.h" />
<ClInclude Include="BpsPatcher.h" />
<ClInclude Include="CamstudioCodec.h" />
<ClInclude Include="CRC32.h" />
@ -448,6 +447,7 @@
<ClInclude Include="PlatformUtilities.h" />
<ClInclude Include="PNGHelper.h" />
<ClInclude Include="RawCodec.h" />
<ClInclude Include="HermiteResampler.h" />
<ClInclude Include="Scale2x\scale2x.h" />
<ClInclude Include="Scale2x\scale3x.h" />
<ClInclude Include="Scale2x\scalebit.h" />
@ -477,7 +477,6 @@
<ClCompile Include="ArchiveReader.cpp" />
<ClCompile Include="AviRecorder.cpp" />
<ClCompile Include="AviWriter.cpp" />
<ClCompile Include="blip_buf.cpp" />
<ClCompile Include="BpsPatcher.cpp" />
<ClCompile Include="CamstudioCodec.cpp" />
<ClCompile Include="CRC32.cpp" />
@ -575,6 +574,7 @@
<ClCompile Include="PlatformUtilities.cpp" />
<ClCompile Include="PNGHelper.cpp" />
<ClCompile Include="AutoResetEvent.cpp" />
<ClCompile Include="HermiteResampler.cpp" />
<ClCompile Include="Scale2x\scale2x.cpp">
<PrecompiledHeader Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">NotUsing</PrecompiledHeader>
<PrecompiledHeader Condition="'$(Configuration)|$(Platform)'=='PGO Profile|Win32'">NotUsing</PrecompiledHeader>

View file

@ -175,9 +175,6 @@
<ClInclude Include="Equalizer.h">
<Filter>Audio</Filter>
</ClInclude>
<ClInclude Include="blip_buf.h">
<Filter>Audio</Filter>
</ClInclude>
<ClInclude Include="Serializer.h">
<Filter>Misc</Filter>
</ClInclude>
@ -196,6 +193,9 @@
<ClInclude Include="IVideoRecorder.h">
<Filter>Avi</Filter>
</ClInclude>
<ClInclude Include="HermiteResampler.h">
<Filter>Audio</Filter>
</ClInclude>
</ItemGroup>
<ItemGroup>
<ClCompile Include="xBRZ\xbrz.cpp">
@ -279,9 +279,6 @@
<ClCompile Include="miniz.cpp">
<Filter>Archives</Filter>
</ClCompile>
<ClCompile Include="blip_buf.cpp">
<Filter>Audio</Filter>
</ClCompile>
<ClCompile Include="Equalizer.cpp">
<Filter>Audio</Filter>
</ClCompile>
@ -330,5 +327,8 @@
<ClCompile Include="GifRecorder.cpp">
<Filter>Avi</Filter>
</ClCompile>
<ClCompile Include="HermiteResampler.cpp">
<Filter>Audio</Filter>
</ClCompile>
</ItemGroup>
</Project>

