Mercurial > audio-send

annotate org/ear.org @ 20:e8ae40c9848c

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
fixed 1,000,000 spelling errors
author Robert McIntyre <rlm@mit.edu>
date Thu, 03 Nov 2011 15:02:18 -0700
parents 22ac5a0367cd
children 0ee04505a37f
rev   line source
rlm@15 1 #+title: Simulated Sense of Hearing
rlm@0 2 #+author: Robert McIntyre
rlm@0 3 #+email: rlm@mit.edu
rlm@15 4 #+description: Simulating multiple listeners and the sense of hearing in jMonkeyEngine3
rlm@15 5 #+keywords: simulated hearing, openal, clojure, jMonkeyEngine3, LWJGL, AI
rlm@15 6 #+SETUPFILE: ../../aurellem/org/setup.org
rlm@15 7 #+INCLUDE: ../../aurellem/org/level-0.org
rlm@0 8 #+BABEL: :exports both :noweb yes :cache no :mkdirp yes
rlm@0 9
rlm@15 10 * Hearing
rlm@0 11
rlm@20 12 I want to be able to place ears in a similar manner to how I place
rlm@0 13 the eyes. I want to be able to place ears in a unique spatial
rlm@20 14 position, and receive as output at every tick the F.F.T. of whatever
rlm@0 15 signals are happening at that point.
rlm@0 16
rlm@15 17 Hearing is one of the more difficult senses to simulate, because there
rlm@15 18 is less support for obtaining the actual sound data that is processed
rlm@15 19 by jMonkeyEngine3.
rlm@15 20
rlm@15 21 jMonkeyEngine's sound system works as follows:
rlm@15 22
rlm@20 23 - jMonkeyEngine uses the =AppSettings= for the particular application
rlm@15 24 to determine what sort of =AudioRenderer= should be used.
rlm@15 25 - although some support is provided for multiple AudioRendering
rlm@15 26 backends, jMonkeyEngine at the time of this writing will either
rlm@20 27 pick no AudioRenderer at all, or the =LwjglAudioRenderer=
rlm@15 28 - jMonkeyEngine tries to figure out what sort of system you're
rlm@20 29 running and extracts the appropriate native libraries.
rlm@18 30 - the =LwjglAudioRenderer= uses the [[http://lwjgl.org/][=LWJGL=]] (LightWeight Java Game
rlm@18 31 Library) bindings to interface with a C library called [[http://kcat.strangesoft.net/openal.html][=OpenAL=]]
rlm@15 32 - =OpenAL= calculates the 3D sound localization and feeds a stream of
rlm@15 33 sound to any of various sound output devices with which it knows
rlm@15 34 how to communicate.
rlm@15 35
rlm@15 36 A consequence of this is that there's no way to access the actual
rlm@20 37 sound data produced by =OpenAL=. Even worse, =OpenAL= only supports
rlm@15 38 one /listener/, which normally isn't a problem for games, but becomes
rlm@15 39 a problem when trying to make multiple AI creatures that can each hear
rlm@15 40 the world from a different perspective.
rlm@15 41
rlm@15 42 To make many AI creatures in jMonkeyEngine that can each hear the
rlm@15 43 world from their own perspective, it is necessary to go all the way
rlm@15 44 back to =OpenAL= and implement support for simulated hearing there.
rlm@15 45
rlm@19 46 * Extending =OpenAL=
rlm@15 47 ** =OpenAL= Devices
rlm@15 48
rlm@15 49 =OpenAL= goes to great lengths to support many different systems, all
rlm@20 50 with different sound capabilities and interfaces. It accomplishes this
rlm@15 51 difficult task by providing code for many different sound backends in
rlm@15 52 pseudo-objects called /Devices/. There's a device for the Linux Open
rlm@20 53 Sound System and the Advanced Linux Sound Architecture, there's one
rlm@15 54 for Direct Sound on Windows, there's even one for Solaris. =OpenAL=
rlm@15 55 solves the problem of platform independence by providing all these
rlm@15 56 Devices.
rlm@15 57
rlm@15 58 Wrapper libraries such as LWJGL are free to examine the system on
rlm@20 59 which they are running and then select an appropriate device for that
rlm@15 60 system.
rlm@15 61
rlm@15 62 There are also a few "special" devices that don't interface with any
rlm@15 63 particular system. These include the Null Device, which doesn't do
rlm@20 64 anything, and the Wave Device, which writes whatever sound it receives
rlm@15 65 to a file, if everything has been set up correctly when configuring
rlm@15 66 =OpenAL=.
rlm@15 67
rlm@15 68 Actual mixing of the sound data happens in the Devices, and they are
rlm@15 69 the only point in the sound rendering process where this data is
rlm@15 70 available.
rlm@15 71
rlm@15 72 Therefore, in order to support multiple listeners, and get the sound
rlm@15 73 data in a form that the AIs can use, it is necessary to create a new
rlm@15 74 Device, which supports this features.
rlm@15 75
rlm@15 76 ** The Send Device
rlm@15 77 Adding a device to OpenAL is rather tricky -- there are five separate
rlm@15 78 files in the =OpenAL= source tree that must be modified to do so. I've
rlm@15 79 documented this process [[./add-new-device.org][here]] for anyone who is interested.
rlm@15 80
rlm@18 81
rlm@18 82 Onward to that actual Device!
rlm@18 83
rlm@18 84 again, my objectives are:
rlm@18 85
rlm@18 86 - Support Multiple Listeners from jMonkeyEngine3
rlm@18 87 - Get access to the rendered sound data for further processing from
rlm@18 88 clojure.
