mirror of
				https://github.com/mfulz/qmk_firmware.git
				synced 2025-10-25 03:29:59 +02:00 
			
		
		
		
	
		
			
				
	
	
		
			562 lines
		
	
	
		
			19 KiB
		
	
	
	
		
			C
		
	
	
	
	
	
			
		
		
	
	
			562 lines
		
	
	
		
			19 KiB
		
	
	
	
		
			C
		
	
	
	
	
	
| /* Copyright 2016-2020 Jack Humbert
 | |
|  * Copyright 2020 JohSchneider
 | |
| 
 | |
|  * This program is free software: you can redistribute it and/or modify
 | |
|  * it under the terms of the GNU General Public License as published by
 | |
|  * the Free Software Foundation, either version 2 of the License, or
 | |
|  * (at your option) any later version.
 | |
|  *
 | |
|  * This program 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 General Public License for more details.
 | |
|  *
 | |
|  * You should have received a copy of the GNU General Public License
 | |
|  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 | |
|  */
 | |
| #include "audio.h"
 | |
| #include "eeconfig.h"
 | |
| #include "timer.h"
 | |
| #include "wait.h"
 | |
| 
 | |
| /* audio system:
 | |
|  *
 | |
|  * audio.[ch] takes care of all overall state, tracking the actively playing
 | |
|  *            notes/tones; the notes a SONG consists of;
 | |
|  *            ...
 | |
|  *            = everything audio-related that is platform agnostic
 | |
|  *
 | |
|  * driver_[avr|chibios]_[dac|pwm] take care of the lower hardware dependent parts,
 | |
|  *            specific to each platform and the used subsystem/driver to drive
 | |
|  *            the output pins/channels with the calculated frequencies for each
 | |
|  *            active tone
 | |
|  *            as part of this, the driver has to trigger regular state updates by
 | |
|  *            calling 'audio_update_state' through some sort of timer - be it a
 | |
|  *            dedicated one or piggybacking on for example the timer used to
 | |
|  *            generate a pwm signal/clock.
 | |
|  *
 | |
|  *
 | |
|  * A Note on terminology:
 | |
|  * tone, pitch and frequency are used somewhat interchangeably, in a strict Wikipedia-sense:
 | |
|  *    "(Musical) tone, a sound characterized by its duration, pitch (=frequency),
 | |
|  *    intensity (=volume), and timbre"
 | |
|  * - intensity/volume is currently not handled at all, although the 'dac_additive' driver could do so
 | |
|  * - timbre is handled globally (TODO: only used with the pwm drivers at the moment)
 | |
|  *
 | |
|  * in musical_note.h a 'note' is the combination of a pitch and a duration
 | |
|  * these are used to create SONG arrays; during playback their frequencies
 | |
|  * are handled as single successive tones, while the durations are
 | |
|  * kept track of in 'audio_update_state'
 | |
|  *
 | |
|  * 'voice' as it is used here, equates to a sort of instrument with its own
 | |
|  * characteristics sound and effects
 | |
|  * the audio system as-is deals only with (possibly multiple) tones of one
 | |
|  * instrument/voice at a time (think: chords). since the number of tones that
 | |
|  * can be reproduced depends on the hardware/driver in use: pwm can only
 | |
|  * reproduce one tone per output/speaker; DACs can reproduce/mix multiple
 | |
|  * when doing additive synthesis.
 | |
|  *
 | |
|  * 'duration' can either be in the beats-per-minute related unit found in
 | |
|  * musical_notes.h, OR in ms; keyboards create SONGs with the former, while
 | |
|  * the internal state of the audio system does its calculations with the later - ms
 | |
|  */
 | |
| 
 | |
| #ifndef AUDIO_TONE_STACKSIZE
 | |
| #    define AUDIO_TONE_STACKSIZE 8
 | |
| #endif
 | |
| uint8_t        active_tones = 0;            // number of tones pushed onto the stack by audio_play_tone - might be more than the hardware is able to reproduce at any single time
 | |
| musical_tone_t tones[AUDIO_TONE_STACKSIZE]; // stack of currently active tones