View file

@ -1,345 +0,0 @@
/* blip_buf 1.1.0. http://www.slack.net/~ant/ */
#include "stdafx.h"
#include "blip_buf.h"
#include <assert.h>
#include <limits.h>
#include <string.h>
#include <stdlib.h>
/* Library Copyright (C) 2003-2009 Shay Green. This library is free software;
you can redistribute it and/or modify it under the terms of the GNU Lesser
General Public License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version. This
library is distributed in the hope that it will be useful, but WITHOUT ANY
WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
details. You should have received a copy of the GNU Lesser General Public
License along with this module; if not, write to the Free Software Foundation,
Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA */
#if defined (BLARGG_TEST) && BLARGG_TEST
#include "blargg_test.h"
#endif
/* Equivalent to ULONG_MAX >= 0xFFFFFFFF00000000.
Avoids constants that don't fit in 32 bits. */
#if ULONG_MAX/0xFFFFFFFF > 0xFFFFFFFF
typedef unsigned long fixed_t;
enum { pre_shift = 32 };
#elif defined(ULLONG_MAX)
typedef unsigned long long fixed_t;
enum { pre_shift = 32 };
#else
typedef unsigned fixed_t;
enum { pre_shift = 0 };
#endif
enum { time_bits = pre_shift + 20 };
static fixed_t const time_unit = (fixed_t) 1 << time_bits;
enum { bass_shift = 9 }; /* affects high-pass filter breakpoint frequency */
enum { end_frame_extra = 2 }; /* allows deltas slightly after frame length */
enum { half_width = 8 };
enum { buf_extra = half_width*2 + end_frame_extra };
enum { phase_bits = 5 };
enum { phase_count = 1 << phase_bits };
enum { delta_bits = 15 };
enum { delta_unit = 1 << delta_bits };
enum { frac_bits = time_bits - pre_shift };
/* We could eliminate avail and encode whole samples in offset, but that would
limit the total buffered samples to blip_max_frame. That could only be
increased by decreasing time_bits, which would reduce resample ratio accuracy.
*/
/** Sample buffer that resamples to output rate and accumulates samples
until they're read out */
struct blip_t
{
fixed_t factor;
fixed_t offset;
int avail;
int size;
int integrator;
};
typedef int buf_t;
/* probably not totally portable */
#define SAMPLES( buf ) ((buf_t*) ((buf) + 1))
/* Arithmetic (sign-preserving) right shift */
#define ARITH_SHIFT( n, shift ) \
((n) >> (shift))
enum { max_sample = +32767 };
enum { min_sample = -32768 };
#define CLAMP( n ) \
{\
if ( (short) n != n )\
n = ARITH_SHIFT( n, 16 ) ^ max_sample;\
}
static void check_assumptions( void )
{
int n;
#if INT_MAX < 0x7FFFFFFF || UINT_MAX < 0xFFFFFFFF
#error "int must be at least 32 bits"
#endif
assert( (-3 >> 1) == -2 ); /* right shift must preserve sign */
n = max_sample * 2;
CLAMP( n );
assert( n == max_sample );
n = min_sample * 2;
CLAMP( n );
assert( n == min_sample );
assert( blip_max_ratio <= time_unit );
assert( blip_max_frame <= (fixed_t) -1 >> time_bits );
}
blip_t* blip_new( int size )
{
blip_t* m;
assert( size >= 0 );
m = (blip_t*) malloc( sizeof *m + (size + buf_extra) * sizeof (buf_t) );
if ( m )
{
m->factor = time_unit / blip_max_ratio;
m->size = size;
blip_clear( m );
check_assumptions();
}
return m;
}
void blip_delete( blip_t* m )
{
if ( m != NULL )
{
/* Clear fields in case user tries to use after freeing */
memset( m, 0, sizeof *m );
free( m );
}
}
void blip_set_rates( blip_t* m, double clock_rate, double sample_rate )
{
double factor = time_unit * sample_rate / clock_rate;
m->factor = (fixed_t) factor;
/* Fails if clock_rate exceeds maximum, relative to sample_rate */
assert( 0 <= factor - m->factor && factor - m->factor < 1 );
/* Avoid requiring math.h. Equivalent to
m->factor = (int) ceil( factor ) */
if ( m->factor < factor )
m->factor++;
/* At this point, factor is most likely rounded up, but could still
have been rounded down in the floating-point calculation. */
}
void blip_clear( blip_t* m )
{
/* We could set offset to 0, factor/2, or factor-1. 0 is suitable if
factor is rounded up. factor-1 is suitable if factor is rounded down.
Since we don't know rounding direction, factor/2 accommodates either,
with the slight loss of showing an error in half the time. Since for
a 64-bit factor this is years, the halving isn't a problem. */
m->offset = m->factor / 2;
m->avail = 0;
m->integrator = 0;
memset( SAMPLES( m ), 0, (m->size + buf_extra) * sizeof (buf_t) );
}
int blip_clocks_needed( const blip_t* m, int samples )
{
fixed_t needed;
/* Fails if buffer can't hold that many more samples */
assert( samples >= 0 && m->avail + samples <= m->size );
needed = (fixed_t) samples * time_unit;
if ( needed < m->offset )
return 0;
return (int)((needed - m->offset + m->factor - 1) / m->factor);
}
void blip_end_frame( blip_t* m, unsigned t )
{
fixed_t off = t * m->factor + m->offset;
m->avail += off >> time_bits;
m->offset = off & (time_unit - 1);
