mirror of
https://github.com/nzp-team/fteqw.git
synced 2024-11-26 13:50:53 +00:00
4d25c073ec
fix crash from qwplayers with invalid modelindexes rework allow_skybox a little. now applies immediately. try to fix the appears-outside-of-map bug, again. dir *.wav now shows a little extra info on mouse-over. try loading music file extensions that we expect to be able to play via ffmpeg, not just the ones that are directly supported. rework the hidden fps_presets, show them with tab completion. fix a possible crash with r_temporalscenecache fix lightmap updates on submodels not happening properly with the scenecache. fix the serious memory leak with scenecache. add r_glsl_pbr cvar to force use of pbr pathways in our glsl. fix bug in alsa output not supporting float output properly. preliminary work to have the mixer use floating point mixing. disabled for now. try to update sys_register_file_associations on linux, still needs work though. try to work around nquake's config quirks, so config files don't get overwritten. repackage quake's conchars in order to add padding. this should avoid extra junk on the outside of glyphs. give fteqcc(commandline version) some extra package extration/creation support. our paks should be more end-user-friendly, and our spanned pk3s are awesome. write rune state to legacy saved games, so we don't ever have the missing-runes bug ever again... git-svn-id: https://svn.code.sf.net/p/fteqw/code/trunk@5780 fc73d0e0-1445-4013-8a0c-d673dee63da5
422 lines
17 KiB
C
422 lines
17 KiB
C
/*
|
|
Copyright (C) 1996-1997 Id Software, Inc.
|
|
|
|
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, write to the Free Software
|
|
Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
|
|
|
|
*/
|
|
// sound.h -- client sound i/o functions
|
|
|
|
#ifndef __SOUND__
|
|
#define __SOUND__
|
|
|
|
//#define MIXER_F32
|
|
#define MAXSOUNDCHANNELS 8 //on a per device basis
|
|
|
|
//pitch/rate changes require that we track stuff with subsample precision.
|
|
//this can result in some awkward overflows.
|
|
#define ssamplepos_t qintptr_t
|
|
#define usamplepos_t quintptr_t
|
|
#define PITCHSHIFT 6 /*max audio file length = ((1<<32)>>PITCHSHIFT)/KHZ*/
|
|
|
|
struct sfx_s;
|
|
|
|
typedef struct {
|
|
struct sfxcache_s *(QDECL *decodedata) (struct sfx_s *sfx, struct sfxcache_s *buf, ssamplepos_t start, int length); //return true when done.
|
|
float (QDECL *querydata) (struct sfx_s *sfx, struct sfxcache_s *buf, char *title, size_t titlesize); //reports length + original format info without actually decoding anything.
|
|
void (QDECL *ended) (struct sfx_s *sfx); //sound stopped playing and is now silent (allow rewinding or something).
|
|
void (QDECL *purge) (struct sfx_s *sfx); //sound is being purged from memory. destroy everything.
|
|
void *buf;
|
|
} sfxdecode_t;
|
|
|
|
enum
|
|
{
|
|
SLS_NOTLOADED, //not tried to load it
|
|
SLS_LOADING, //loading it on a worker thread.
|
|
SLS_LOADED, //currently in memory and usable.
|
|
SLS_FAILED //already tried to load it. it won't work. not found, invalid format, etc
|
|
};
|
|
typedef struct sfx_s
|
|
{
|
|
char name[MAX_OSPATH];
|
|
sfxdecode_t decoder;
|
|
|
|
int loadstate; //no more super-spammy
|
|
qboolean touched:1; //if the sound is still relevent
|
|
qboolean syspath:1; //if the sound is still relevent
|
|
|
|
int loopstart; //-1 or sample index to begin looping at once the sample ends
|
|
} sfx_t;
|
|
|
|
typedef enum
|
|
{
|
|
#ifdef FTE_TARGET_WEB
|
|
QAF_BLOB=0,
|
|
#endif
|
|
QAF_S8=1,
|
|
//QAF_U8=0x80|1,
|
|
QAF_S16=2,
|
|
//QAF_S32=4,
|
|
#ifdef MIXER_F32
|
|
QAF_F32=0x80|4,
|
|
#endif
|
|
#define QAF_BYTES(v) (v&0x7f) //to make memory allocation easier.
