mirror of
https://github.com/trudnai/Steve2.git
synced 2025-04-14 20:37:33 +00:00
Some improvements, probably cannot even hear it
This commit is contained in:
tudnai
parent
60b0c3917a
commit
1aa9270124
@ -91,6 +91,7 @@ ALCcontext *ctx = NULL;
|
||||
int spkr_att = 0;
|
||||
int spkr_level = SPKR_LEVEL_ZERO;
|
||||
int spkr_level_ema = SPKR_LEVEL_ZERO;
|
||||
int spkr_level_ema1 = SPKR_LEVEL_ZERO;
|
||||
int spkr_level_ema2 = SPKR_LEVEL_ZERO;
|
||||
int spkr_level_ema3 = SPKR_LEVEL_ZERO;
|
||||
int spkr_level_ema4 = SPKR_LEVEL_ZERO;
|
||||
@ -128,7 +129,7 @@ unsigned spkr_fps = DEFAULT_FPS;
|
||||
unsigned spkr_frame_cntr = 0;
|
||||
unsigned spkr_clk = 0;
|
||||
|
||||
#define SPKR_BUF_SIZE ( spkr_sample_rate * SPKR_CHANNELS / DEFAULT_FPS ) // stereo
|
||||
#define SPKR_BUF_SIZE ( (int)(spkr_sample_rate * SPKR_CHANNELS / DEFAULT_FPS) ) // stereo
|
||||
#define SPKR_BUF_SLOT(n) ( SPKR_BUF_SIZE * (n) )
|
||||
#define SPKR_BUF_SLOT_SIZE(n) ( SPKR_BUF_SLOT(n) * sizeof(spkr_sample_t) )
|
||||
|
||||
@ -144,7 +145,7 @@ const unsigned spkr_seconds = 1;
|
||||
|
||||
#ifdef SPKR_OVERSAMPLING
|
||||
//const unsigned spkr_stream_rate = SPKR_STREAM_RATE; // Optimal? 220500; // Best: 321000; // Acceptable: 192000; // higher the sample rate, the better the sound gets
|
||||
const unsigned spkr_sample_rate = SPKR_STREAM_RATE * SPKR_OVERSAMPLING;
|
||||
const double spkr_sample_rate = MHZ(DEFAULT_MHZ_6502); // SPKR_STREAM_RATE * SPKR_OVERSAMPLING;
|
||||
#else
|
||||
const unsigned spkr_sample_rate = 321000; // Optimal? 220500; // Best: 321000; // Acceptable: 192000; // higher the sample rate, the better
|
||||
#endif
|
||||
@ -559,7 +560,7 @@ float SPKR_INITIAL_TRAILING_EDGE = 0.64; // need a bit of slope to get Xonix so
|
||||
void spkr_toggle() {
|
||||
// do not sleep while sound is playing
|
||||
m6502.ecoSpindown = ecoSpindown;
|
||||
|
||||
|
||||
if ( diskAccelerator_count ) {
|
||||
// turn off disk acceleration immediately
|
||||
diskAccelerator_count = 0;
|
||||
@ -570,16 +571,15 @@ void spkr_toggle() {
|
||||
|
||||
// push a click into the speaker buffer
|
||||
// (we will play the entire buffer at the end of the frame)
|
||||
double multiplier = (double)spkr_sample_rate / MHZ(MHz_6502);
|
||||
#ifdef SPKR_KEEP_PITCH
|
||||
if ( MHz_6502 < default_MHz_6502 ) {
|
||||
multiplier = (double)spkr_sample_rate / MHZ(default_MHz_6502);
|
||||
multiplier = spkr_sample_rate / MHZ(default_MHz_6502);
|
||||
// indexer = (double)spkr_sample_rate / MHZ(default_MHz_6502) * MHz_6502;
|
||||
}
|
||||
#endif
|
||||
|
||||
// spkr_sample_idx = round( (spkr_clk + m6502.clkfrm) / ( MHZ(default_MHz_6502) / (double)spkr_sample_rate)) * SPKR_CHANNELS;
|
||||
spkr_sample_idx = round( (double)(spkr_clk + m6502.clkfrm) * multiplier ) * SPKR_CHANNELS;
|
||||
spkr_sample_idx = round( (double)(spkr_clk + m6502.clkfrm) ) * SPKR_CHANNELS;
|
||||
|
||||
spkr_sample_idx %= SPKR_BUF_SLOT_SIZE(BUFFER_COUNT);
|
||||
|
||||
@ -613,53 +613,76 @@ void spkr_toggle() {
|
||||
int playDelay = SPKR_PLAY_DELAY;
|
||||
|
||||
|
||||
#define SPKR_FILTER_RESET
|
||||
#undef SPKR_FILTER_RESET
|
||||
|
||||
#ifdef SPKR_FILTER_RESET
|
||||
INLINE static void spkr_filter_reset() {
|
||||
spkr_level = SPKR_LEVEL_ZERO;
|
||||
spkr_level_ema = SPKR_LEVEL_ZERO;
|
||||
spkr_level_ema1 = SPKR_LEVEL_ZERO;
|
||||