 | |
| 
 | |
| bool playing_melody = false; // playing a SONG?
 | |
| bool playing_note   = false; // or (possibly multiple simultaneous) tones
 | |
| bool state_changed  = false; // global flag, which is set if anything changes with the active_tones
 | |
| 
 | |
| // melody/SONG related state variables
 | |
| float (*notes_pointer)[][2];                           // SONG, an array of MUSICAL_NOTEs
 | |
| uint16_t notes_count;                                  // length of the notes_pointer array
 | |
| bool     notes_repeat;                                 // PLAY_SONG or PLAY_LOOP?
 | |
| uint16_t melody_current_note_duration = 0;             // duration of the currently playing note from the active melody, in ms
 | |
| uint8_t  note_tempo                   = TEMPO_DEFAULT; // beats-per-minute
 | |
| uint16_t current_note                 = 0;             // index into the array at notes_pointer
 | |
| bool     note_resting                 = false;         // if a short pause was introduced between two notes with the same frequency while playing a melody
 | |
| uint16_t last_timestamp               = 0;
 | |
| 
 | |
| #ifdef AUDIO_ENABLE_TONE_MULTIPLEXING
 | |
| #    ifndef AUDIO_MAX_SIMULTANEOUS_TONES
 | |
| #        define AUDIO_MAX_SIMULTANEOUS_TONES 3
 | |
| #    endif
 | |
| uint16_t tone_multiplexing_rate        = AUDIO_TONE_MULTIPLEXING_RATE_DEFAULT;
 | |
| uint8_t  tone_multiplexing_index_shift = 0; // offset used on active-tone array access
 | |
| #endif
 | |
| 
 | |
| // provided and used by voices.c
 | |
| extern uint8_t  note_timbre;
 | |
| extern bool     glissando;
 | |
| extern bool     vibrato;
 | |
| extern uint16_t voices_timer;
 | |
| 
 | |
| #ifndef STARTUP_SONG
 | |
| #    define STARTUP_SONG SONG(STARTUP_SOUND)
 | |
| #endif
 | |
| #ifndef AUDIO_ON_SONG
 | |
| #    define AUDIO_ON_SONG SONG(AUDIO_ON_SOUND)
 | |
| #endif
 | |
| #ifndef AUDIO_OFF_SONG
 | |
| #    define AUDIO_OFF_SONG SONG(AUDIO_OFF_SOUND)
 | |
| #endif
 | |
| float startup_song[][2]   = STARTUP_SONG;
 | |
| float audio_on_song[][2]  = AUDIO_ON_SONG;
 | |
| float audio_off_song[][2] = AUDIO_OFF_SONG;
 | |
| 
 | |
| static bool    audio_initialized    = false;
 | |
| static bool    audio_driver_stopped = true;
 | |
| audio_config_t audio_config;
 | |
| 
 | |
| void audio_init() {
 | |
|     if (audio_initialized) {
 | |
|         return;
 | |
|     }
 | |
| 
 | |
|     // Check EEPROM
 | |
| #ifdef EEPROM_ENABLE
 | |
|     if (!eeconfig_is_enabled()) {
 | |
|         eeconfig_init();
 | |
|     }
 | |
|     audio_config.raw = eeconfig_read_audio();
 | |
| #else // EEPROM settings
 | |
|     audio_config.enable        = true;
 | |
| #    ifdef AUDIO_CLICKY_ON
 | |
|     audio_config.clicky_enable = true;
 | |
| #    endif
 | |
| #endif // EEPROM settings
 | |
| 
 | |
|     for (uint8_t i = 0; i < AUDIO_TONE_STACKSIZE; i++) {
 | |
|         tones[i] = (musical_tone_t){.time_started = 0, .pitch = -1.0f, .duration = 0};
 | |
|     }
 | |
| 
 | |
|     if (!audio_initialized) {
 | |
|         audio_driver_initialize();
 | |
|         audio_initialized = true;
 | |
|     }
 | |
|     stop_all_notes();
 | |
| #ifndef AUDIO_INIT_DELAY
 | |
|     audio_startup();
 | |
| #endif
 | |
| }
 | |
| 
 | |
| void audio_startup(void) {
 | |
|     if (audio_config.enable) {
 | |
|         PLAY_SONG(startup_song);
 | |
|     }
 | |
| 
 | |
|     last_timestamp = timer_read();
 | |
| }
 | |
| 
 | |
| void audio_toggle(void) {
 | |
|     if (audio_config.enable) {
 | |
|         stop_all_notes();
 | |
|     }
 | |
|     audio_config.enable ^= 1;
 | |
|     eeconfig_update_audio(audio_config.raw);
 | |
|     if (audio_config.enable) {
 | |
|         audio_on_user();
 | |
|     } else {
 | |
|         audio_off_user();
 | |
|     }
 | |
| }
 | |
| 
 | |
| void audio_on(void) {
 | |