|
|
} qaudiofmt_t;
|
|
|
|
// !!! if this is changed, it much be changed in asm_i386.h too !!!
|
|
typedef struct sfxcache_s
|
|
{
|
|
usamplepos_t length; //sample count
|
|
unsigned int speed;
|
|
qaudiofmt_t format;
|
|
unsigned int numchannels;
|
|
usamplepos_t soundoffset; //byte index into the sound
|
|
qbyte *data; // variable sized
|
|
} sfxcache_t;
|
|
|
|
typedef struct
|
|
{
|
|
int numchannels; // this many samples per frame
|
|
int samples; // mono samples in buffer (individual, non grouped)
|
|
int samplepos; // in mono samples
|
|
int samplebytes; // per channel (NOT per frame)
|
|
enum
|
|
{
|
|
QSF_INVALID, //not selected yet...
|
|
QSF_EXTERNALMIXER, //this sample format is totally irrelevant as this device uses some sort of external mixer.
|
|
QSF_U8, //FIXME: more unsigned formats need changes to S_ClearBuffer
|
|
QSF_S8, //signed 8bit format is actually quite rare.
|
|
QSF_S16, //normal format
|
|
// QSF_X8_S24, //upper 8 bits unused. hopefully we don't need any packed thing
|
|
// QSF_S32, //lower 8 bits probably unused. this makes overflow detection messy.
|
|
QSF_F32, //modern mixers can use SSE/SIMD stuff, and we can skip clamping so this can be quite nippy.
|
|
} sampleformat;
|
|
int speed; // this many frames per second
|
|
unsigned char *buffer; // pointer to mixed pcm buffer (not directly used by mixer)
|
|
} dma_t;
|
|
|
|
//client and server
|
|
#define CF_SV_RELIABLE 1 // send reliably
|
|
#define CF_NET_SENTVELOCITY CF_SV_RELIABLE
|
|
#define CF_FORCELOOP 2 // forces looping. set on static sounds.
|
|
#define CF_NOSPACIALISE 4 // these sounds are played at a fixed volume in both speakers, but still gets quieter with distance.
|
|
//#define CF_PAUSED 8 // rate = 0. or something.
|
|
#define CF_CL_ABSVOLUME 16 // ignores volume cvar (but not mastervolume). this is ignored if received from the server because there's no practical way for the server to respect the client's preferences.
|
|
//#define CF_SV_RESERVED CF_CL_ABSVOLUME
|
|
#define CF_NOREVERB 32 // disables reverb on this channel, if possible.
|
|
#define CF_FOLLOW 64 // follows the owning entity (stops moving if we lose track)
|
|
//#define CF_RESERVEDN 128 // reserved for things that should be networked.
|
|
|
|
#define CF_SV_UNICAST 256 // serverside only. the sound is sent to msg_entity only.
|
|
#define CF_SV_SENDVELOCITY 512 // serverside hint that velocity is important
|
|
#define CF_CLI_AUTOSOUND 1024 // generated from q2 entities, which avoids breaking regular sounds, using it outside the sound system will probably break things.
|
|
#define CF_CLI_INACTIVE 2048 // try to play even when inactive
|
|
#ifdef Q3CLIENT
|
|
#define CF_CLI_NODUPES 4096 // block multiple identical sounds being started on the same entity within rapid succession. required by quake3.
|
|
#endif
|
|
#define CF_NETWORKED (CF_NOSPACIALISE|CF_NOREVERB|CF_FORCELOOP|CF_FOLLOW/*|CF_RESERVEDN*/)
|
|
|
|
typedef struct
|
|
{
|
|
sfx_t *sfx; // sfx number
|
|
int vol[MAXSOUNDCHANNELS]; // volume, 0.8 fixed point.
|
|
ssamplepos_t pos; // sample position in sfx, <0 means delay sound start (shifted up by PITCHSHIFT)
|
|
int rate; // fixed point rate scaling
|
|
int flags; // cf_ flags
|
|
int entnum; // to allow overriding a specific sound
|
|
int entchannel; // to avoid overriding a specific sound too easily
|
|
vec3_t origin; // origin of sound effect
|
|
vec3_t velocity; // velocity of sound effect
|
|
vec_t dist_mult; // distance multiplier (attenuation/clipK)
|
|
int master_vol; // 0-255 master volume
|
|
#ifdef Q3CLIENT
|
|
unsigned int starttime; // start time, to replicate q3's 50ms embargo on duped sounds.