rlm@18 89
rlm@18 90 ** =send.c=
rlm@18 91
rlm@18 92 ** Header
rlm@18 93 #+srcname: send-header
rlm@15 94 #+begin_src C
rlm@15 95 #include "config.h"
rlm@15 96 #include <stdlib.h>
rlm@15 97 #include "alMain.h"
rlm@15 98 #include "AL/al.h"
rlm@15 99 #include "AL/alc.h"
rlm@15 100 #include "alSource.h"
rlm@15 101 #include <jni.h>
rlm@15 102
rlm@15 103 //////////////////// Summary
rlm@15 104
rlm@15 105 struct send_data;
rlm@15 106 struct context_data;
rlm@15 107
rlm@15 108 static void addContext(ALCdevice *, ALCcontext *);
rlm@15 109 static void syncContexts(ALCcontext *master, ALCcontext *slave);
rlm@15 110 static void syncSources(ALsource *master, ALsource *slave,
rlm@15 111 ALCcontext *masterCtx, ALCcontext *slaveCtx);
rlm@15 112
rlm@15 113 static void syncSourcei(ALuint master, ALuint slave,
rlm@15 114 ALCcontext *masterCtx, ALCcontext *ctx2, ALenum param);
rlm@15 115 static void syncSourcef(ALuint master, ALuint slave,
rlm@15 116 ALCcontext *masterCtx, ALCcontext *ctx2, ALenum param);
rlm@15 117 static void syncSource3f(ALuint master, ALuint slave,
rlm@15 118 ALCcontext *masterCtx, ALCcontext *ctx2, ALenum param);
rlm@15 119
rlm@15 120 static void swapInContext(ALCdevice *, struct context_data *);
rlm@15 121 static void saveContext(ALCdevice *, struct context_data *);
rlm@15 122 static void limitContext(ALCdevice *, ALCcontext *);
rlm@15 123 static void unLimitContext(ALCdevice *);
rlm@15 124
rlm@15 125 static void init(ALCdevice *);
rlm@15 126 static void renderData(ALCdevice *, int samples);
rlm@15 127
rlm@15 128 #define UNUSED(x) (void)(x)
rlm@18 129 #+end_src
rlm@15 130
rlm@20 131 The main idea behind the Send device is to take advantage of the fact
rlm@18 132 that LWJGL only manages one /context/ when using OpenAL. A /context/
rlm@18 133 is like a container that holds samples and keeps track of where the
rlm@18 134 listener is. In order to support multiple listeners, the Send device
rlm@18 135 identifies the LWJGL context as the master context, and creates any
rlm@18 136 number of slave contexts to represent additional listeners. Every
rlm@18 137 time the device renders sound, it synchronizes every source from the
rlm@18 138 master LWJGL context to the slave contexts. Then, it renders each
rlm@18 139 context separately, using a different listener for each one. The
rlm@18 140 rendered sound is made available via JNI to jMonkeyEngine.
rlm@18 141
rlm@18 142 To recap, the process is:
rlm@18 143 - Set the LWJGL context as "master" in the =init()= method.
rlm@18 144 - Create any number of additional contexts via =addContext()=
rlm@18 145 - At every call to =renderData()= sync the master context with the
rlm@20 146 slave contexts with =syncContexts()=
rlm@18 147 - =syncContexts()= calls =syncSources()= to sync all the sources
rlm@18 148 which are in the master context.
rlm@18 149 - =limitContext()= and =unLimitContext()= make it possible to render
rlm@18 150 only one context at a time.
rlm@18 151
rlm@18 152 ** Necessary State
rlm@18 153 #+begin_src C
rlm@15 154 //////////////////// State
rlm@15 155
rlm@15 156 typedef struct context_data {
rlm@15 157 ALfloat ClickRemoval[MAXCHANNELS];
rlm@15 158 ALfloat PendingClicks[MAXCHANNELS];
rlm@15 159 ALvoid *renderBuffer;
rlm@15 160 ALCcontext *ctx;
rlm@15 161 } context_data;
rlm@15 162
rlm@15 163 typedef struct send_data {
rlm@15 164 ALuint size;
rlm@15 165 context_data **contexts;
rlm@15 166 ALuint numContexts;
rlm@15 167 ALuint maxContexts;
rlm@15 168 } send_data;
rlm@18 169 #+end_src
rlm@15 170
rlm@18 171 Switching between contexts is not the normal operation of a Device,
rlm@18 172 and one of the problems with doing so is that a Device normally keeps
rlm@18 173 around a few pieces of state such as the =ClickRemoval= array above
rlm@18 174 which will become corrupted if the contexts are not done in
rlm@18 175 parallel. The solution is to create a copy of this normally global
rlm@18 176 device state for each context, and copy it back and forth into and out
rlm@18 177 of the actual device state whenever a context is rendered.
rlm@15 178
rlm@18 179 ** Synchronization Macros
rlm@15 180
rlm@18 181 #+begin_src C
rlm@15 182 //////////////////// Context Creation / Synchronization
rlm@15 183
rlm@15 184 #define _MAKE_SYNC(NAME, INIT_EXPR, GET_EXPR, SET_EXPR) \
rlm@15 185 void NAME (ALuint sourceID1, ALuint sourceID2, \
rlm@15 186 ALCcontext *ctx1, ALCcontext *ctx2, \
rlm@15 187 ALenum param){ \
rlm@15 188 INIT_EXPR; \
rlm@15 189 ALCcontext *current = alcGetCurrentContext(); \
rlm@15 190 alcMakeContextCurrent(ctx1); \
rlm@15 191 GET_EXPR; \
rlm@15 192 alcMakeContextCurrent(ctx2); \
rlm@15 193 SET_EXPR; \
rlm@15 194 alcMakeContextCurrent(current); \
rlm@15 195 }
rlm@15 196
rlm@15 197 #define MAKE_SYNC(NAME, TYPE, GET, SET) \
rlm@15 198 _MAKE_SYNC(NAME, \
rlm@15 199 TYPE value, \
rlm@15 200 GET(sourceID1, param, &value), \
rlm@15 201 SET(sourceID2, param, value))
rlm@15 202
rlm@15 203 #define MAKE_SYNC3(NAME, TYPE, GET, SET) \
rlm@15 204 _MAKE_SYNC(NAME, \
rlm@15 205 TYPE value1; TYPE value2; TYPE value3;, \
rlm@15 206 GET(sourceID1, param, &value1, &value2, &value3), \
rlm@15 207 SET(sourceID2, param, value1, value2, value3))
rlm@15 208
rlm@15 209 MAKE_SYNC( syncSourcei, ALint, alGetSourcei, alSourcei);
rlm@15 210 MAKE_SYNC( syncSourcef, ALfloat, alGetSourcef, alSourcef);
rlm@15 211 MAKE_SYNC3(syncSource3i, ALint, alGetSource3i, alSource3i);
rlm@15 212 MAKE_SYNC3(syncSource3f, ALfloat, alGetSource3f, alSource3f);
rlm@15 213
rlm@18 214 #+end_src
rlm@18 215
rlm@20 216 Setting the state of an =OpenAL= source is done with the =alSourcei=,
rlm@18 217 =alSourcef=, =alSource3i=, and =alSource3f= functions. In order to
rlm@20 218 completely synchronize two sources, it is necessary to use all of
rlm@18 219 them. These macros help to condense the otherwise repetitive
rlm@20 220 synchronization code involving these similar low-level =OpenAL= functions.
