344 lines
10 KiB
Java
344 lines
10 KiB
Java
|
|
||
|
package ibxm;
|
||
|
|
||
|
public class IBXM {
|
||
|
public static final String VERSION = "ibxm alpha 51 (c)2008 mumart@gmail.com";
|
||
|
|
||
|
public static final int FP_SHIFT = 15;
|
||
|
public static final int FP_ONE = 1 << FP_SHIFT;
|
||
|
public static final int FP_MASK = FP_ONE - 1;
|
||
|
|
||
|
private int sampling_rate, resampling_quality, volume_ramp_length;
|
||
|
private int tick_length_samples, current_tick_samples;
|
||
|
private int[] mixing_buffer, volume_ramp_buffer;
|
||
|
|
||
|
private Module module;
|
||
|
private Channel[] channels;
|
||
|
private int[] global_volume, note;
|
||
|
private int current_sequence_index, next_sequence_index;
|
||
|
private int current_row, next_row;
|
||
|
private int tick_counter, ticks_per_row;
|
||
|
private int pattern_loop_count, pattern_loop_channel;
|
||
|
|
||
|
public IBXM( int sample_rate ) {
|
||
|
|
||
|
/** MODIFIED 13 Oct 2009 by Paul Lamb **/
|
||
|
// System.out.println( VERSION );
|
||
|
/***************************************/
|
||
|
|
||
|
if( sample_rate < 8000 ) {
|
||
|
sample_rate = 8000;
|
||
|
}
|
||
|
sampling_rate = sample_rate;
|
||
|
volume_ramp_length = sampling_rate >> 10;
|
||
|
volume_ramp_buffer = new int[ volume_ramp_length * 2 ];
|
||
|
mixing_buffer = new int[ sampling_rate / 6 ];
|
||
|
global_volume = new int[ 1 ];
|
||
|
note = new int[ 5 ];
|
||
|
set_module( new Module() );
|
||
|
set_resampling_quality( 1 );
|
||
|
}
|
||
|
|
||
|
public void set_module( Module m ) {
|
||
|
int channel_idx;
|
||
|
module = m;
|
||
|
channels = new Channel[ module.get_num_channels() ];
|
||
|
for( channel_idx = 0; channel_idx < channels.length; channel_idx++ ) {
|
||
|
channels[ channel_idx ] = new Channel( module, sampling_rate, global_volume );
|
||
|
}
|
||
|
set_sequence_index( 0, 0 );
|
||
|
}
|
||
|
|
||
|
public void set_resampling_quality( int quality ) {
|
||
|
resampling_quality = quality;
|
||
|
}
|
||
|
|
||
|
public int calculate_song_duration() {
|
||
|
int song_duration;
|
||
|
set_sequence_index( 0, 0 );
|
||
|
next_tick();
|
||
|
song_duration = tick_length_samples;
|
||
|
while( !next_tick() ) {
|
||
|
song_duration += tick_length_samples;
|
||
|
}
|
||
|
set_sequence_index( 0, 0 );
|
||
|
return song_duration;
|
||
|
}
|
||
|
|
||
|
public void set_sequence_index( int sequence_index, int row ) {
|
||
|
int channel_idx;
|
||
|
global_volume[ 0 ] = 64;
|
||
|
for( channel_idx = 0; channel_idx < channels.length; channel_idx++ ) {
|
||
|
channels[ channel_idx ].reset();
|
||
|
channels[ channel_idx ].set_panning( module.get_initial_panning( channel_idx ) );
|
||
|
}
|
||
|
set_global_volume( module.global_volume );
|
||
|
set_speed( 6 );
|
||
|
set_speed( module.default_speed );
|
||
|
set_tempo( 125 );
|
||
|
set_tempo( module.default_tempo );