|     audio_config.enable = 1;
 | |
|     eeconfig_update_audio(audio_config.raw);
 | |
|     audio_on_user();
 | |
|     PLAY_SONG(audio_on_song);
 | |
| }
 | |
| 
 | |
| void audio_off(void) {
 | |
|     PLAY_SONG(audio_off_song);
 | |
|     audio_off_user();
 | |
|     wait_ms(100);
 | |
|     audio_stop_all();
 | |
|     audio_config.enable = 0;
 | |
|     eeconfig_update_audio(audio_config.raw);
 | |
| }
 | |
| 
 | |
| bool audio_is_on(void) {
 | |
|     return (audio_config.enable != 0);
 | |
| }
 | |
| 
 | |
| void audio_stop_all() {
 | |
|     if (audio_driver_stopped) {
 | |
|         return;
 | |
|     }
 | |
| 
 | |
|     active_tones = 0;
 | |
| 
 | |
|     audio_driver_stop();
 | |
| 
 | |
|     playing_melody = false;
 | |
|     playing_note   = false;
 | |
| 
 | |
|     melody_current_note_duration = 0;
 | |
| 
 | |
|     for (uint8_t i = 0; i < AUDIO_TONE_STACKSIZE; i++) {
 | |
|         tones[i] = (musical_tone_t){.time_started = 0, .pitch = -1.0f, .duration = 0};
 | |
|     }
 | |
| 
 | |
|     audio_driver_stopped = true;
 | |
| }
 | |
| 
 | |
| void audio_stop_tone(float pitch) {
 | |
|     if (pitch < 0.0f) {
 | |
|         pitch = -1 * pitch;
 | |
|     }
 | |
| 
 | |
|     if (playing_note) {
 | |
|         if (!audio_initialized) {
 | |
|             audio_init();
 | |
|         }
 | |
|         bool found = false;
 | |
|         for (int i = AUDIO_TONE_STACKSIZE - 1; i >= 0; i--) {
 | |
|             found = (tones[i].pitch == pitch);
 | |
|             if (found) {
 | |
|                 tones[i] = (musical_tone_t){.time_started = 0, .pitch = -1.0f, .duration = 0};
 | |
|                 for (int j = i; (j < AUDIO_TONE_STACKSIZE - 1); j++) {
 | |
|                     tones[j]     = tones[j + 1];
 | |
|                     tones[j + 1] = (musical_tone_t){.time_started = 0, .pitch = -1.0f, .duration = 0};
 | |
|                 }
 | |
|                 break;
 | |
|             }
 | |
|         }
 | |
|         if (!found) {
 | |
|             return;
 | |
|         }
 | |
| 
 | |
|         state_changed = true;
 | |
|         active_tones--;
 | |
|         if (active_tones < 0) active_tones = 0;
 | |
| #ifdef AUDIO_ENABLE_TONE_MULTIPLEXING
 | |
|         if (tone_multiplexing_index_shift >= active_tones) {
 | |
|             tone_multiplexing_index_shift = 0;
 | |
|         }
 | |
| #endif
 | |
|         if (active_tones == 0) {
 | |
|             audio_driver_stop();
 | |
|             audio_driver_stopped = true;
 | |
|             playing_note         = false;
 | |
|         }
 | |
|     }
 | |
| }
 | |
| 
 | |
| void audio_play_note(float pitch, uint16_t duration) {
 | |
|     if (!audio_config.enable) {
 | |
|         return;
 | |
|     }
 | |
| 
 | |
|     if (!audio_initialized) {
 | |
|         audio_init();
 | |
|     }
 | |
| 
 | |
|     if (pitch < 0.0f) {
 | |
|         pitch = -1 * pitch;
 | |
|     }
 | |
| 
 | |
|     // round-robin: shifting out old tones, keeping only unique ones
 | |
|     // if the new frequency is already amongst the active tones, shift it to the top of the stack
 | |
|     bool found = false;
 | |
|     for (int i = active_tones - 1; i >= 0; i--) {
 | |
|         found = (tones[i].pitch == pitch);
 | |
|         if (found) {
 | |
|             for (int j = i; (j < active_tones - 1); j++) {
 | |
|                 tones[j]     = tones[j + 1];
 | |
|                 tones[j + 1] = (musical_tone_t){.time_started = timer_read(), .pitch = pitch, .duration = duration};
 | |
|             }
 | |
|             return; // since this frequency played already, the hardware was already started
 | |
|         }
 | |
|     }
 | |
| 
 | |
|     // frequency/tone is actually new, so we put it on the top of the stack
 | |
|     active_tones++;
 | |
|     if (active_tones > AUDIO_TONE_STACKSIZE) {
 | |
|         active_tones = AUDIO_TONE_STACKSIZE;