/* Fails if buffer size was exceeded */
assert( m->avail <= m->size );
}
int blip_samples_avail( const blip_t* m )
{
return m->avail;
}
static void remove_samples( blip_t* m, int count )
{
buf_t* buf = SAMPLES( m );
int remain = m->avail + buf_extra - count;
m->avail -= count;
memmove( &buf [0], &buf [count], remain * sizeof buf [0] );
memset( &buf [remain], 0, count * sizeof buf [0] );
}
int blip_read_samples( blip_t* m, short out [], int count, int stereo )
{
assert( count >= 0 );
if ( count > m->avail )
count = m->avail;
if ( count )
{
int const step = stereo ? 2 : 1;
buf_t const* in = SAMPLES( m );
buf_t const* end = in + count;
int sum = m->integrator;
do
{
/* Eliminate fraction */
int s = ARITH_SHIFT( sum, delta_bits );
sum += *in++;
CLAMP( s );
*out = s;
out += step;
/* High-pass filter */
sum -= s << (delta_bits - bass_shift);
}
while ( in != end );
m->integrator = sum;
remove_samples( m, count );
}
return count;
}
/* Things that didn't help performance on x86:
__attribute__((aligned(128)))
#define short int
restrict
*/
/* Sinc_Generator( 0.9, 0.55, 4.5 ) */
static short const bl_step [phase_count + 1] [half_width] =
{
{ 43, -115, 350, -488, 1136, -914, 5861,21022},
{ 44, -118, 348, -473, 1076, -799, 5274,21001},
{ 45, -121, 344, -454, 1011, -677, 4706,20936},
{ 46, -122, 336, -431, 942, -549, 4156,20829},
{ 47, -123, 327, -404, 868, -418, 3629,20679},
{ 47, -122, 316, -375, 792, -285, 3124,20488},
{ 47, -120, 303, -344, 714, -151, 2644,20256},
{ 46, -117, 289, -310, 634, -17, 2188,19985},
{ 46, -114, 273, -275, 553, 117, 1758,19675},
{ 44, -108, 255, -237, 471, 247, 1356,19327},
{ 43, -103, 237, -199, 390, 373, 981,18944},
{ 42, -98, 218, -160, 310, 495, 633,18527},
{ 40, -91, 198, -121, 231, 611, 314,18078},
{ 38, -84, 178, -81, 153, 722, 22,17599},
{ 36, -76, 157, -43, 80, 824, -241,17092},
{ 34, -68, 135, -3, 8, 919, -476,16558},
{ 32, -61, 115, 34, -60, 1006, -683,16001},
{ 29, -52, 94, 70, -123, 1083, -862,15422},
{ 27, -44, 73, 106, -184, 1152,-1015,14824},
{ 25, -36, 53, 139, -239, 1211,-1142,14210},
{ 22, -27, 34, 170, -290, 1261,-1244,13582},
{ 20, -20, 16, 199, -335, 1301,-1322,12942},
{ 18, -12, -3, 226, -375, 1331,-1376,12293},
{ 15, -4, -19, 250, -410, 1351,-1408,11638},
{ 13, 3, -35, 272, -439, 1361,-1419,10979},
{ 11, 9, -49, 292, -464, 1362,-1410,10319},
{ 9, 16, -63, 309, -483, 1354,-1383, 9660},
{ 7, 22, -75, 322, -496, 1337,-1339, 9005},
{ 6, 26, -85, 333, -504, 1312,-1280, 8355},
{ 4, 31, -94, 341, -507, 1278,-1205, 7713},
{ 3, 35, -102, 347, -506, 1238,-1119, 7082},
{ 1, 40, -110, 350, -499, 1190,-1021, 6464},
{ 0, 43, -115, 350, -488, 1136, -914, 5861}
};
/* Shifting by pre_shift allows calculation using unsigned int rather than
possibly-wider fixed_t. On 32-bit platforms, this is likely more efficient.
And by having pre_shift 32, a 32-bit platform can easily do the shift by
simply ignoring the low half. */
void blip_add_delta( blip_t* m, unsigned time, int delta )
{
unsigned fixed = (unsigned) ((time * m->factor + m->offset) >> pre_shift);
buf_t* out = SAMPLES( m ) + m->avail + (fixed >> frac_bits);
int const phase_shift = frac_bits - phase_bits;
int phase = fixed >> phase_shift & (phase_count - 1);
short const* in = bl_step [phase];
short const* rev = bl_step [phase_count - phase];
int interp = fixed >> (phase_shift - delta_bits) & (delta_unit - 1);
int delta2 = (delta * interp) >> delta_bits;
delta -= delta2;
/* Fails if buffer size was exceeded */
assert( out <= &SAMPLES( m ) [m->size + end_frame_extra] );
out [0] += in[0]*delta + in[half_width+0]*delta2;
out [1] += in[1]*delta + in[half_width+1]*delta2;
out [2] += in[2]*delta + in[half_width+2]*delta2;
out [3] += in[3]*delta + in[half_width+3]*delta2;
out [4] += in[4]*delta + in[half_width+4]*delta2;
out [5] += in[5]*delta + in[half_width+5]*delta2;
out [6] += in[6]*delta + in[half_width+6]*delta2;
out [7] += in[7]*delta + in[half_width+7]*delta2;
in = rev;
out [ 8] += in[7]*delta + in[7-half_width]*delta2;
out [ 9] += in[6]*delta + in[6-half_width]*delta2;
out [10] += in[5]*delta + in[5-half_width]*delta2;
out [11] += in[4]*delta + in[4-half_width]*delta2;
out [12] += in[3]*delta + in[3-half_width]*delta2;
out [13] += in[2]*delta + in[2-half_width]*delta2;
out [14] += in[1]*delta + in[1-half_width]*delta2;
out [15] += in[0]*delta + in[0-half_width]*delta2;
}
void blip_add_delta_fast( blip_t* m, unsigned time, int delta )
{
unsigned fixed = (unsigned) ((time * m->factor + m->offset) >> pre_shift);
buf_t* out = SAMPLES( m ) + m->avail + (fixed >> frac_bits);
int interp = fixed >> (frac_bits - delta_bits) & (delta_unit - 1);
int delta2 = delta * interp;
/* Fails if buffer size was exceeded */
assert( out <= &SAMPLES( m ) [m->size + end_frame_extra] );
out [7] += delta * delta_unit - delta2;
out [8] += delta2;
}

