Add a hardware-accelerated 3DS audio backend
Hell yes
This commit is contained in:
parent
268d68b078
commit
d391ab77f9
3 changed files with 272 additions and 3 deletions
|
@ -25,7 +25,7 @@ option(LANCZOS_RESAMPLER "Use Lanczos filtering for audio resampling instead of
|
|||
option(FREETYPE_FONTS "Use FreeType2 to render the DejaVu Mono (English) or Migu1M (Japanese) fonts, instead of using pre-rendered copies of Courier New (English) and MS Gothic (Japanese)" OFF)
|
||||
|
||||
set(BACKEND_RENDERER "SDLTexture" CACHE STRING "Which renderer the game should use: 'OpenGL3' for an OpenGL 3.2 renderer, 'OpenGLES2' for an OpenGL ES 2.0 renderer, 'SDLTexture' for SDL2's hardware-accelerated Texture API, 'SDLSurface' for SDL2's software-rendered Surface API, 'Wii U' for the Wii U's hardware-accelerated GX2 API, '3DS' for the 3DS's hardware accelerated Citro2D/Citro3D API, or 'Software' for a handwritten software renderer")
|
||||
set(BACKEND_AUDIO "SDL2" CACHE STRING "Which audio backend the game should use: 'SDL2', 'miniaudio', 'WiiU-Hardware', 'WiiU-Software', '3DS', or 'Null'")
|
||||
set(BACKEND_AUDIO "SDL2" CACHE STRING "Which audio backend the game should use: 'SDL2', 'miniaudio', 'WiiU-Hardware', 'WiiU-Software', '3DS-Hardware', '3DS-Software', or 'Null'")
|
||||
set(BACKEND_PLATFORM "SDL2" CACHE STRING "Which platform backend the game should use: 'SDL2', 'GLFW3', 'WiiU', '3DS', or 'Null'")
|
||||
|
||||
option(LTO "Enable link-time optimisation" OFF)
|
||||
|
@ -371,7 +371,11 @@ elseif(BACKEND_AUDIO MATCHES "WiiU-Software")
|
|||
"src/Backends/Audio/SoftwareMixer/Backend.h"
|
||||
"src/Backends/Audio/SoftwareMixer/WiiU-Software.cpp"
|
||||
)
|
||||
elseif(BACKEND_AUDIO MATCHES "3DS")
|
||||
elseif(BACKEND_AUDIO MATCHES "3DS-Hardware")
|
||||
target_sources(CSE2 PRIVATE
|
||||
"src/Backends/Audio/3DS.cpp"
|
||||
)
|
||||
elseif(BACKEND_AUDIO MATCHES "3DS-Software")
|
||||
target_sources(CSE2 PRIVATE
|
||||
"src/Backends/Audio/SoftwareMixer.cpp"
|
||||
"src/Backends/Audio/SoftwareMixer/Mixer.cpp"
|
||||
|
|
|
@ -109,7 +109,8 @@ Name | Function
|
|||
`-DBACKEND_AUDIO=miniaudio` | Deliver audio with miniaudio (software-mixer)
|
||||
`-DBACKEND_AUDIO=WiiU-Hardware` | Deliver audio with Wii U's AXVoice API (hardware-accelerated) (WARNING - currently broken: voices randomly disappear for unknown reasons)
|
||||
`-DBACKEND_AUDIO=WiiU-Software` | Deliver audio with Wii U's AXVoice API (software-mixer)
|
||||
`-DBACKEND_AUDIO=3DS` | Deliver audio with 3DS's NDSP API (software-mixer)
|
||||
`-DBACKEND_AUDIO=3DS-Hardware` | Deliver audio with 3DS's NDSP API (hardware-accelerated)
|
||||
`-DBACKEND_AUDIO=3DS-Software` | Deliver audio with 3DS's NDSP API (software-mixer)
|
||||
`-DBACKEND_AUDIO=Null` | Don't deliver audio at all (WARNING - game will have no audio)
|
||||
`-DBACKEND_PLATFORM=SDL2` | (Default) Use SDL2 for miscellaneous platform-dependant operations
|
||||
`-DBACKEND_PLATFORM=GLFW3` | Use GLFW3 for miscellaneous platform-dependant operations
|
||||
|
|
264
src/Backends/Audio/3DS.cpp
Normal file
264
src/Backends/Audio/3DS.cpp
Normal file
|
@ -0,0 +1,264 @@
|
|||
#include "../Audio.h"
|
||||
|
||||
#include <math.h>
|
||||
#include <stddef.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
|
||||
#include <3ds.h>
|
||||
|
||||
#include "../Misc.h"
|
||||
|
||||
#define MIN(a, b) ((a) < (b) ? (a) : (b))
|
||||
#define MAX(a, b) ((a) > (b) ? (a) : (b))
|
||||
#define CLAMP(x, y, z) MIN(MAX((x), (y)), (z))
|
||||
|
||||
typedef struct AudioBackend_Sound
|
||||
{
|
||||
signed char *samples;
|
||||
ndspWaveBuf wave_buffer;
|
||||
unsigned int frequency;
|
||||