 | |
|         // shift out the oldest tone to make room
 | |
|         for (int i = 0; i < active_tones - 1; i++) {
 | |
|             tones[i] = tones[i + 1];
 | |
|         }
 | |
|     }
 | |
|     state_changed           = true;
 | |
|     playing_note            = true;
 | |
|     tones[active_tones - 1] = (musical_tone_t){.time_started = timer_read(), .pitch = pitch, .duration = duration};
 | |
| 
 | |
|     // TODO: needs to be handled per note/tone -> use its timestamp instead?
 | |
|     voices_timer = timer_read(); // reset to zero, for the effects added by voices.c
 | |
| 
 | |
|     if (audio_driver_stopped) {
 | |
|         audio_driver_start();
 | |
|         audio_driver_stopped = false;
 | |
|     }
 | |
| }
 | |
| 
 | |
| void audio_play_tone(float pitch) {
 | |
|     audio_play_note(pitch, 0xffff);
 | |
| }
 | |
| 
 | |
| void audio_play_melody(float (*np)[][2], uint16_t n_count, bool n_repeat) {
 | |
|     if (!audio_config.enable) {
 | |
|         audio_stop_all();
 | |
|         return;
 | |
|     }
 | |
| 
 | |
|     if (!audio_initialized) {
 | |
|         audio_init();
 | |
|     }
 | |
| 
 | |
|     // Cancel note if a note is playing
 | |
|     if (playing_note) audio_stop_all();
 | |
| 
 | |
|     playing_melody = true;
 | |
|     note_resting   = false;
 | |
| 
 | |
|     notes_pointer = np;
 | |
|     notes_count   = n_count;
 | |
|     notes_repeat  = n_repeat;
 | |
| 
 | |
|     current_note = 0; // note in the melody-array/list at note_pointer
 | |
| 
 | |
|     // start first note manually, which also starts the audio_driver
 | |
|     // all following/remaining notes are played by 'audio_update_state'
 | |
|     audio_play_note((*notes_pointer)[current_note][0], audio_duration_to_ms((*notes_pointer)[current_note][1]));
 | |
|     last_timestamp               = timer_read();
 | |
|     melody_current_note_duration = audio_duration_to_ms((*notes_pointer)[current_note][1]);
 | |
| }
 | |
| 
 | |
| float click[2][2];
 | |
| void  audio_play_click(uint16_t delay, float pitch, uint16_t duration) {
 | |
|     uint16_t duration_tone  = audio_ms_to_duration(duration);
 | |
|     uint16_t duration_delay = audio_ms_to_duration(delay);
 | |
| 
 | |
|     if (delay <= 0.0f) {
 | |
|         click[0][0] = pitch;
 | |
|         click[0][1] = duration_tone;
 | |
|         click[1][0] = 0.0f;
 | |
|         click[1][1] = 0.0f;
 | |
|         audio_play_melody(&click, 1, false);
 | |
|     } else {
 | |
|         // first note is a rest/pause
 | |
|         click[0][0] = 0.0f;
 | |
|         click[0][1] = duration_delay;
 | |
|         // second note is the actual click
 | |
|         click[1][0] = pitch;
 | |
|         click[1][1] = duration_tone;
 | |
|         audio_play_melody(&click, 2, false);
 | |
|     }
 | |
| }
 | |
| 
 | |
| bool audio_is_playing_note(void) {
 | |
|     return playing_note;
 | |
| }
 | |
| 
 | |
| bool audio_is_playing_melody(void) {
 | |
|     return playing_melody;
 | |
| }
 | |
| 
 | |
| uint8_t audio_get_number_of_active_tones(void) {
 | |
|     return active_tones;
 | |
| }
 | |
| 
 | |
| float audio_get_frequency(uint8_t tone_index) {
 | |
|     if (tone_index >= active_tones) {
 | |
|         return 0.0f;
 | |
|     }
 | |
|     return tones[active_tones - tone_index - 1].pitch;
 | |
| }
 | |
| 
 | |
| float audio_get_processed_frequency(uint8_t tone_index) {
 | |
|     if (tone_index >= active_tones) {
 | |
|         return 0.0f;
 | |
|     }
 | |
| 
 | |
|     int8_t index = active_tones - tone_index - 1;
 | |
|     // new tones are stacked on top (= appended at the end), so the most recent/current is MAX-1
 | |
| 
 | |
| #ifdef AUDIO_ENABLE_TONE_MULTIPLEXING
 | |
|     index = index - tone_multiplexing_index_shift;
 | |
|     if (index < 0) // wrap around
 | |
|         index += active_tones;
 | |
| #endif
 | |
| 
