1 /*! \mainpage The RtAudio Tutorial
3 <CENTER>\ref intro \ref changes \ref download \ref start \ref error \ref probing \ref settings \ref playbackb \ref playbackc \ref recording \ref duplex \ref multi \ref methods \ref compiling \ref debug \ref apinotes \ref wishlist \ref acknowledge \ref license</CENTER>
5 \section intro Introduction
7 RtAudio is a set of C++ classes which provide a common API (Application Programming Interface) for realtime audio input/output across Linux (native ALSA, JACK, and OSS), Macintosh OS X, SGI, and Windows (DirectSound and ASIO) operating systems. RtAudio significantly simplifies the process of interacting with computer audio hardware. It was designed with the following goals:
10 <LI>object oriented C++ design</LI>
11 <LI>simple, common API across all supported platforms</LI>
12 <LI>only two header files and one source file for easy inclusion in programming projects</LI>
13 <LI>allow simultaneous multi-api support</LI>
14 <LI>blocking functionality</LI>
15 <LI>callback functionality</LI>
16 <LI>extensive audio device parameter control</LI>
17 <LI>audio device capability probing</LI>
18 <LI>automatic internal conversion for data format, channel number compensation, de-interleaving, and byte-swapping</LI>
21 RtAudio incorporates the concept of audio streams, which represent audio output (playback) and/or input (recording). Available audio devices and their capabilities can be enumerated and then specified when opening a stream. Where applicable, multiple API support can be compiled and a particular API specified when creating an RtAudio instance. See the \ref apinotes section for information specific to each of the supported audio APIs.
23 The RtAudio API provides both blocking (synchronous) and callback (asynchronous) functionality. Callbacks are typically used in conjunction with graphical user interfaces (GUI). Blocking functionality is often necessary for explicit control of multiple input/output stream synchronization or when audio must be synchronized with other system events.
25 \section changes What's New (Version 3.0)
27 RtAudio now allows simultaneous multi-api support. For example, you can compile RtAudio to provide both DirectSound and ASIO support on Windows platforms or ALSA, JACK, and OSS support on Linux platforms. This was accomplished by creating an abstract base class, RtApi, with subclasses for each supported API (RtApiAlsa, RtApiJack, RtApiOss, RtApiDs, RtApiAsio, RtApiCore, and RtApiAl). The class RtAudio is now a "controller" which creates an instance of an RtApi subclass based on the user's API choice via an optional RtAudio::RtAudioApi instantiation argument. If no API is specified, RtAudio attempts to make a "logical" API selection.
29 Support for the JACK low-latency audio server has been added with this version of RtAudio. It is necessary to have the JACK server running before creating an instance of RtAudio.
31 Several API changes have been made in version 3.0 of RtAudio in an effort to provide more consistent behavior across all supported audio APIs. The most significant of these changes is that multiple stream support from a single RtAudio instance has been discontinued. As a result, stream identifier input arguments are no longer required. Also, the RtAudio::streamWillBlock() function was poorly supported by most APIs and has been deprecated (though the function still exists in those subclasses of RtApi that do allow it to be implemented).
33 The RtAudio::getDeviceInfo() function was modified to return a globally defined RtAudioDeviceInfo structure. This structure is a simplified version of the previous RTAUDIO_DEVICE structure. In addition, the RTAUDIO_FORMAT structure was renamed RtAudioFormat and defined globally within RtAudio.h. These changes were made for clarity and to better conform with standard C++ programming practices.
35 The RtError class declaration and definition have been extracted to a separate file (RtError.h). This was done in preparation for a new release of the RtMidi class (planned for Summer 2004).
37 \section download Download
39 Latest Release (18 November 2005): <A href="http://music.mcgill.ca/~gary/rtaudio/release/rtaudio-3.0.3.tar.gz">Version 3.0.3</A>
41 \section start Getting Started
43 With version 3.0, it is now possible to compile multiple API support on a given platform and to specify an API choice during class instantiation. In the examples that follow, no API will be specified (in which case, RtAudio attempts to select the most "logical" available API).