rlm@18 221
rlm@18 222 ** Source Synchronization
rlm@18 223 #+begin_src C
rlm@15 224 void syncSources(ALsource *masterSource, ALsource *slaveSource,
rlm@15 225 ALCcontext *masterCtx, ALCcontext *slaveCtx){
rlm@15 226 ALuint master = masterSource->source;
rlm@15 227 ALuint slave = slaveSource->source;
rlm@15 228 ALCcontext *current = alcGetCurrentContext();
rlm@15 229
rlm@15 230 syncSourcef(master,slave,masterCtx,slaveCtx,AL_PITCH);
rlm@15 231 syncSourcef(master,slave,masterCtx,slaveCtx,AL_GAIN);
rlm@15 232 syncSourcef(master,slave,masterCtx,slaveCtx,AL_MAX_DISTANCE);
rlm@15 233 syncSourcef(master,slave,masterCtx,slaveCtx,AL_ROLLOFF_FACTOR);
rlm@15 234 syncSourcef(master,slave,masterCtx,slaveCtx,AL_REFERENCE_DISTANCE);
rlm@15 235 syncSourcef(master,slave,masterCtx,slaveCtx,AL_MIN_GAIN);
rlm@15 236 syncSourcef(master,slave,masterCtx,slaveCtx,AL_MAX_GAIN);
rlm@15 237 syncSourcef(master,slave,masterCtx,slaveCtx,AL_CONE_OUTER_GAIN);
rlm@15 238 syncSourcef(master,slave,masterCtx,slaveCtx,AL_CONE_INNER_ANGLE);
rlm@15 239 syncSourcef(master,slave,masterCtx,slaveCtx,AL_CONE_OUTER_ANGLE);
rlm@15 240 syncSourcef(master,slave,masterCtx,slaveCtx,AL_SEC_OFFSET);
rlm@15 241 syncSourcef(master,slave,masterCtx,slaveCtx,AL_SAMPLE_OFFSET);
rlm@15 242 syncSourcef(master,slave,masterCtx,slaveCtx,AL_BYTE_OFFSET);
rlm@15 243
rlm@15 244 syncSource3f(master,slave,masterCtx,slaveCtx,AL_POSITION);
rlm@15 245 syncSource3f(master,slave,masterCtx,slaveCtx,AL_VELOCITY);
rlm@15 246 syncSource3f(master,slave,masterCtx,slaveCtx,AL_DIRECTION);
rlm@15 247
rlm@15 248 syncSourcei(master,slave,masterCtx,slaveCtx,AL_SOURCE_RELATIVE);
rlm@15 249 syncSourcei(master,slave,masterCtx,slaveCtx,AL_LOOPING);
rlm@15 250
rlm@15 251 alcMakeContextCurrent(masterCtx);
rlm@15 252 ALint source_type;
rlm@15 253 alGetSourcei(master, AL_SOURCE_TYPE, &source_type);
rlm@15 254
rlm@15 255 // Only static sources are currently synchronized!
rlm@15 256 if (AL_STATIC == source_type){
rlm@15 257 ALint master_buffer;
rlm@15 258 ALint slave_buffer;
rlm@15 259 alGetSourcei(master, AL_BUFFER, &master_buffer);
rlm@15 260 alcMakeContextCurrent(slaveCtx);
rlm@15 261 alGetSourcei(slave, AL_BUFFER, &slave_buffer);
rlm@15 262 if (master_buffer != slave_buffer){
rlm@15 263 alSourcei(slave, AL_BUFFER, master_buffer);
rlm@15 264 }
rlm@15 265 }
rlm@15 266
rlm@15 267 // Synchronize the state of the two sources.
rlm@15 268 alcMakeContextCurrent(masterCtx);
rlm@15 269 ALint masterState;
rlm@15 270 ALint slaveState;
rlm@15 271
rlm@15 272 alGetSourcei(master, AL_SOURCE_STATE, &masterState);
rlm@15 273 alcMakeContextCurrent(slaveCtx);
rlm@15 274 alGetSourcei(slave, AL_SOURCE_STATE, &slaveState);
rlm@15 275
rlm@15 276 if (masterState != slaveState){
rlm@15 277 switch (masterState){
rlm@15 278 case AL_INITIAL : alSourceRewind(slave); break;
rlm@15 279 case AL_PLAYING : alSourcePlay(slave); break;
rlm@15 280 case AL_PAUSED : alSourcePause(slave); break;
rlm@15 281 case AL_STOPPED : alSourceStop(slave); break;
rlm@15 282 }
rlm@15 283 }
rlm@15 284 // Restore whatever context was previously active.