 | |
|     if (tones[index].pitch <= 0.0f) {
 | |
|         return 0.0f;
 | |
|     }
 | |
| 
 | |
|     return voice_envelope(tones[index].pitch);
 | |
| }
 | |
| 
 | |
| bool audio_update_state(void) {
 | |
|     if (!playing_note && !playing_melody) {
 | |
|         return false;
 | |
|     }
 | |
| 
 | |
|     bool     goto_next_note = false;
 | |
|     uint16_t current_time   = timer_read();
 | |
| 
 | |
|     if (playing_melody) {
 | |
|         goto_next_note = timer_elapsed(last_timestamp) >= melody_current_note_duration;
 | |
|         if (goto_next_note) {
 | |
|             uint16_t delta         = timer_elapsed(last_timestamp) - melody_current_note_duration;
 | |
|             last_timestamp         = current_time;
 | |
|             uint16_t previous_note = current_note;
 | |
|             current_note++;
 | |
|             voices_timer = timer_read(); // reset to zero, for the effects added by voices.c
 | |
| 
 | |
|             if (current_note >= notes_count) {
 | |
|                 if (notes_repeat) {
 | |
|                     current_note = 0;
 | |
|                 } else {
 | |
|                     audio_stop_all();
 | |
|                     return false;
 | |
|                 }
 | |
|             }
 | |
| 
 | |
|             if (!note_resting && (*notes_pointer)[previous_note][0] == (*notes_pointer)[current_note][0]) {
 | |
|                 note_resting = true;
 | |
| 
 | |
|                 // special handling for successive notes of the same frequency:
 | |
|                 // insert a short pause to separate them audibly
 | |
|                 audio_play_note(0.0f, audio_duration_to_ms(2));
 | |
|                 current_note                 = previous_note;
 | |
|                 melody_current_note_duration = audio_duration_to_ms(2);
 | |
| 
 | |
|             } else {
 | |
|                 note_resting = false;
 | |
| 
 | |
|                 // TODO: handle glissando here (or remember previous and current tone)
 | |
|                 /* there would need to be a freq(here we are) -> freq(next note)
 | |
|                  * and do slide/glissando in between problem here is to know which
 | |
|                  * frequency on the stack relates to what other? e.g. a melody starts
 | |
|                  * tones in a sequence, and stops expiring one, so the most recently
 | |
|                  * stopped is the starting point for a glissando to the most recently started?
 | |
|                  * how to detect and preserve this relation?
 | |
|                  * and what about user input, chords, ...?
 | |
|                  */
 | |
| 
 | |
|                 // '- delta': Skip forward in the next note's length if we've over shot
 | |
|                 //            the last, so the overall length of the song is the same
 | |
|                 uint16_t duration = audio_duration_to_ms((*notes_pointer)[current_note][1]);
 | |
| 
 | |
|                 // Skip forward past any completely missed notes
 | |
|                 while (delta > duration && current_note < notes_count - 1) {
 | |
|                     delta -= duration;
 | |
|                     current_note++;
 | |
|                     duration = audio_duration_to_ms((*notes_pointer)[current_note][1]);
 | |
|                 }
 | |
| 
 | |
|                 if (delta < duration) {
 | |
|                     duration -= delta;
 | |
|                 } else {
 | |
|                     // Only way to get here is if it is the last note and
 | |
|                     // we have completely missed it. Play it for 1ms...
 | |
|                     duration = 1;
 | |
|                 }
 | |
| 
 | |
|                 audio_play_note((*notes_pointer)[current_note][0], duration);
 | |
|                 melody_current_note_duration = duration;
 | |
|             }
 | |
|         }
 | |
|     }
 | |