|
||
|
pattern_loop_count = -1;
|
||
|
next_sequence_index = sequence_index;
|
||
|
next_row = row;
|
||
|
tick_counter = 0;
|
||
|
current_tick_samples = tick_length_samples;
|
||
|
clear_vol_ramp_buffer();
|
||
|
}
|
||
|
|
||
|
public void seek( int sample_position ) {
|
||
|
int idx;
|
||
|
set_sequence_index( 0, 0 );
|
||
|
next_tick();
|
||
|
while( sample_position > tick_length_samples ) {
|
||
|
sample_position -= tick_length_samples;
|
||
|
next_tick();
|
||
|
}
|
||
|
mix_tick();
|
||
|
current_tick_samples = sample_position;
|
||
|
}
|
||
|
|
||
|
public void get_audio( byte[] output_buffer, int frames ) {
|
||
|
int output_idx, mix_idx, mix_end, count, amplitude;
|
||
|
output_idx = 0;
|
||
|
while( frames > 0 ) {
|
||
|
count = tick_length_samples - current_tick_samples;
|
||
|
if( count > frames ) {
|
||
|
count = frames;
|
||
|
}
|
||
|
mix_idx = current_tick_samples << 1;
|
||
|
mix_end = mix_idx + ( count << 1 ) - 1;
|
||
|
while( mix_idx <= mix_end ) {
|
||
|
amplitude = mixing_buffer[ mix_idx ];
|
||
|
if( amplitude > 32767 ) {
|
||
|
amplitude = 32767;
|
||
|
}
|
||
|
if( amplitude < -32768 ) {
|
||
|
amplitude = -32768;
|
||
|
}
|
||
|
output_buffer[ output_idx ] = ( byte ) ( amplitude >> 8 );
|
||
|
output_buffer[ output_idx + 1 ] = ( byte ) ( amplitude & 0xFF );
|
||
|
output_idx += 2;
|
||
|
mix_idx += 1;
|
||
|
}
|
||
|
current_tick_samples = mix_idx >> 1;
|
||
|
frames -= count;
|
||
|
if( frames > 0 ) {
|
||
|
next_tick();
|
||
|
mix_tick();
|
||
|
current_tick_samples = 0;
|
||
|
}
|
||
|
}
|
||
|
}
|
||
|
|
||
|
private void mix_tick() {
|
||
|
int channel_idx, mix_idx, mix_len;
|
||
|
mix_idx = 0;
|
||
|
mix_len = tick_length_samples + volume_ramp_length << 1;
|
||
|
while( mix_idx < mix_len ) {
|
||
|
mixing_buffer[ mix_idx ] = 0;
|
||
|
mix_idx += 1;
|
||
|
}
|
||
|
for( channel_idx = 0; channel_idx < channels.length; channel_idx++ ) {
|
||
|
mix_len = tick_length_samples + volume_ramp_length;
|
||
|
channels[ channel_idx ].resample( mixing_buffer, 0, mix_len, resampling_quality );
|
||
|
}
|
||
|
volume_ramp();
|
||
|
}
|
||
|
|
||
|
private boolean next_tick() {
|
||
|
int channel_idx;
|
||
|
boolean song_end;
|
||
|
for( channel_idx = 0; channel_idx < channels.length; channel_idx++ ) {
|
||
|
channels[ channel_idx ].update_sample_idx( tick_length_samples );
|
||
|
}
|
||
|
tick_counter -= 1;
|
||
|
if( tick_counter <= 0 ) {
|
||
|
tick_counter = ticks_per_row;
|
||
|
song_end = next_row();
|
||
|
} else {
|
||
|
for( channel_idx = 0; channel_idx < channels.length; channel_idx++ ) {
|
||
|
channels[ channel_idx ].tick();
|
||
|
}
|
||
|
song_end = false;
|
||
|
}
|
||
|
return song_end;
|
||
|
}
|
||
|
|
||
|
private boolean next_row() {
|
||
|
int channel_idx, effect, effect_param;
|
||
|
boolean song_end;
|
||
|
Pattern pattern;
|
||
|