|
|
#endif
|
|
} channel_t;
|
|
|
|
struct soundcardinfo_s;
|
|
typedef struct soundcardinfo_s soundcardinfo_t;
|
|
|
|
extern struct sndreverbproperties_s
|
|
{
|
|
int modificationcount;
|
|
struct reverbproperties_s
|
|
{ //note: this struct originally comes from openal's eaxreverb
|
|
//it is shared with gamecode
|
|
float flDensity;
|
|
float flDiffusion;
|
|
float flGain;
|
|
float flGainHF;
|
|
float flGainLF;
|
|
float flDecayTime;
|
|
float flDecayHFRatio;
|
|
float flDecayLFRatio;
|
|
float flReflectionsGain;
|
|
float flReflectionsDelay;
|
|
float flReflectionsPan[3];
|
|
float flLateReverbGain;
|
|
float flLateReverbDelay;
|
|
float flLateReverbPan[3];
|
|
float flEchoTime;
|
|
float flEchoDepth;
|
|
float flModulationTime;
|
|
float flModulationDepth;
|
|
float flAirAbsorptionGainHF;
|
|
float flHFReference;
|
|
float flLFReference;
|
|
float flRoomRolloffFactor;
|
|
int iDecayHFLimit;
|
|
} props;
|
|
} *reverbproperties;
|
|
extern size_t numreverbproperties;
|
|
|
|
//reverbproperties_s presets, from efx-presets.h
|
|
//mostly for testing
|
|
#define REVERB_PRESET_PSYCHOTIC \
|
|
{ 0.0625f, 0.5000f, 0.3162f, 0.8404f, 1.0000f, 7.5600f, 0.9100f, 1.0000f, 0.4864f, 0.0200f, { 0.0000f, 0.0000f, 0.0000f }, 2.4378f, 0.0300f, { 0.0000f, 0.0000f, 0.0000f }, 0.2500f, 0.0000f, 4.0000f, 1.0000f, 0.9943f, 5000.0000f, 250.0000f, 0.0000f, 0x0 }
|
|
//default reverb 1
|
|
#define REVERB_PRESET_UNDERWATER \
|
|
{ 0.3645f, 1.0000f, 0.3162f, 0.0100f, 1.0000f, 1.4900f, 0.1000f, 1.0000f, 0.5963f, 0.0070f, { 0.0000f, 0.0000f, 0.0000f }, 7.0795f, 0.0110f, { 0.0000f, 0.0000f, 0.0000f }, 0.2500f, 0.0000f, 1.1800f, 0.3480f, 0.9943f, 5000.0000f, 250.0000f, 0.0000f, 0x1 }
|
|
|
|
void S_Init (void);
|
|
void S_Startup (void);
|
|
void S_EnumerateDevices(void);
|
|
void S_Shutdown (qboolean final);
|
|
float S_GetSoundTime(int entnum, int entchannel);
|
|
float S_GetChannelLevel(int entnum, int entchannel);
|
|
void S_StartSound (int entnum, int entchannel, sfx_t *sfx, vec3_t origin, vec3_t velocity, float fvol, float attenuation, float timeofs, float pitchadj, unsigned int flags);
|
|
float S_UpdateSound(int entnum, int entchannel, sfx_t *sfx, vec3_t origin, vec3_t velocity, float fvol, float attenuation, float timeofs, float pitchadj, unsigned int flags);
|
|
void S_StaticSound (sfx_t *sfx, vec3_t origin, float vol, float attenuation);
|
|
void S_StopSound (int entnum, int entchannel);
|
|
void S_StopAllSounds(qboolean clear);
|
|
void S_UpdateListener(int seat, int entnum, vec3_t origin, vec3_t forward, vec3_t right, vec3_t up, size_t reverbtype, vec3_t velocity);
|
|
qboolean S_UpdateReverb(size_t reverbtype, void *reverb, size_t reverbsize);
|
|
void S_GetListenerInfo(int seat, float *origin, float *forward, float *right, float *up);
|
|
void S_Update (void);
|
|
void S_ExtraUpdate (void);
|
|
void S_MixerThread(soundcardinfo_t *sc);
|
|
void S_Purge(qboolean retaintouched);
|
|
|
|
void S_LockMixer(void);
|
|
void S_UnlockMixer(void);
|
|
|
|
qboolean S_HaveOutput(void);
|
|
|
|
void S_Music_Clear(sfx_t *onlyifsample);
|
|
void S_Music_Seek(float time);
|
|
qboolean S_GetMusicInfo(int musicchannel, float *time, float *duration, char *title, size_t titlesize);
|
|
qboolean S_Music_Playing(int musicchannel);
|
|
float Media_CrossFade(int musicchanel, float vol, float time); //queries the volume we're meant to be playing (checks for fade out). -1 for no more, otherwise returns vol.