rlm@15 285 alcMakeContextCurrent(current);
rlm@15 286 }
rlm@18 287 #+end_src
rlm@20 288 This function is long because it has to exhaustively go through all the
rlm@18 289 possible state that a source can have and make sure that it is the
rlm@18 290 same between the master and slave sources. I'd like to take this
rlm@18 291 moment to salute the [[http://connect.creativelabs.com/openal/Documentation/Forms/AllItems.aspx][=OpenAL= Reference Manual]], which provides a very
rlm@18 292 good description of =OpenAL='s internals.
rlm@15 293
rlm@18 294 ** Context Synchronization
rlm@18 295 #+begin_src C
rlm@15 296 void syncContexts(ALCcontext *master, ALCcontext *slave){
rlm@15 297 /* If there aren't sufficient sources in slave to mirror
rlm@15 298 the sources in master, create them. */
rlm@15 299 ALCcontext *current = alcGetCurrentContext();
rlm@15 300
rlm@15 301 UIntMap *masterSourceMap = &(master->SourceMap);
rlm@15 302 UIntMap *slaveSourceMap = &(slave->SourceMap);
rlm@15 303 ALuint numMasterSources = masterSourceMap->size;
rlm@15 304 ALuint numSlaveSources = slaveSourceMap->size;
rlm@15 305
rlm@15 306 alcMakeContextCurrent(slave);
rlm@15 307 if (numSlaveSources < numMasterSources){
rlm@15 308 ALuint numMissingSources = numMasterSources - numSlaveSources;
rlm@15 309 ALuint newSources[numMissingSources];
rlm@15 310 alGenSources(numMissingSources, newSources);
rlm@15 311 }
rlm@15 312
rlm@20 313 /* Now, slave is guaranteed to have at least as many sources
rlm@15 314 as master. Sync each source from master to the corresponding
rlm@15 315 source in slave. */
rlm@15 316 int i;
rlm@15 317 for(i = 0; i < masterSourceMap->size; i++){
rlm@15 318 syncSources((ALsource*)masterSourceMap->array[i].value,
rlm@15 319 (ALsource*)slaveSourceMap->array[i].value,
rlm@15 320 master, slave);
rlm@15 321 }
rlm@15 322 alcMakeContextCurrent(current);
rlm@15 323 }
rlm@18 324 #+end_src
rlm@15 325
rlm@18 326 Most of the hard work in Context Synchronization is done in
rlm@18 327 =syncSources()=. The only thing that =syncContexts()= has to worry
rlm@20 328 about is automatically creating new sources whenever a slave context
rlm@18 329 does not have the same number of sources as the master context.
rlm@18 330
rlm@19 331 ** Context Creation
rlm@18 332 #+begin_src C
rlm@15 333 static void addContext(ALCdevice *Device, ALCcontext *context){
rlm@15 334 send_data *data = (send_data*)Device->ExtraData;
rlm@15 335 // expand array if necessary
rlm@15 336 if (data->numContexts >= data->maxContexts){
rlm@15 337 ALuint newMaxContexts = data->maxContexts*2 + 1;
rlm@15 338 data->contexts = realloc(data->contexts, newMaxContexts*sizeof(context_data));
rlm@15 339 data->maxContexts = newMaxContexts;
rlm@15 340 }
rlm@15 341 // create context_data and add it to the main array
rlm@15 342 context_data *ctxData;
rlm@15 343 ctxData = (context_data*)calloc(1, sizeof(*ctxData));
rlm@15 344 ctxData->renderBuffer =
rlm@15 345 malloc(BytesFromDevFmt(Device->FmtType) *
rlm@15 346 Device->NumChan * Device->UpdateSize);
rlm@15 347 ctxData->ctx = context;
rlm@15 348
rlm@15 349 data->contexts[data->numContexts] = ctxData;
rlm@15 350 data->numContexts++;
rlm@15 351 }
rlm@18 352 #+end_src
rlm@15 353
rlm@18 354 Here, the slave context is created, and it's data is stored in the
rlm@18 355 device-wide =ExtraData= structure. The =renderBuffer= that is created
rlm@18 356 here is where the rendered sound samples for this slave context will
rlm@18 357 eventually go.
rlm@15 358
rlm@19 359 ** Context Switching
rlm@18 360 #+begin_src C
rlm@15 361 //////////////////// Context Switching
rlm@15 362
rlm@15 363 /* A device brings along with it two pieces of state
rlm@15 364 * which have to be swapped in and out with each context.
rlm@15 365 */
rlm@15 366 static void swapInContext(ALCdevice *Device, context_data *ctxData){
rlm@15 367 memcpy(Device->ClickRemoval, ctxData->ClickRemoval, sizeof(ALfloat)*MAXCHANNELS);
rlm@15 368 memcpy(Device->PendingClicks, ctxData->PendingClicks, sizeof(ALfloat)*MAXCHANNELS);
rlm@15 369 }
rlm@15 370
rlm@15 371 static void saveContext(ALCdevice *Device, context_data *ctxData){
rlm@15 372 memcpy(ctxData->ClickRemoval, Device->ClickRemoval, sizeof(ALfloat)*MAXCHANNELS);
rlm@15 373 memcpy(ctxData->PendingClicks, Device->PendingClicks, sizeof(ALfloat)*MAXCHANNELS);
rlm@15 374 }
rlm@15 375
rlm@15 376 static ALCcontext **currentContext;
rlm@15 377 static ALuint currentNumContext;
rlm@15 378
rlm@15 379 /* By default, all contexts are rendered at once for each call to aluMixData.
rlm@20 380 * This function uses the internals of the ALCdevice struct to temporally
rlm@15 381 * cause aluMixData to only render the chosen context.