spkr_level_ema2 = SPKR_LEVEL_ZERO;
|
||||
spkr_level_ema3 = SPKR_LEVEL_ZERO;
|
||||
}
|
||||
#endif
|
||||
|
||||
// Exponential Moving Average
|
||||
INLINE int ema( int val, int prev, float ema_len ) {
|
||||
static const float ema_sensitivity = 2;
|
||||
// static const float ema_len = 42;
|
||||
float m = ema_sensitivity / ema_len;
|
||||
float n = 1 - m;
|
||||
|
||||
/// Exponential Moving Average - general implementation
|
||||
/// @note Calculate the multiplier for smoothing (weighting) the EMA,
|
||||
/// which typically follows the formula: [2 ÷ (number of observations + 1)].
|
||||
/// @param val Current value
|
||||
/// @param prev Previous value
|
||||
/// @param len Length of the period
|
||||
/// @param sensitivity Sensitivity (1: Wilders, 2: Exponential)
|
||||
INLINE int exp_ma( const int val, const int prev, const float len, const float sensitivity ) {
|
||||
const float m = sensitivity / len;
|
||||
const float n = 1 - m;
|
||||
return m * val + n * prev;
|
||||
}
|
||||
|
||||
|
||||
/// calculate the multiplier for smoothing (weighting) the EMA,
|
||||
/// which typically follows the formula: [2 ÷ (number of observations + 1)].
|
||||
///
|
||||
/// IMPORTANT!: ema1 must be precalculated!
|
||||
/// Exponential Moving Average
|
||||
/// @param val Current value
|
||||
/// @param prev Previous value
|
||||
/// @param len Length of the period
|
||||
INLINE int ema( const int val, const int prev, const float len ) {
|
||||
return exp_ma(val, prev, len, 2);
|
||||
}
|
||||
|
||||
|
||||
/// Wilders Moving Average
|
||||
/// @param val Current value
|
||||
/// @param prev Previous value
|
||||
/// @param len Length of the period
|
||||
INLINE int wma( const int val, const int prev, const float len ) {
|
||||
return exp_ma(val, prev, len, 1);
|
||||
}
|
||||
|
||||
|
||||
/// Double EMA
|
||||
/// @note ema1 must be precalculated!
|
||||
/// * You Must Call ema() before this to get ema1
|
||||
/// * ema2 is updated!
|
||||
INLINE static const double dema ( const double ema1, double ema2 ) {
|
||||
/// @param ema1 EMA (must be precalculated)
|
||||
/// @param ema2 EMA of EMA (must be precalculated)
|
||||
INLINE static const double dema ( const double ema1, const double ema2 ) {
|
||||
return 2 * ema1 - ema2;
|
||||
}
|
||||
|
||||
|
||||
/// Calculate the multiplier for smoothing (weighting) the EMA,
|
||||
/// which typically follows the formula: [2 ÷ (number of observations + 1)].
|
||||
///
|
||||
/// IMPORTANT!: ema1 and ema2 must be precalculated!
|
||||
/// Tripple EMA
|
||||
/// @note ema1 and ema2 must be precalculated!
|
||||
/// * You Must Call dema() before this to get ema2 and ema() to get ema1
|
||||
/// * ema3 is updated!
|
||||
INLINE static const double tema ( const double ema1, const double ema2, double ema3 ) {
|
||||
/// @param ema1 EMA (must be precalculated)
|
||||
/// @param ema2 EMA of EMA (must be precalculated)
|
||||
/// @param ema3 EMA of EMA of EMA (must be precalculated)
|
||||
INLINE static const double tema ( const double ema1, const double ema2, const double ema3 ) {
|
||||
return 3 * ema1 - 3 * ema2 + ema3;
|
||||
}
|
||||
|
||||
|
||||
/// T3 Moving Average
|
||||
///
|
||||
/// Tillson's T3 is a kind of Moving average. Tim Tillson described it in "Technical Analysis of Stocks and Commodities", January 1998.