song_end = false;
|
||
|
if( next_sequence_index < 0 ) {
|
||
|
/* Bad next sequence index.*/
|
||
|
next_sequence_index = 0;
|
||
|
next_row = 0;
|
||
|
}
|
||
|
if( next_sequence_index >= module.get_sequence_length() ) {
|
||
|
/* End of sequence.*/
|
||
|
song_end = true;
|
||
|
next_sequence_index = module.restart_sequence_index;
|
||
|
if( next_sequence_index < 0 ) {
|
||
|
next_sequence_index = 0;
|
||
|
}
|
||
|
if( next_sequence_index >= module.get_sequence_length() ) {
|
||
|
next_sequence_index = 0;
|
||
|
}
|
||
|
next_row = 0;
|
||
|
}
|
||
|
if( next_sequence_index < current_sequence_index ) {
|
||
|
/* Jump to previous pattern. */
|
||
|
song_end = true;
|
||
|
}
|
||
|
if( next_sequence_index == current_sequence_index ) {
|
||
|
if( next_row <= current_row ) {
|
||
|
if( pattern_loop_count < 0 ) {
|
||
|
/* Jump to previous row in the same pattern, but not a pattern loop. */
|
||
|
song_end = true;
|
||
|
}
|
||
|
}
|
||
|
}
|
||
|
current_sequence_index = next_sequence_index;
|
||
|
pattern = module.get_pattern_from_sequence( current_sequence_index );
|
||
|
if( next_row < 0 || next_row >= pattern.num_rows ) {
|
||
|
/* Bad next row.*/
|
||
|
next_row = 0;
|
||
|
}
|
||
|
current_row = next_row;
|
||
|
next_row = current_row + 1;
|
||
|
if( next_row >= pattern.num_rows ) {
|
||
|
next_sequence_index = current_sequence_index + 1;
|
||
|
next_row = 0;
|
||
|
}
|
||
|
for( channel_idx = 0; channel_idx < channels.length; channel_idx++ ) {
|
||
|
pattern.get_note( note, current_row * channels.length + channel_idx );
|
||
|
effect = note[ 3 ];
|
||
|
effect_param = note[ 4 ];
|
||
|
channels[ channel_idx ].row( note[ 0 ], note[ 1 ], note[ 2 ], effect, effect_param );
|
||
|
switch( effect ) {
|
||
|
case 0x0B:
|
||
|
/* Pattern Jump.*/
|
||
|
if( pattern_loop_count < 0 ) {
|
||
|
next_sequence_index = effect_param;
|
||
|
next_row = 0;
|
||
|
}
|
||
|
break;
|
||
|
case 0x0D:
|
||
|
/* Pattern Break.*/
|
||
|
if( pattern_loop_count < 0 ) {
|
||
|
next_sequence_index = current_sequence_index + 1;
|
||
|
next_row = ( effect_param >> 4 ) * 10 + ( effect_param & 0x0F );
|
||
|
}
|
||
|
break;
|
||
|
case 0x0E:
|
||
|
/* Extended.*/
|
||
|
switch( effect_param & 0xF0 ) {
|
||
|
case 0x60:
|
||
|
/* Pattern loop.*/
|
||
|
if( ( effect_param & 0x0F ) == 0 ) {
|
||
|
/* Set loop marker on this channel. */
|
||
|
channels[ channel_idx ].pattern_loop_row = current_row;
|
||
|
}
|
||
|
if( channels[ channel_idx ].pattern_loop_row < current_row ) {
|
||
|
/* Marker and parameter are valid. Begin looping. */
|
||
|
if( pattern_loop_count < 0 ) {
|
||
|
/* Not already looping, begin. */
|
||
|
pattern_loop_count = effect_param & 0x0F;
|
||
|
pattern_loop_channel = channel_idx;
|
||
|
}
|
||
|
if( pattern_loop_channel == channel_idx ) {
|
||
|
/* Loop in progress on this channel. Next iteration. */