float volume;
|
||||
float pan_l;
|
||||
float pan_r;
|
||||
bool looping;
|
||||
int channel;
|
||||
unsigned int identifier;
|
||||
} AudioBackend_Sound;
|
||||
|
||||
static struct
|
||||
{
|
||||
unsigned int sound_identifier;
|
||||
AudioBackend_Sound *sound;
|
||||
} channels[24];
|
||||
|
||||
static void (*organya_callback)(void);
|
||||
static unsigned int organya_callback_timer;
|
||||
|
||||
static Thread audio_thread;
|
||||
static bool audio_thread_die;
|
||||
|
||||
static void OrganyaThread(void *user_data)
|
||||
{
|
||||
(void)user_data;
|
||||
|
||||
while (!audio_thread_die)
|
||||
{
|
||||
if (organya_callback_timer == 0)
|
||||
{
|
||||
// Is Organya isn't currently playing, idle for 10ms and check again
|
||||
svcSleepThread(10 * 1000000);
|
||||
}
|
||||
else
|
||||
{
|
||||
organya_callback();
|
||||
|
||||
svcSleepThread(organya_callback_timer * 1000000);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
static float MillibelToScale(long volume)
|
||||
{
|
||||
// Volume is in hundredths of a decibel, from 0 to -10000
|
||||
volume = CLAMP(volume, -10000, 0);
|
||||
return pow(10.0f, volume / 2000.0f);
|
||||
}
|
||||
|
||||
static int AllocateChannel(AudioBackend_Sound *sound)
|
||||
{
|
||||
// Search for a channel which either doesn't have an assigned sound,
|
||||
// or whose assigned sound has since stopped playing.
|
||||
for (int i = 0; i < 24; ++i)
|
||||
{
|
||||
if (channels[i].sound_identifier == 0
|
||||
|| channels[i].sound->wave_buffer.status == NDSP_WBUF_FREE
|
||||
|| channels[i].sound->wave_buffer.status == NDSP_WBUF_DONE)
|
||||
{
|
||||
channels[i].sound_identifier = sound->identifier;
|
||||
channels[i].sound = sound;
|
||||
|
||||
return i;
|
||||
}
|
||||
}
|
||||
|
||||
Backend_PrintInfo("Ran out of sound channels - hey you, whatever you're doing, stop it!");
|
||||
|
||||
return -1;
|
||||
}
|
||||
|
||||
bool AudioBackend_Init(void)
|
||||
{
|
||||
ndspInit();
|
||||
|
||||
ndspSetOutputMode(NDSP_OUTPUT_STEREO);
|
||||
|
||||
audio_thread_die = false;
|
||||
audio_thread = threadCreate(OrganyaThread, NULL, 32 * 1024, 0x18, -1, false);
|
||||
|
||||
return true;
|
||||
}
|
||||
|
||||
void AudioBackend_Deinit(void)
|
||||
{
|
||||
audio_thread_die = true;
|
||||
threadJoin(audio_thread, UINT64_MAX);
|
||||
threadFree(audio_thread);
|
||||
|
||||
ndspExit();
|
||||
}
|
||||
|
||||
AudioBackend_Sound* AudioBackend_CreateSound(unsigned int frequency, const unsigned char *samples, size_t length)
|
||||
{
|
||||
static unsigned int identifier_allocator;
|
||||
|
||||
AudioBackend_Sound *sound = (AudioBackend_Sound*)malloc(sizeof(AudioBackend_Sound));
|
||||
|
||||
if (sound != NULL)
|
||||
{
|
||||
sound->samples = (signed char*)linearAlloc(length);
|
||||
|
||||
if (sound->samples != NULL)
|
||||
{
|
||||
for (size_t i = 0; i < length; ++i)
|
||||
sound->samples[i] = samples[i] - 0x80;
|
||||
|
||||
DSP_FlushDataCache(sound->samples, length);
|
||||
|
||||
memset(&sound->wave_buffer, 0, sizeof(sound->wave_buffer));
|
||||
sound->wave_buffer.data_vaddr = sound->samples;
|
||||
sound->wave_buffer.nsamples = length;
|
||||
|
||||
sound->frequency = frequency;
|
||||
sound->volume = 1.0f;
|
||||
sound->pan_l = 1.0f;
|
||||
sound->pan_r = 1.0f;
|
||||
sound->looping = false;
|
||||
|
||||
sound->channel = -1;
|
||||
|
||||
do
|
||||
{
|
||||
sound->identifier = ++identifier_allocator;
|
||||
} while (sound->identifier == 0); // 0 is reserved
|
||||
|
||||
return sound;
|
||||
}
|
||||
else
|
||||
{
|
||||
Backend_PrintError("linearAlloc failed in AudioBackend_CreateSound");
|
||||
}
|
||||
|
||||
free(sound);
|