|
||||
/// @note Tillson's T3 is a kind of Moving average. Tim Tillson described it in "Technical Analysis of Stocks and Commodities", January 1998.
|
||||
/// He named his article "Better Moving Averages". Tillson’s moving average becomes a popular indicator of technical analysis.
|
||||
/// Its advantage is that it gets less lag with the price chart and its curve is considerably smoother. By using it, the trader can get
|
||||
/// early entry and less number of false signals.
|
||||
@ -679,6 +702,11 @@ INLINE static const double tema ( const double ema1, const double ema2, double e
|
||||
///
|
||||
/// IMPORTANT!: ema1, ema2 ema3, ema4, ame5 and ema6 must be precalculated!
|
||||
/// * You Must Call dema() before this to get ema2 and ema() to get ema1
|
||||
/// @param ema3 ema3 = ema (ema2)
|
||||
/// @param ema4 ema4 = ema (ema3)
|
||||
/// @param ema5 ema5 = ema (ema4)
|
||||
/// @param ema6 ema6 = ema (ema5)
|
||||
/// @param a Volume factor (typical: 0.7)
|
||||
INLINE static const double t3 (
|
||||
const double ema3,
|
||||
const double ema4,
|
||||
@ -701,7 +729,56 @@ INLINE static const double t3 (
|
||||
}
|
||||
|
||||
|
||||
INLINE static void spkr_filter_ema(spkr_sample_t * buf, int buf_size) {
|
||||
/// Generalized Dema - GD (n,v) = EMA (n)*(1+v)-EMA( EMA (n))*v,
|
||||
/// @note Where v ranges between 0 and 1. When v=0, GD is just an EMA , and when v=1, GD is DEMA . In between, GD is a cooler DEMA . By using a value for v less than 1 (I like .7), we cure the multiple DEMA overshoot problem, at the cost of accepting some additional phase delay. Now we can run GD through itself multiple times to define a new, smoother moving average T3 that does not overshoot the data:
|
||||
///
|
||||
/// @param val Value to filter
|
||||
/// @param ema1 Pointer to EMA - will be updated
|
||||
/// @param ema2 Pointer to EMA of EMA - will be updated
|
||||
/// @param len EMA length of period
|
||||
/// @param v Volume Factor
|
||||
INLINE static const double gd ( const double val, double * ema1, double * ema2, const double len, const double v ) {
|
||||
*ema1 = ema(val, *ema1, len);
|
||||
*ema2 = ema(*ema1, *ema2, len);
|
||||
return *ema1 * (1 + v) - *ema2 * v;
|
||||
}
|
||||
|
||||
|
||||
/// T3 Moving Average - T3(n) = GD ( GD ( GD (n)))
|
||||
///
|
||||
/// @note Tillson's T3 is a kind of Moving average. Tim Tillson described it in "Technical Analysis of Stocks and Commodities", January 1998.
|
||||
/// He named his article "Better Moving Averages". Tillson’s moving average becomes a popular indicator of technical analysis.
|
||||
/// Its advantage is that it gets less lag with the price chart and its curve is considerably smoother. By using it, the trader can get
|
||||
/// early entry and less number of false signals.