rlm@15 382 */
rlm@15 383 static void limitContext(ALCdevice *Device, ALCcontext *ctx){
rlm@15 384 currentContext = Device->Contexts;
rlm@15 385 currentNumContext = Device->NumContexts;
rlm@15 386 Device->Contexts = &ctx;
rlm@15 387 Device->NumContexts = 1;
rlm@15 388 }
rlm@15 389
rlm@15 390 static void unLimitContext(ALCdevice *Device){
rlm@15 391 Device->Contexts = currentContext;
rlm@15 392 Device->NumContexts = currentNumContext;
rlm@15 393 }
rlm@18 394 #+end_src
rlm@15 395
rlm@20 396 =OpenAL= normally renders all Contexts in parallel, outputting the
rlm@18 397 whole result to the buffer. It does this by iterating over the
rlm@18 398 Device->Contexts array and rendering each context to the buffer in
rlm@20 399 turn. By temporally setting Device->NumContexts to 1 and adjusting
rlm@18 400 the Device's context list to put the desired context-to-be-rendered
rlm@18 401 into position 0, we can get trick =OpenAL= into rendering each slave
rlm@18 402 context separate from all the others.
rlm@15 403
rlm@18 404 ** Main Device Loop
rlm@18 405 #+begin_src C
rlm@15 406 //////////////////// Main Device Loop
rlm@15 407
rlm@18 408 /* Establish the LWJGL context as the master context, which will
rlm@15 409 * be synchronized to all the slave contexts
rlm@15 410 */
rlm@15 411 static void init(ALCdevice *Device){
rlm@15 412 ALCcontext *masterContext = alcGetCurrentContext();
rlm@15 413 addContext(Device, masterContext);
rlm@15 414 }
rlm@15 415
rlm@15 416
rlm@15 417 static void renderData(ALCdevice *Device, int samples){
rlm@15 418 if(!Device->Connected){return;}
rlm@15 419 send_data *data = (send_data*)Device->ExtraData;
rlm@15 420 ALCcontext *current = alcGetCurrentContext();
rlm@15 421
rlm@15 422 ALuint i;
rlm@15 423 for (i = 1; i < data->numContexts; i++){
rlm@15 424 syncContexts(data->contexts[0]->ctx , data->contexts[i]->ctx);
rlm@15 425 }
rlm@15 426
rlm@15 427 if ((uint) samples > Device->UpdateSize){
rlm@15 428 printf("exceeding internal buffer size; dropping samples\n");
rlm@15 429 printf("requested %d; available %d\n", samples, Device->UpdateSize);
rlm@15 430 samples = (int) Device->UpdateSize;
rlm@15 431 }
rlm@15 432
rlm@15 433 for (i = 0; i < data->numContexts; i++){
rlm@15 434 context_data *ctxData = data->contexts[i];
rlm@15 435 ALCcontext *ctx = ctxData->ctx;
rlm@15 436 alcMakeContextCurrent(ctx);
rlm@15 437 limitContext(Device, ctx);
rlm@15 438 swapInContext(Device, ctxData);
rlm@15 439 aluMixData(Device, ctxData->renderBuffer, samples);
rlm@15 440 saveContext(Device, ctxData);
rlm@15 441 unLimitContext(Device);
rlm@15 442 }
rlm@15 443 alcMakeContextCurrent(current);
rlm@15 444 }
rlm@18 445 #+end_src
rlm@15 446
rlm@18 447 The main loop synchronizes the master LWJGL context with all the slave
rlm@18 448 contexts, then walks each context, rendering just that context to it's
rlm@18 449 audio-sample storage buffer.
rlm@15 450
rlm@19 451 ** JNI Methods
rlm@19 452
rlm@19 453 At this point, we have the ability to create multiple listeners by
rlm@19 454 using the master/slave context trick, and the rendered audio data is
rlm@19 455 waiting patiently in internal buffers, one for each listener. We need
rlm@19 456 a way to transport this information to Java, and also a way to drive
rlm@19 457 this device from Java. The following JNI interface code is inspired
rlm@19 458 by the way LWJGL interfaces with =OpenAL=.
rlm@19 459
rlm@19 460 *** step
rlm@19 461
rlm@18 462 #+begin_src C
rlm@15 463 //////////////////// JNI Methods
rlm@15 464
rlm@15 465 #include "com_aurellem_send_AudioSend.h"
rlm@15 466
rlm@15 467 /*
rlm@15 468 * Class: com_aurellem_send_AudioSend
rlm@15 469 * Method: nstep
rlm@15 470 * Signature: (JI)V
rlm@15 471 */
rlm@15 472 JNIEXPORT void JNICALL Java_com_aurellem_send_AudioSend_nstep
rlm@15 473 (JNIEnv *env, jclass clazz, jlong device, jint samples){
rlm@15 474 UNUSED(env);UNUSED(clazz);UNUSED(device);
rlm@15 475 renderData((ALCdevice*)((intptr_t)device), samples);
rlm@15 476 }
rlm@19 477 #+end_src
rlm@19 478 This device, unlike most of the other devices in =OpenAL=, does not
rlm@19 479 render sound unless asked. This enables the system to slow down or
rlm@19 480 speed up depending on the needs of the AIs who are using it to
rlm@19 481 listen. If the device tried to render samples in real-time, a
rlm@19 482 complicated AI whose mind takes 100 seconds of computer time to
rlm@19 483 simulate 1 second of AI-time would miss almost all of the sound in
rlm@19 484 its environment.