|
|
sfx_t *Media_NextTrack(int musicchanel, float *time); //queries the track we're meant to be playing now.
|
|
|
|
sfx_t *S_FindName (const char *name, qboolean create, qboolean syspath);
|
|
sfx_t *S_PrecacheSound2 (const char *sample, qboolean syspath);
|
|
#define S_PrecacheSound(s) S_PrecacheSound2(s,false)
|
|
void S_TouchSound (char *sample);
|
|
void S_UntouchAll(void);
|
|
void S_ClearPrecache (void);
|
|
void S_BeginPrecaching (void);
|
|
void S_EndPrecaching (void);
|
|
|
|
void S_PaintChannels(soundcardinfo_t *sc, int endtime);
|
|
void S_InitPaintChannels (soundcardinfo_t *sc);
|
|
|
|
soundcardinfo_t *S_SetupDeviceSeat(char *driver, char *device, int seat);
|
|
void S_ShutdownCard (soundcardinfo_t *sc);
|
|
|
|
void S_DefaultSpeakerConfiguration(soundcardinfo_t *sc);
|
|
void S_ResetFailedLoad(void);
|
|
|
|
#ifdef PEXT2_VOICECHAT
|
|
void S_Voip_Parse(void);
|
|
#endif
|
|
#ifdef VOICECHAT
|
|
extern cvar_t snd_voip_showmeter;
|
|
void S_Voip_Transmit(unsigned char clc, sizebuf_t *buf);
|
|
void S_Voip_MapChange(void);
|
|
int S_Voip_Loudness(qboolean ignorevad); //-1 for not capturing, otherwise between 0 and 100
|
|
int S_Voip_ClientLoudness(unsigned int plno);
|
|
qboolean S_Voip_Speaking(unsigned int plno);
|
|
void S_Voip_Ignore(unsigned int plno, qboolean ignore);
|
|
#else
|
|
#define S_Voip_Loudness() -1
|
|
#define S_Voip_Speaking(p) false
|
|
#define S_Voip_Ignore(p,s)
|
|
#endif
|
|
|
|
qboolean S_IsPlayingSomewhere(sfx_t *s);
|
|
//qboolean ResampleSfx (sfx_t *sfx, int inrate, int inchannels, int inwidth, int insamps, int inloopstart, qbyte *data);
|
|
|
|
// picks a channel based on priorities, empty slots, number of channels
|
|
channel_t *SND_PickChannel(soundcardinfo_t *sc, int entnum, int entchannel);
|
|
|
|
void SND_ResampleStream (void *in, int inrate, qaudiofmt_t inwidth, int inchannels, int insamps, void *out, int outrate, qaudiofmt_t outwidth, int outchannels, int resampstyle);
|
|
|
|
// restart entire sound subsystem (doesn't flush old sounds, so make sure that happens)
|
|
void S_DoRestart (qboolean onlyifneeded);
|
|
|
|
void S_Restart_f (void);
|
|
|
|
//plays streaming audio
|
|
void S_RawAudio(int sourceid, qbyte *data, int speed, int samples, int channels, qaudiofmt_t width, float volume);
|
|
|
|
void CLVC_Poll (void);
|
|
|
|
void SNDVC_MicInput(qbyte *buffer, int samples, int freq, int width);
|
|
|
|
|
|
|
|
// ====================================================================
|
|
// User-setable variables
|
|
// ====================================================================
|
|
|
|
#define MAX_DYNAMIC_CHANNELS 64 /*playing sounds (identical ones merge)*/
|
|
|
|
|
|
#define NUM_MUSICS 1
|
|
|
|
#define AMBIENT_FIRST 0
|
|
#define AMBIENT_STOP NUM_AMBIENTS
|
|
#define MUSIC_FIRST AMBIENT_STOP
|
|
#define MUSIC_STOP (MUSIC_FIRST + NUM_MUSICS)
|
|
#define DYNAMIC_FIRST MUSIC_STOP
|
|
#define DYNAMIC_STOP (DYNAMIC_FIRST + MAX_DYNAMIC_CHANNELS)
|
|
|
|
//
|
|
// Fake dma is a synchronous faking of the DMA progress used for
|
|
// isolating performance in the renderer. The fakedma_updates is
|
|
// number of times S_Update() is called per second.