||||
}
|
||||
else
|
||||
{
|
||||
Backend_PrintError("malloc failed in AudioBackend_CreateSound");
|
||||
}
|
||||
|
||||
return NULL;
|
||||
}
|
||||
|
||||
void AudioBackend_DestroySound(AudioBackend_Sound *sound)
|
||||
{
|
||||
if (sound->channel != -1 && channels[sound->channel].sound_identifier == sound->identifier)
|
||||
{
|
||||
ndspChnWaveBufClear(sound->channel);
|
||||
channels[sound->channel].sound_identifier = 0;
|
||||
channels[sound->channel].sound = NULL;
|
||||
}
|
||||
|
||||
linearFree(sound->samples);
|
||||
free(sound);
|
||||
}
|
||||
|
||||
void AudioBackend_PlaySound(AudioBackend_Sound *sound, bool looping)
|
||||
{
|
||||
if (sound->channel == -1 || channels[sound->channel].sound_identifier != sound->identifier)
|
||||
sound->channel = AllocateChannel(sound);
|
||||
|
||||
bool previous_looping = sound->looping;
|
||||
sound->looping = looping;
|
||||
|
||||
if (sound->channel != -1)
|
||||
{
|
||||
if (sound->wave_buffer.status == NDSP_WBUF_FREE
|
||||
|| sound->wave_buffer.status == NDSP_WBUF_DONE
|
||||
|| previous_looping != looping)
|
||||
{
|
||||
ndspChnWaveBufClear(sound->channel);
|
||||
|
||||
ndspChnSetInterp(sound->channel, NDSP_INTERP_LINEAR);
|
||||
ndspChnSetRate(sound->channel, sound->frequency);
|
||||
ndspChnSetFormat(sound->channel, NDSP_FORMAT_MONO_PCM8);
|
||||
|
||||
sound->wave_buffer.looping = looping;
|
||||
|
||||
float mix[12];
|
||||
memset(mix, 0, sizeof(mix));
|
||||
mix[0] = sound->pan_l * sound->volume;
|
||||
mix[1] = sound->pan_r * sound->volume;
|
||||
ndspChnSetMix(sound->channel, mix);
|
||||
|
||||
ndspChnWaveBufAdd(sound->channel, &sound->wave_buffer);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
void AudioBackend_StopSound(AudioBackend_Sound *sound)
|
||||
{
|
||||
if (sound->channel != -1 && channels[sound->channel].sound_identifier == sound->identifier)
|
||||
ndspChnWaveBufClear(sound->channel);
|
||||
}
|
||||
|
||||
void AudioBackend_RewindSound(AudioBackend_Sound *sound)
|
||||
{
|
||||
(void)sound;
|
||||
}
|
||||
|
||||
void AudioBackend_SetSoundFrequency(AudioBackend_Sound *sound, unsigned int frequency)
|
||||
{
|
||||
sound->frequency = frequency;
|
||||
|
||||
if (sound->channel != -1 && channels[sound->channel].sound_identifier == sound->identifier)
|
||||
ndspChnSetRate(sound->channel, frequency);
|
||||
}
|
||||
|
||||
void AudioBackend_SetSoundVolume(AudioBackend_Sound *sound, long volume)
|
||||
{
|
||||
sound->volume = MillibelToScale(volume);
|
||||
|
||||
if (sound->channel != -1 && channels[sound->channel].sound_identifier == sound->identifier)
|
||||
{
|
||||
float mix[12];
|
||||
memset(mix, 0, sizeof(mix));
|
||||
mix[0] = sound->pan_l * sound->volume;
|
||||
mix[1] = sound->pan_r * sound->volume;
|
||||
ndspChnSetMix(sound->channel, mix);
|
||||
}
|
||||
}
|
||||
|
||||
void AudioBackend_SetSoundPan(AudioBackend_Sound *sound, long pan)
|
||||
{
|
||||
sound->pan_l = MillibelToScale(-pan);
|
||||
sound->pan_r = MillibelToScale(pan);
|
||||
|
||||
if (sound->channel != -1 && channels[sound->channel].sound_identifier == sound->identifier)
|
||||
{
|
||||
float mix[12];
|
||||
memset(mix, 0, sizeof(mix));
|
||||
mix[0] = sound->pan_l * sound->volume;
|
||||
mix[1] = sound->pan_r * sound->volume;
|
||||
ndspChnSetMix(sound->channel, mix);
|
||||
}
|
||||
}
|
||||
|
||||
void AudioBackend_SetOrganyaCallback(void (*callback)(void))
|
||||
{
|
||||
organya_callback = callback;
|
||||
}
|
||||
|
||||
void AudioBackend_SetOrganyaTimer(unsigned int milliseconds)
|
||||
{
|
||||
organya_callback_timer = milliseconds;
|
||||
}
|
Loading…
Add table
Reference in a new issue