rlm@15 485
rlm@19 486
rlm@19 487 *** getSamples
rlm@19 488 #+begin_src C
rlm@15 489 /*
rlm@15 490 * Class: com_aurellem_send_AudioSend
rlm@15 491 * Method: ngetSamples
rlm@15 492 * Signature: (JLjava/nio/ByteBuffer;III)V
rlm@15 493 */
rlm@15 494 JNIEXPORT void JNICALL Java_com_aurellem_send_AudioSend_ngetSamples
rlm@15 495 (JNIEnv *env, jclass clazz, jlong device, jobject buffer, jint position,
rlm@15 496 jint samples, jint n){
rlm@15 497 UNUSED(clazz);
rlm@15 498
rlm@15 499 ALvoid *buffer_address =
rlm@15 500 ((ALbyte *)(((char*)(*env)->GetDirectBufferAddress(env, buffer)) + position));
rlm@15 501 ALCdevice *recorder = (ALCdevice*) ((intptr_t)device);
rlm@15 502 send_data *data = (send_data*)recorder->ExtraData;
rlm@15 503 if ((ALuint)n > data->numContexts){return;}
rlm@15 504 memcpy(buffer_address, data->contexts[n]->renderBuffer,
rlm@15 505 BytesFromDevFmt(recorder->FmtType) * recorder->NumChan * samples);
rlm@15 506 }
rlm@19 507 #+end_src
rlm@15 508
rlm@19 509 This is the transport layer between C and Java that will eventually
rlm@19 510 allow us to access rendered sound data from clojure.
rlm@19 511
rlm@19 512 *** Listener Management
rlm@19 513
rlm@19 514 =addListener=, =setNthListenerf=, and =setNthListener3f= are
rlm@19 515 necessary to change the properties of any listener other than the
rlm@19 516 master one, since only the listener of the current active context is
rlm@19 517 affected by the normal =OpenAL= listener calls.
rlm@19 518
rlm@19 519 #+begin_src C
rlm@15 520 /*
rlm@15 521 * Class: com_aurellem_send_AudioSend
rlm@15 522 * Method: naddListener
rlm@15 523 * Signature: (J)V
rlm@15 524 */
rlm@15 525 JNIEXPORT void JNICALL Java_com_aurellem_send_AudioSend_naddListener
rlm@15 526 (JNIEnv *env, jclass clazz, jlong device){
rlm@15 527 UNUSED(env); UNUSED(clazz);
rlm@15 528 //printf("creating new context via naddListener\n");
rlm@15 529 ALCdevice *Device = (ALCdevice*) ((intptr_t)device);
rlm@15 530 ALCcontext *new = alcCreateContext(Device, NULL);
rlm@15 531 addContext(Device, new);
rlm@15 532 }
rlm@15 533
rlm@15 534 /*
rlm@15 535 * Class: com_aurellem_send_AudioSend
rlm@15 536 * Method: nsetNthListener3f
rlm@15 537 * Signature: (IFFFJI)V
rlm@15 538 */
rlm@15 539 JNIEXPORT void JNICALL Java_com_aurellem_send_AudioSend_nsetNthListener3f
rlm@15 540 (JNIEnv *env, jclass clazz, jint param,
rlm@15 541 jfloat v1, jfloat v2, jfloat v3, jlong device, jint contextNum){
rlm@15 542 UNUSED(env);UNUSED(clazz);
rlm@15 543
rlm@15 544 ALCdevice *Device = (ALCdevice*) ((intptr_t)device);
rlm@15 545 send_data *data = (send_data*)Device->ExtraData;
rlm@15 546
rlm@15 547 ALCcontext *current = alcGetCurrentContext();
rlm@15 548 if ((ALuint)contextNum > data->numContexts){return;}
rlm@15 549 alcMakeContextCurrent(data->contexts[contextNum]->ctx);
rlm@15 550 alListener3f(param, v1, v2, v3);
rlm@15 551 alcMakeContextCurrent(current);
rlm@15 552 }
rlm@15 553
rlm@15 554 /*
rlm@15 555 * Class: com_aurellem_send_AudioSend
rlm@15 556 * Method: nsetNthListenerf
rlm@15 557 * Signature: (IFJI)V
rlm@15 558 */
rlm@15 559 JNIEXPORT void JNICALL Java_com_aurellem_send_AudioSend_nsetNthListenerf
rlm@15 560 (JNIEnv *env, jclass clazz, jint param, jfloat v1, jlong device,
rlm@15 561 jint contextNum){
rlm@15 562
rlm@15 563 UNUSED(env);UNUSED(clazz);
rlm@15 564
rlm@15 565 ALCdevice *Device = (ALCdevice*) ((intptr_t)device);
rlm@15 566 send_data *data = (send_data*)Device->ExtraData;
rlm@15 567
rlm@15 568 ALCcontext *current = alcGetCurrentContext();
rlm@15 569 if ((ALuint)contextNum > data->numContexts){return;}
rlm@15 570 alcMakeContextCurrent(data->contexts[contextNum]->ctx);
rlm@15 571 alListenerf(param, v1);
rlm@15 572 alcMakeContextCurrent(current);
rlm@15 573 }
rlm@19 574 #+end_src
rlm@15 575
rlm@20 576 *** Initialization
rlm@19 577 =initDevice= is called from the Java side after LWJGL has created its
rlm@19 578 context, and before any calls to =addListener=. It establishes the
rlm@19 579 LWJGL context as the master context.
rlm@19 580
rlm@20 581 =getAudioFormat= is a convenience function that uses JNI to build up a
rlm@19 582 =javax.sound.sampled.AudioFormat= object from data in the Device. This
rlm@19 583 way, there is no ambiguity about what the bits created by =step= and
rlm@19 584 returned by =getSamples= mean.