|
|
//
|
|
|
|
extern int snd_speed;
|
|
|
|
extern cvar_t snd_nominaldistance;
|
|
|
|
extern cvar_t snd_loadas8bit;
|
|
extern cvar_t bgmvolume;
|
|
extern cvar_t volume, mastervolume;
|
|
extern cvar_t snd_capture;
|
|
extern cvar_t nosound;
|
|
|
|
extern float voicevolumemod;
|
|
|
|
extern qboolean snd_initialized;
|
|
extern cvar_t snd_mixerthread;
|
|
|
|
extern int snd_blocked;
|
|
|
|
void S_LocalSound (const char *s);
|
|
void S_LocalSound2 (const char *sound, int channel, float volume);
|
|
qboolean S_LoadSound (sfx_t *s, qboolean forcedecode);
|
|
|
|
typedef qboolean (QDECL *S_LoadSound_t) (sfx_t *s, qbyte *data, size_t datalen, int sndspeed, qboolean forcedecode);
|
|
qboolean S_RegisterSoundInputPlugin(void *module, S_LoadSound_t loadfnc); //called to register additional sound input plugins
|
|
void S_UnregisterSoundInputModule(void *module);
|
|
|
|
void S_AmbientOff (void);
|
|
void S_AmbientOn (void);
|
|
|
|
|
|
//inititalisation functions.
|
|
typedef struct
|
|
{
|
|
const char *name; //must be a single token, with no :
|
|
qboolean (QDECL *InitCard) (soundcardinfo_t *sc, const char *cardname); //NULL for default device.
|
|
qboolean (QDECL *Enumerate) (void (QDECL *callback) (const char *drivername, const char *devicecode, const char *readablename));
|
|
void (QDECL *RegisterCvars) (void);
|
|
} sounddriver_t;
|
|
/*typedef int (*sounddriver) (soundcardinfo_t *sc, int cardnum);
|
|
extern sounddriver pOPENAL_InitCard;
|
|
extern sounddriver pDSOUND_InitCard;
|
|
extern sounddriver pALSA_InitCard;
|
|
extern sounddriver pSNDIO_InitCard;
|
|
extern sounddriver pOSS_InitCard;
|
|
extern sounddriver pSDL_InitCard;
|
|
extern sounddriver pWAV_InitCard;
|
|
extern sounddriver pAHI_InitCard;
|
|
*/
|
|
|
|
typedef enum
|
|
{
|
|
CUR_SPACIALISEONLY = 0, //for ticking over, respacialising, etc
|
|
CUR_UPDATE = (1u<<1), //flags/rate/offset changed without changing the sound itself
|
|
CUR_SOUNDCHANGE = (1u<<2), //the audio file changed too. reset everything.
|
|
CUR_EVERYTHING = CUR_UPDATE|CUR_SOUNDCHANGE
|
|
} chanupdatereason_t;
|
|
|
|
struct soundcardinfo_s { //windows has one defined AFTER directsound
|
|
char name[256]; //a description of the card.
|
|
char guid[256]; //device name as detected (so input code can create sound devices without bugging out too much)
|
|
struct soundcardinfo_s *next;
|
|
int seat;
|
|
|
|
//speaker orientations for spacialisation.