|
||||
///
|
||||
/// In filter theory parlance, T3 is a six-pole non-linear Kalman filter. Kalman filters are ones which use the error (in this case (time series - EMA (n)) to correct themselves. In Technical Analysis , these are called Adaptive Moving Averages; they track the time series more aggressively when it is making large moves
|
||||
///
|
||||
/// @param val Value to filter
|
||||
/// @param ema1 Pointer to first EMA - will be updated
|
||||
/// @param ema2 Pointer to first EMA of EMA - will be updated
|
||||
/// @param ema3 Pointer to second EMA - will be updated
|
||||
/// @param ema4 Pointer to second EMA of EMA - will be updated
|
||||
/// @param ema5 Pointer to third EMA - will be updated
|
||||
/// @param ema6 Pointer to third EMA of EMA - will be updated
|
||||
/// @param len EMA length of period
|
||||
INLINE static const double t3_new(
|
||||
const double val,
|
||||
double * ema1,
|
||||
double * ema2,
|
||||
double * ema3,
|
||||
double * ema4,
|
||||
double * ema5,
|
||||
double * ema6,
|
||||
const double len
|
||||
) {
|
||||
double gd1 = gd(val, ema1, ema2, len, 0.7);
|
||||
double gd2 = gd(gd1, ema3, ema4, len, 0.7);
|
||||
double gd3 = gd(gd2, ema5, ema6, len, 0.7);
|
||||
return gd3;
|
||||
}
|
||||
|
||||
|
||||
INLINE static void spkr_filter_ema(spkr_sample_t * buf, const int buf_size) {
|
||||
for ( int i = 0; i < buf_size; ) {
|
||||
spkr_level_ema = ema(buf[i], spkr_level_ema, spkr_ema_len);
|
||||
|
||||
@ -715,10 +792,10 @@ INLINE static void spkr_filter_ema(spkr_sample_t * buf, int buf_size) {
|
||||
}
|
||||
|
||||
|
||||
INLINE static void spkr_filter_ema3(spkr_sample_t * buf, int buf_size) {
|
||||
INLINE static void spkr_filter_ema3(spkr_sample_t * buf, const int buf_size) {
|
||||
for ( int i = 0; i < buf_size; ) {
|
||||
spkr_level_ema = ema(buf[i], spkr_level_ema, spkr_ema3_len);
|
||||
spkr_level_ema2 = ema(spkr_level_ema, spkr_level_ema2, spkr_ema3_len);
|
||||
spkr_level_ema1 = ema(buf[i], spkr_level_ema1, spkr_ema3_len);
|
||||
spkr_level_ema2 = ema(spkr_level_ema1, spkr_level_ema2, spkr_ema3_len);
|
||||
spkr_level_ema3 = ema(spkr_level_ema2, spkr_level_ema3, spkr_ema3_len);
|
||||
|
||||
// smoothing with EMA3
|
||||
@ -731,12 +808,12 @@ INLINE static void spkr_filter_ema3(spkr_sample_t * buf, int buf_size) {
|
||||
}
|
||||
|
||||
|
||||
INLINE static void spkr_filter_dema(spkr_sample_t * buf, int buf_size) {
|
||||
INLINE static void spkr_filter_dema(spkr_sample_t * buf, const int buf_size) {
|
||||
for ( int i = 0; i < buf_size; ) {
|
||||
spkr_level_ema = ema(buf[i], spkr_level_ema, spkr_ema_len);
|
||||
spkr_level_ema2 = ema(spkr_level_ema, spkr_level_ema2, spkr_ema_len);
|
||||
spkr_level_ema1 = ema(buf[i], spkr_level_ema1, spkr_ema_len);
|
||||
spkr_level_ema2 = ema(spkr_level_ema1, spkr_level_ema2, spkr_ema_len);
|
||||
|
||||
double level = dema(spkr_level_ema, spkr_level_ema2);
|
||||
double level = dema(spkr_level_ema1, spkr_level_ema2);
|
||||
// smoothing with DEMA
|
||||
buf[i++] = level;
|
||||
buf[i++] = level;
|
||||
@ -749,13 +826,13 @@ INLINE static void spkr_filter_dema(spkr_sample_t * buf, int buf_size) {
|
||||
}
|
||||
|
||||
|
||||
INLINE static void spkr_filter_tema(spkr_sample_t * buf, int buf_size) {
|
||||
INLINE static void spkr_filter_tema(spkr_sample_t * buf, const int buf_size) {
|
||||
for ( int i = 0; i < buf_size; ) {
|
||||