| 
 | |
|     if (playing_note) {
 | |
| #ifdef AUDIO_ENABLE_TONE_MULTIPLEXING
 | |
|         tone_multiplexing_index_shift = (int)(current_time / tone_multiplexing_rate) % MIN(AUDIO_MAX_SIMULTANEOUS_TONES, active_tones);
 | |
|         goto_next_note                = true;
 | |
| #endif
 | |
|         if (vibrato || glissando) {
 | |
|             // force update on each cycle, since vibrato shifts the frequency slightly
 | |
|             goto_next_note = true;
 | |
|         }
 | |
| 
 | |
|         // housekeeping: stop notes that have no playtime left
 | |
|         for (int i = 0; i < active_tones; i++) {
 | |
|             if ((tones[i].duration != 0xffff) // indefinitely playing notes, started by 'audio_play_tone'
 | |
|                 && (tones[i].duration != 0)   // 'uninitialized'
 | |
|             ) {
 | |
|                 if (timer_elapsed(tones[i].time_started) >= tones[i].duration) {
 | |
|                     audio_stop_tone(tones[i].pitch); // also sets 'state_changed=true'
 | |
|                 }
 | |
|             }
 | |
|         }
 | |
|     }
 | |
| 
 | |
|     // state-changes have a higher priority, always triggering the hardware to update
 | |
|     if (state_changed) {
 | |
|         state_changed = false;
 | |
|         return true;
 | |
|     }
 | |
| 
 | |
|     return goto_next_note;
 | |
| }
 | |
| 
 | |
| // Tone-multiplexing functions
 | |
| #ifdef AUDIO_ENABLE_TONE_MULTIPLEXING
 | |
| void audio_set_tone_multiplexing_rate(uint16_t rate) {
 | |
|     tone_multiplexing_rate = rate;
 | |
| }
 | |
| void audio_enable_tone_multiplexing(void) {
 | |
|     tone_multiplexing_rate = AUDIO_TONE_MULTIPLEXING_RATE_DEFAULT;
 | |
| }
 | |
| void audio_disable_tone_multiplexing(void) {
 | |
|     tone_multiplexing_rate = 0;
 | |
| }
 | |
| void audio_increase_tone_multiplexing_rate(uint16_t change) {
 | |
|     if ((0xffff - change) > tone_multiplexing_rate) {
 | |
|         tone_multiplexing_rate += change;
 | |
|     }
 | |
| }
 | |
| void audio_decrease_tone_multiplexing_rate(uint16_t change) {
 | |
|     if (change <= tone_multiplexing_rate) {
 | |
|         tone_multiplexing_rate -= change;
 | |
|     }
 | |
| }
 | |
| #endif
 | |
| 
 | |
| // Tempo functions
 | |
| 
 | |
| void audio_set_tempo(uint8_t tempo) {
 | |
|     if (tempo < 10) note_tempo = 10;
 | |
|     //  else if (tempo > 250)
 | |
|     //      note_tempo = 250;
 | |
|     else
 | |
|         note_tempo = tempo;
 | |
| }
 | |
| 
 | |
| void audio_increase_tempo(uint8_t tempo_change) {
 | |
|     if (tempo_change > 255 - note_tempo)
 | |
|         note_tempo = 255;
 | |
|     else
 | |
|         note_tempo += tempo_change;
 | |
| }
 | |
| 
 | |
| void audio_decrease_tempo(uint8_t tempo_change) {
 | |
|     if (tempo_change >= note_tempo - 10)
 | |
|         note_tempo = 10;
 | |
|     else
 | |
|         note_tempo -= tempo_change;
 | |
| }
 | |
| 
 | |
| // TODO in the int-math version are some bugs; songs sometimes abruptly end - maybe an issue with the timer/system-tick wrapping around?
 | |
| uint16_t audio_duration_to_ms(uint16_t duration_bpm) {
 | |
| #if defined(__AVR__)
 | |
|     // doing int-math saves us some bytes in the overall firmware size, but the intermediate result is less accurate before being cast to/returned as uint
 | |
|     return ((uint32_t)duration_bpm * 60 * 1000) / (64 * note_tempo);
 | |
|     // NOTE: beware of uint16_t overflows when note_tempo is low and/or the duration is long
 | |
| #else
 | |
|     return ((float)duration_bpm * 60) / (64 * note_tempo) * 1000;
 | |
| #endif
 | |
| }
 | |
| uint16_t audio_ms_to_duration(uint16_t duration_ms) {
 | |
| #if defined(__AVR__)
 | |
|     return ((uint32_t)duration_ms * 64 * note_tempo) / 60 / 1000;
 | |
| #else
 | |
|     return ((float)duration_ms * 64 * note_tempo) / 60 / 1000;
 | |
| #endif
 | |
| }
 | 