|
|
float dist[MAXSOUNDCHANNELS];
|
|
|
|
vec3_t speakerdir[MAXSOUNDCHANNELS];
|
|
|
|
//info on which sound effects are playing
|
|
//FIXME: use a linked list
|
|
channel_t *channel;
|
|
size_t total_chans;
|
|
size_t max_chans;
|
|
|
|
float ambientlevels[NUM_AMBIENTS]; //we use a float instead of the channel's int volume value to avoid framerate dependancies with slow transitions.
|
|
|
|
//mixer
|
|
volatile dma_t sn; //why is this volatile?
|
|
qboolean inactive_sound; //continue mixing for this card even when the window isn't active.
|
|
qboolean selfpainting; //allow the sound code to call the right functions when it feels the need (not properly supported).
|
|
|
|
int paintedtime; //used in the mixer as last-written pos (in frames)
|
|
int oldsamplepos; //this is used to track buffer wraps
|
|
int buffers; //used to keep track of how many buffer wraps for consistant sound
|
|
int samplequeue; //this is the number of samples the device can enqueue. if set, DMAPos returns the write point (rather than hardware read point) (in samplepairs).
|
|
|
|
//callbacks
|
|
void *(*Lock) (soundcardinfo_t *sc, unsigned int *startoffset); //grab a pointer to the hardware ringbuffer or whatever. startoffset is the starting offset. you can set it to 0 and bump the start offset if you need.
|
|
void (*Unlock) (soundcardinfo_t *sc, void *buffer); //release the hardware ringbuffer memory
|
|
void (*Submit) (soundcardinfo_t *sc, int start, int end); //if the ringbuffer is emulated, this is where you should push it to the device.
|
|
void (*Shutdown) (soundcardinfo_t *sc); //kill the device
|
|
unsigned int (*GetDMAPos) (soundcardinfo_t *sc); //get the current point that the hardware is reading from (the return value should not wrap, at least not very often)
|
|
void (*SetEnvironmentReverb) (soundcardinfo_t *sc, size_t reverb); //if you have eax enabled, change the environment. fixme. generally this is a stub. optional.
|
|
void (*Restore) (soundcardinfo_t *sc); //called before lock/unlock/lock/unlock/submit. optional
|
|
void (*ChannelUpdate) (soundcardinfo_t *sc, channel_t *channel, chanupdatereason_t schanged); //properties of a sound effect changed. this is to notify hardware mixers. optional.
|
|
void (*ListenerUpdate) (soundcardinfo_t *sc, int entnum, vec3_t origin, vec3_t forward, vec3_t right, vec3_t up, vec3_t velocity); //player moved or something. this is to notify hardware mixers. optional.
|
|
ssamplepos_t (*GetChannelPos) (soundcardinfo_t *sc, channel_t *channel); //queries a hardware mixer's channel position (essentially returns channel->pos, except more up to date)
|
|
|
|
//driver-specific - if you need more stuff, you should just shove it in the handle pointer
|
|
void *thread;
|
|
void *handle;
|
|
int snd_sent;
|
|
int snd_completed;
|
|
int audio_fd;
|
|
};
|
|
|
|
extern soundcardinfo_t *sndcardinfo;
|
|
|
|
typedef struct
|
|
{
|
|
int apiver;
|
|
char *drivername;
|
|
qboolean (QDECL *Enumerate) (void (QDECL *callback) (const char *drivername, const char *devicecode, const char *readablename));
|
|
void *(QDECL *Init) (int samplerate, const char *device); /*create a new context*/
|
|
void (QDECL *Start) (void *ctx); /*begin grabbing new data, old data is potentially flushed*/
|
|
unsigned int (QDECL *Update) (void *ctx, unsigned char *buffer, unsigned int minbytes, unsigned int maxbytes); /*grab the data into a different buffer*/
|
|
void (QDECL *Stop) (void *ctx); /*stop grabbing new data, old data may remain*/
|
|
void (QDECL *Shutdown) (void *ctx); /*destroy everything*/
|
|
} snd_capture_driver_t;
|
|
|
|
#endif
|