rlm@19 585
rlm@19 586 #+begin_src C
rlm@15 587 /*
rlm@15 588 * Class: com_aurellem_send_AudioSend
rlm@15 589 * Method: ninitDevice
rlm@15 590 * Signature: (J)V
rlm@15 591 */
rlm@15 592 JNIEXPORT void JNICALL Java_com_aurellem_send_AudioSend_ninitDevice
rlm@15 593 (JNIEnv *env, jclass clazz, jlong device){
rlm@15 594 UNUSED(env);UNUSED(clazz);
rlm@15 595 ALCdevice *Device = (ALCdevice*) ((intptr_t)device);
rlm@15 596 init(Device);
rlm@15 597 }
rlm@15 598
rlm@15 599 /*
rlm@15 600 * Class: com_aurellem_send_AudioSend
rlm@15 601 * Method: ngetAudioFormat
rlm@15 602 * Signature: (J)Ljavax/sound/sampled/AudioFormat;
rlm@15 603 */
rlm@15 604 JNIEXPORT jobject JNICALL Java_com_aurellem_send_AudioSend_ngetAudioFormat
rlm@15 605 (JNIEnv *env, jclass clazz, jlong device){
rlm@15 606 UNUSED(clazz);
rlm@15 607 jclass AudioFormatClass =
rlm@15 608 (*env)->FindClass(env, "javax/sound/sampled/AudioFormat");
rlm@15 609 jmethodID AudioFormatConstructor =
rlm@15 610 (*env)->GetMethodID(env, AudioFormatClass, "<init>", "(FIIZZ)V");
rlm@15 611
rlm@15 612 ALCdevice *Device = (ALCdevice*) ((intptr_t)device);
rlm@15 613 int isSigned;
rlm@15 614 switch (Device->FmtType)
rlm@15 615 {
rlm@15 616 case DevFmtUByte:
rlm@15 617 case DevFmtUShort: isSigned = 0; break;
rlm@15 618 default : isSigned = 1;
rlm@15 619 }
rlm@15 620 float frequency = Device->Frequency;
rlm@15 621 int bitsPerFrame = (8 * BytesFromDevFmt(Device->FmtType));
rlm@15 622 int channels = Device->NumChan;
rlm@15 623 jobject format = (*env)->
rlm@15 624 NewObject(
rlm@15 625 env,AudioFormatClass,AudioFormatConstructor,
rlm@15 626 frequency,
rlm@15 627 bitsPerFrame,
rlm@15 628 channels,
rlm@15 629 isSigned,
rlm@15 630 0);
rlm@15 631 return format;
rlm@15 632 }
rlm@19 633 #+end_src
rlm@15 634
rlm@19 635 *** Boring Device management stuff
rlm@19 636 This code is more-or-less copied verbatim from the other =OpenAL=
rlm@19 637 backends. It's the basis for =OpenAL='s primitive object system.
rlm@19 638
rlm@19 639 #+begin_src C
rlm@20 640 //////////////////// Device Initialization / Management
rlm@15 641
rlm@15 642 static const ALCchar sendDevice[] = "Multiple Audio Send";
rlm@15 643
rlm@15 644 static ALCboolean send_open_playback(ALCdevice *device,
rlm@15 645 const ALCchar *deviceName)
rlm@15 646 {
rlm@15 647 send_data *data;
rlm@15 648 // stop any buffering for stdout, so that I can
rlm@20 649 // see the printf statements in my terminal immediately
rlm@15 650 setbuf(stdout, NULL);
rlm@15 651
rlm@15 652 if(!deviceName)
rlm@15 653 deviceName = sendDevice;
rlm@15 654 else if(strcmp(deviceName, sendDevice) != 0)
rlm@15 655 return ALC_FALSE;
rlm@15 656 data = (send_data*)calloc(1, sizeof(*data));
rlm@15 657 device->szDeviceName = strdup(deviceName);
rlm@15 658 device->ExtraData = data;
rlm@15 659 return ALC_TRUE;
rlm@15 660 }
rlm@15 661
rlm@15 662 static void send_close_playback(ALCdevice *device)
rlm@15 663 {
rlm@15 664 send_data *data = (send_data*)device->ExtraData;
rlm@15 665 alcMakeContextCurrent(NULL);
rlm@15 666 ALuint i;
rlm@15 667 // Destroy all slave contexts. LWJGL will take care of
rlm@15 668 // its own context.
rlm@15 669 for (i = 1; i < data->numContexts; i++){
rlm@15 670 context_data *ctxData = data->contexts[i];
rlm@15 671 alcDestroyContext(ctxData->ctx);
rlm@15 672 free(ctxData->renderBuffer);
rlm@15 673 free(ctxData);
rlm@15 674 }
rlm@15 675 free(data);
rlm@15 676 device->ExtraData = NULL;
rlm@15 677 }
rlm@15 678
rlm@15 679 static ALCboolean send_reset_playback(ALCdevice *device)
rlm@15 680 {
rlm@15 681 SetDefaultWFXChannelOrder(device);
rlm@15 682 return ALC_TRUE;
rlm@15 683 }
rlm@15 684
rlm@15 685 static void send_stop_playback(ALCdevice *Device){
rlm@15 686 UNUSED(Device);
rlm@15 687 }
rlm@15 688
rlm@15 689 static const BackendFuncs send_funcs = {
rlm@15 690 send_open_playback,
rlm@15 691 send_close_playback,
rlm@15 692 send_reset_playback,
rlm@15 693 send_stop_playback,
rlm@15 694 NULL,
rlm@15 695 NULL, /* These would be filled with functions to */
rlm@15 696 NULL, /* handle capturing audio if we we into that */
rlm@15 697 NULL, /* sort of thing... */
rlm@15 698 NULL,
rlm@15 699 NULL
rlm@15 700 };
rlm@15 701
rlm@15 702 ALCboolean alc_send_init(BackendFuncs *func_list){
rlm@15 703 *func_list = send_funcs;
rlm@15 704 return ALC_TRUE;
rlm@15 705 }
rlm@15 706
rlm@15 707 void alc_send_deinit(void){}
rlm@15 708
rlm@15 709 void alc_send_probe(enum DevProbe type)
rlm@15 710 {
rlm@15 711 switch(type)
rlm@15 712 {
rlm@15 713 case DEVICE_PROBE:
rlm@15 714 AppendDeviceList(sendDevice);
rlm@15 715 break;
rlm@15 716 case ALL_DEVICE_PROBE:
rlm@15 717 AppendAllDeviceList(sendDevice);
rlm@15 718 break;
rlm@15 719 case CAPTURE_DEVICE_PROBE:
rlm@15 720 break;
rlm@15 721 }
rlm@15 722 }
rlm@15 723 #+end_src
rlm@15 724
rlm@19 725 * The Java interface, =AudioSend=
rlm@15 726
rlm@19 727 The Java interface to the Send Device follows naturally from the JNI
rlm@19 728 definitions. It is included here for completeness. The only thing here
rlm@19 729 of note is the =deviceID=. This is available from LWJGL, but to only
rlm@20 730 way to get it is reflection. Unfortunately, there is no other way to
rlm@19 731 control the Send device than to obtain a pointer to it.