|
||
|
if( pattern_loop_count == 0 ) {
|
||
|
/* Loop finished. */
|
||
|
/* Invalidate current marker. */
|
||
|
channels[ channel_idx ].pattern_loop_row = current_row + 1;
|
||
|
} else {
|
||
|
/* Count must be higher than zero. */
|
||
|
/* Loop and cancel any breaks on this row. */
|
||
|
next_row = channels[ channel_idx ].pattern_loop_row;
|
||
|
next_sequence_index = current_sequence_index;
|
||
|
}
|
||
|
pattern_loop_count -= 1;
|
||
|
}
|
||
|
}
|
||
|
break;
|
||
|
case 0xE0:
|
||
|
/* Pattern delay.*/
|
||
|
tick_counter += ticks_per_row * ( effect_param & 0x0F );
|
||
|
break;
|
||
|
}
|
||
|
break;
|
||
|
case 0x0F:
|
||
|
/* Set Speed/Tempo.*/
|
||
|
if( effect_param < 32 ) {
|
||
|
set_speed( effect_param );
|
||
|
tick_counter = ticks_per_row;
|
||
|
} else {
|
||
|
set_tempo( effect_param );
|
||
|
}
|
||
|
break;
|
||
|
case 0x25:
|
||
|
/* S3M Set Speed.*/
|
||
|
set_speed( effect_param );
|
||
|
tick_counter = ticks_per_row;
|
||
|
break;
|
||
|
}
|
||
|
}
|
||
|
return song_end;
|
||
|
}
|
||
|
|
||
|
private void set_global_volume( int volume ) {
|
||
|
if( volume < 0 ) {
|
||
|
volume = 0;
|
||
|
}
|
||
|
if( volume > 64 ) {
|
||
|
volume = 64;
|
||
|
}
|
||
|
global_volume[ 0 ] = volume;
|
||
|
}
|
||
|
|
||
|
private void set_speed( int speed ) {
|
||
|
if( speed > 0 && speed < 256 ) {
|
||
|
ticks_per_row = speed;
|
||
|
}
|
||
|
}
|
||
|
|
||
|
private void set_tempo( int bpm ) {
|
||
|
if( bpm > 31 && bpm < 256 ) {
|
||
|
tick_length_samples = ( sampling_rate * 5 ) / ( bpm * 2 );
|
||
|
}
|
||
|
}
|
||
|
|
||
|
private void volume_ramp() {
|
||
|
int ramp_idx, next_idx, ramp_end;
|
||
|
int volume_ramp_delta, volume, sample;
|
||
|
sample = 0;
|
||
|
volume_ramp_delta = FP_ONE / volume_ramp_length;
|
||
|
volume = 0;
|
||
|
ramp_idx = 0;
|
||
|
next_idx = 2 * tick_length_samples;
|
||
|
ramp_end = volume_ramp_length * 2 - 1;
|
||
|
while( ramp_idx <= ramp_end ) {
|
||
|
sample = volume_ramp_buffer[ ramp_idx ] * ( FP_ONE - volume ) >> FP_SHIFT;
|
||
|
mixing_buffer[ ramp_idx ] = sample + ( mixing_buffer[ ramp_idx ] * volume >> FP_SHIFT );
|
||
|
volume_ramp_buffer[ ramp_idx ] = mixing_buffer[ next_idx + ramp_idx ];
|
||
|
sample = volume_ramp_buffer[ ramp_idx + 1 ] * ( FP_ONE - volume ) >> FP_SHIFT;
|
||
|
mixing_buffer[ ramp_idx + 1 ] = sample + ( mixing_buffer[ ramp_idx + 1 ] * volume >> FP_SHIFT );
|
||
|
volume_ramp_buffer[ ramp_idx + 1 ] = mixing_buffer[ next_idx + ramp_idx + 1 ];
|
||
|
volume += volume_ramp_delta;
|
||
|
ramp_idx += 2;
|
||
|
}
|
||
|
}
|
||
|
|
||
|
private void clear_vol_ramp_buffer() {
|
||
|
int ramp_idx, ramp_end;
|
||
|
ramp_idx = 0;
|
||
|
ramp_end = volume_ramp_length * 2 - 1;
|
||
|
while( ramp_idx <= ramp_end ) {
|
||
|
volume_ramp_buffer[ ramp_idx ] = 0;
|
||
|
ramp_idx += 1;
|
||
|
}
|
||
|
}
|
||
|
}
|
||
|
|