spkr_level_ema = ema(buf[i], spkr_level_ema, spkr_ema_len);
|
||||
spkr_level_ema2 = ema(spkr_level_ema, spkr_level_ema2, spkr_ema_len);
|
||||
spkr_level_ema1 = ema(buf[i], spkr_level_ema1, spkr_ema_len);
|
||||
spkr_level_ema2 = ema(spkr_level_ema1, spkr_level_ema2, spkr_ema_len);
|
||||
spkr_level_ema3 = ema(spkr_level_ema2, spkr_level_ema3, spkr_ema_len);
|
||||
|
||||
double level = tema(spkr_level_ema, spkr_level_ema2, spkr_level_ema3);
|
||||
double level = tema(spkr_level_ema1, spkr_level_ema2, spkr_level_ema3);
|
||||
// smoothing with TEMA
|
||||
buf[i++] = level;
|
||||
buf[i++] = level;
|
||||
@ -768,10 +845,10 @@ INLINE static void spkr_filter_tema(spkr_sample_t * buf, int buf_size) {
|
||||
}
|
||||
|
||||
|
||||
INLINE static void spkr_filter_t3(spkr_sample_t * buf, int buf_size) {
|
||||
INLINE static void spkr_filter_t3(spkr_sample_t * buf, const int buf_size) {
|
||||
for ( int i = 0; i < buf_size; ) {
|
||||
spkr_level_ema = ema(buf[i], spkr_level_ema, spkr_ema_len);
|
||||
spkr_level_ema2 = ema(spkr_level_ema, spkr_level_ema2, spkr_ema_len);
|
||||
spkr_level_ema1 = ema(buf[i], spkr_level_ema1, spkr_ema_len);
|
||||
spkr_level_ema2 = ema(spkr_level_ema1, spkr_level_ema2, spkr_ema_len);
|
||||
spkr_level_ema3 = ema(spkr_level_ema2, spkr_level_ema3, spkr_ema_len);
|
||||
spkr_level_ema4 = ema(spkr_level_ema3, spkr_level_ema4, spkr_ema_len);
|
||||
spkr_level_ema5 = ema(spkr_level_ema4, spkr_level_ema5, spkr_ema_len);
|
||||
@ -858,7 +935,7 @@ INLINE static spkr_sample_t spkr_avg_new(const spkr_sample_t * buf, int len) {
|
||||
|
||||
|
||||
INLINE static void spkr_downsample() {
|
||||
for (int i = 0; i < SPKR_STRM_SLOT_SIZE(1); ) {
|
||||
for (int i = 0; i < SPKR_BUF_SIZE; ) {
|
||||
int buf_idx = i * SPKR_OVERSAMPLING;
|
||||
|
||||
// get the average for that section
|
||||
@ -884,6 +961,7 @@ INLINE static void spkr_filter() {
|
||||
// spkr_filter_tema( spkr_samples, SPKR_BUF_SIZE );
|
||||
// spkr_filter_t3( spkr_samples, SPKR_BUF_SIZE );
|
||||
spkr_downsample();
|
||||
// spkr_filter_ema( spkr_stream, SPKR_STRM_SLOT_SIZE(1) );
|
||||
#else
|
||||
spkr_filter_ema( spkr_samples, SPKR_BUF_SIZE );
|
||||
// spkr_filter_dema( spkr_samples, SPKR_BUF_SIZE );
|
||||
@ -1105,10 +1183,7 @@ void spkr_update() {
|
||||
spkr_fade(spkr_level, 0);
|
||||
spkr_filter();
|
||||
|
||||
#ifdef SPKR_DEBUG
|
||||
spkr_debug(spkr_debug_raw_file);
|
||||
spkr_debug(spkr_debug_ema_file);
|
||||
#endif
|
||||
|
||||
spkr_buffer_with_prebuf();
|
||||
|
||||
@ -1145,7 +1220,13 @@ void spkr_update() {
|
||||
|
||||
spkr_sample_idx = 0;
|
||||
spkr_sample_last_idx = 0;
|
||||
|
||||
|
||||
// spkr_sample_idx -= SPKR_BUF_SLOT_SIZE(1);
|
||||
// if ( spkr_sample_idx < 0 ) spkr_sample_idx = 0;
|
||||
// spkr_sample_last_idx -= SPKR_BUF_SLOT_SIZE(1);
|
||||
// if ( spkr_sample_last_idx < 0 ) spkr_sample_last_idx = 0;
|
||||
|
||||
|
||||
// make sure it never goes below 0 (never overflows)
|
||||
if ( (int)spkr_play_time < 0 ) {
|
||||
spkr_play_time = 0;
|
||||
|
||||
@ -80,7 +80,7 @@ enum {
|
||||
|
||||
typedef int16_t spkr_sample_t;
|
||||
|
||||
extern const unsigned spkr_sample_rate;
|
||||
extern const double spkr_sample_rate;
|
||||
extern const unsigned spkr_buf_alloc_size;
|
||||
extern const unsigned spkr_buf_size;
|
||||
extern unsigned spkr_fps;
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user