rlm@15 732
rlm@19 733 #+include: "../java/src/com/aurellem/send/AudioSend.java" src java :exports code
rlm@15 734
rlm@19 735 * Finally, Ears in clojure!
rlm@15 736
rlm@20 737 Now that the infrastructure is complete (modulo a few patches to
rlm@19 738 jMonkeyEngine3 to support accessing this modified version of =OpenAL=
rlm@19 739 that are not worth discussing), the clojure ear abstraction is rather
rlm@19 740 simple. Just as there were =SceneProcessors= for vision, there are
rlm@19 741 now =SoundProcessors= for hearing.
rlm@15 742
rlm@19 743 #+include "../../jmeCapture/src/com/aurellem/capture/audio/SoundProcessor.java" src java
rlm@15 744
rlm@15 745 #+srcname: ears
rlm@0 746 #+begin_src clojure
rlm@19 747 (ns cortex.hearing
rlm@19 748 "Simulate the sense of hearing in jMonkeyEngine3. Enables multiple
rlm@19 749 listeners at different positions in the same world. Passes vectors
rlm@20 750 of floats in the range [-1.0 -- 1.0] in PCM format to any arbitrary
rlm@19 751 function."
rlm@19 752 {:author "Robert McIntyre"}
rlm@19 753 (:use (cortex world util))
rlm@19 754 (:import java.nio.ByteBuffer)
rlm@19 755 (:import org.tritonus.share.sampled.FloatSampleTools)
rlm@19 756 (:import com.aurellem.capture.audio.SoundProcessor)
rlm@19 757 (:import javax.sound.sampled.AudioFormat))
rlm@19 758
rlm@0 759 (defn sound-processor
rlm@19 760 "Deals with converting ByteBuffers into Vectors of floats so that
rlm@19 761 the continuation functions can be defined in terms of immutable
rlm@19 762 stuff."
rlm@0 763 [continuation]
rlm@0 764 (proxy [SoundProcessor] []
rlm@0 765 (cleanup [])
rlm@0 766 (process
rlm@19 767 [#^ByteBuffer audioSamples numSamples #^AudioFormat audioFormat]
rlm@19 768 (let [bytes (byte-array numSamples)
rlm@19 769 floats (float-array numSamples)]
rlm@19 770 (.get audioSamples bytes 0 numSamples)
rlm@19 771 (FloatSampleTools/byte2floatInterleaved
rlm@19 772 bytes 0 floats 0
rlm@19 773 (/ numSamples (.getFrameSize audioFormat)) audioFormat)
rlm@19 774 (continuation
rlm@19 775 (vec floats))))))
rlm@0 776
rlm@0 777 (defn add-ear
rlm@19 778 "Add an ear to the world. The continuation function will be called
rlm@0 779 on the FFT or the sounds which the ear hears in the given
rlm@0 780 timeframe. Sound is 3D."
rlm@0 781 [world listener continuation]
rlm@0 782 (let [renderer (.getAudioRenderer world)]
rlm@0 783 (.addListener renderer listener)
rlm@0 784 (.registerSoundProcessor renderer listener
rlm@0 785 (sound-processor continuation))
rlm@0 786 listener))
rlm@0 787 #+end_src
rlm@0 788
rlm@19 789 * Example
rlm@0 790
rlm@0 791 #+srcname: test-hearing
rlm@0 792 #+begin_src clojure :results silent
rlm@19 793 (ns test.hearing
rlm@19 794 (:use (cortex world util hearing))
rlm@19 795 (:import (com.jme3.audio AudioNode Listener))
rlm@19 796 (:import com.jme3.scene.Node))
rlm@0 797
rlm@0 798 (defn setup-fn [world]
rlm@0 799 (let [listener (Listener.)]
rlm@19 800 (add-ear world listener #(println-repl (nth % 0)))))
rlm@0 801
rlm@0 802 (defn play-sound [node world value]
rlm@0 803 (if (not value)
rlm@0 804 (do
rlm@0 805 (.playSource (.getAudioRenderer world) node))))
rlm@0 806
rlm@19 807 (defn test-basic-hearing []
rlm@19 808 (.start
rlm@19 809 (let [node1 (AudioNode. (asset-manager) "Sounds/pure.wav" false false)]
rlm@19 810 (world
rlm@19 811 (Node.)
rlm@19 812 {"key-space" (partial play-sound node1)}
rlm@19 813 setup-fn
rlm@19 814 no-op))))
rlm@0 815 #+end_src
rlm@0 816
rlm@19 817 This extremely basic program prints out the first sample it encounters
rlm@19 818 at every time stamp. You can see the rendered sound begin printed at
rlm@19 819 the REPL.
rlm@15 820
rlm@0 821 * COMMENT Code Generation
rlm@0 822
rlm@15 823 #+begin_src clojure :tangle ../../cortex/src/cortex/hearing.clj
rlm@15 824 <<ears>>
rlm@0 825 #+end_src
rlm@0 826
rlm@15 827 #+begin_src clojure :tangle ../../cortex/src/test/hearing.clj
rlm@0 828 <<test-hearing>>
rlm@0 829 #+end_src
rlm@0 830
rlm@0 831
rlm@15 832 #+begin_src C :tangle ../Alc/backends/send.c
rlm@15 833 <<send>>
rlm@15 834 #+end_src
rlm@19 835
rlm@19 836