View file

@ -1,81 +0,0 @@
#pragma once
#include "stdafx.h"
/** \file
Sample buffer that resamples from input clock rate to output sample rate */
/* blip_buf 1.1.0 */
#ifndef BLIP_BUF_H
#define BLIP_BUF_H
#if defined(_MSC_VER)
#define EXPORT __declspec(dllexport)
#else
#define EXPORT
#endif
#ifdef __cplusplus
extern "C" {
#endif
/** First parameter of most functions is blip_t*, or const blip_t* if nothing
is changed. */
typedef struct blip_t blip_t;
/** Creates new buffer that can hold at most sample_count samples. Sets rates
so that there are blip_max_ratio clocks per sample. Returns pointer to new
buffer, or NULL if insufficient memory. */
EXPORT blip_t* blip_new( int sample_count );
/** Sets approximate input clock rate and output sample rate. For every
clock_rate input clocks, approximately sample_rate samples are generated. */
EXPORT void blip_set_rates( blip_t*, double clock_rate, double sample_rate );
enum { /** Maximum clock_rate/sample_rate ratio. For a given sample_rate,
clock_rate must not be greater than sample_rate*blip_max_ratio. */
blip_max_ratio = 1 << 20 };
/** Clears entire buffer. Afterwards, blip_samples_avail() == 0. */
EXPORT void blip_clear( blip_t* );
/** Adds positive/negative delta into buffer at specified clock time. */
EXPORT void blip_add_delta( blip_t*, unsigned int clock_time, int delta );
/** Same as blip_add_delta(), but uses faster, lower-quality synthesis. */
void blip_add_delta_fast( blip_t*, unsigned int clock_time, int delta );
/** Length of time frame, in clocks, needed to make sample_count additional
samples available. */
int blip_clocks_needed( const blip_t*, int sample_count );
enum { /** Maximum number of samples that can be generated from one time frame. */
blip_max_frame = 4000 };
/** Makes input clocks before clock_duration available for reading as output
samples. Also begins new time frame at clock_duration, so that clock time 0 in
the new time frame specifies the same clock as clock_duration in the old time
frame specified. Deltas can have been added slightly past clock_duration (up to
however many clocks there are in two output samples). */
EXPORT void blip_end_frame( blip_t*, unsigned int clock_duration );
/** Number of buffered samples available for reading. */
int blip_samples_avail( const blip_t* );
/** Reads and removes at most 'count' samples and writes them to 'out'. If
'stereo' is true, writes output to every other element of 'out', allowing easy
interleaving of two buffers into a stereo sample stream. Outputs 16-bit signed
samples. Returns number of samples actually read. */
EXPORT int blip_read_samples( blip_t*, short out [], int count, int stereo );
/** Frees buffer. No effect if NULL is passed. */
EXPORT void blip_delete( blip_t* );
/* Deprecated */
typedef blip_t blip_buffer_t;
#ifdef __cplusplus
}
#endif
#endif

View file

@ -248,6 +248,9 @@
#define __forceinline
#define alloca __builtin_alloca
#elif !defined(_MSC_VER)
#ifdef __forceinline
#undef __forceinline
#endif
#if __GNUC__
#define __forceinline inline
#else

View file

@ -21,4 +21,14 @@ using std::istream;
using std::string;
using std::vector;
using std::atomic;
using std::atomic_flag;
using std::atomic_flag;
#ifndef __MINGW32__
#ifdef __clang__
#define __forceinline __attribute__((always_inline)) inline
#else
#ifdef __GNUC__
#define __forceinline __attribute__((always_inline)) inline
#endif
#endif
#endif