45 The first thing that must be done when using RtAudio is to create an instance of the class. The default constructor scans the underlying audio system to verify that at least one device is available. RtAudio often uses C++ exceptions to report errors, necessitating try/catch blocks around most member functions. The following code example demonstrates default object construction and destruction:
55 // Default RtAudio constructor
57 audio = new RtAudio();
59 catch (RtError &error) {
60 // Handle the exception here
69 Obviously, this example doesn't demonstrate any of the real functionality of RtAudio. However, all uses of RtAudio must begin with a constructor (either default or overloaded varieties) and must end with class destruction. Further, it is necessary that all class methods that can throw a C++ exception be called within a try/catch block.
72 \section error Error Handling
74 RtAudio uses a C++ exception handler called RtError, which is declared and defined in RtError.h. The RtError class is quite simple but it does allow errors to be "caught" by RtError::Type. Almost all RtAudio methods can "throw" an RtError, most typically if a driver error occurs or a stream function is called when no stream is open. There are a number of cases within RtAudio where warning messages may be displayed but an exception is not thrown. There is a protected RtAudio method, error(), that can be modified to globally control how these messages are handled and reported. By default, error messages are not automatically displayed in RtAudio unless the preprocessor definition __RTAUDIO_DEBUG__ is defined. Messages associated with caught exceptions can be displayed with, for example, the RtError::printMessage() function.
77 \section probing Probing Device Capabilities
79 A programmer may wish to query the available audio device capabilities before deciding which to use. The following example outlines how this can be done.
92 // Default RtAudio constructor
94 audio = new RtAudio();
96 catch (RtError &error) {
101 // Determine the number of devices available
102 int devices = audio->getDeviceCount();
104 // Scan through devices for various capabilities
105 RtAudioDeviceInfo info;
106 for (int i=1; i<=devices; i++) {
109 info = audio->getDeviceInfo(i);
111 catch (RtError &error) {
112 error.printMessage();
116 // Print, for example, the maximum number of output channels for each device
117 std::cout << "device = " << i;
118 std::cout << ": maximum output channels = " << info.outputChannels << "\n";
128 The RtAudioDeviceInfo structure is defined in RtAudio.h and provides a variety of information useful in assessing the capabilities of a device:
131 typedef struct RtAudioDeviceInfo{
132 std::string name; // Character string device identifier.
133 bool probed; // true if the device capabilities were successfully probed.
134 int outputChannels; // Maximum output channels supported by device.
135 int inputChannels; // Maximum input channels supported by device.
136 int duplexChannels; // Maximum simultaneous input/output channels supported by device.
137 bool isDefault; // true if this is the default output or input device.
138 std::vector<int> sampleRates; // Supported sample rates.
139 RtAudioFormat nativeFormats; // Bit mask of supported data formats.
143 The following data formats are defined and fully supported by RtAudio:
146 typedef unsigned long RtAudioFormat;
147 static const RtAudioFormat RTAUDIO_SINT8; // Signed 8-bit integer
148 static const RtAudioFormat RTAUDIO_SINT16; // Signed 16-bit integer
149 static const RtAudioFormat RTAUDIO_SINT24; // Signed 24-bit integer (upper 3 bytes of 32-bit signed integer.)
150 static const RtAudioFormat RTAUDIO_SINT32; // Signed 32-bit integer
151 static const RtAudioFormat RTAUDIO_FLOAT32; // 32-bit float normalized between +/- 1.0
152 static const RtAudioFormat RTAUDIO_FLOAT64; // 64-bit double normalized between +/- 1.0
155 The <I>nativeFormats</I> member of the RtAudioDeviceInfo structure is a bit mask of the above formats that are natively supported by the device. However, RtAudio will automatically provide format conversion if a particular format is not natively supported. When the <I>probed</I> member of the RtAudioDeviceInfo structure is false, the remaining structure members are undefined and the device is probably unuseable.
157 While some audio devices may require a minimum channel value greater than one, RtAudio will provide automatic channel number compensation when the number of channels set by the user is less than that required by the device. Channel compensation is <I>NOT</I> possible when the number of channels set by the user is greater than that supported by the device.
159 It should be noted that the capabilities reported by a device driver or underlying audio API are not always accurate and/or may be dependent on a combination of device settings. For this reason, RtAudio does not typically rely on the queried values when attempting to open a stream.
162 \section settings Device Settings
164 The next step in using RtAudio is to open a stream with particular device and parameter settings.
173 int sampleRate = 44100;
174 int bufferSize = 256; // 256 sample frames
175 int nBuffers = 4; // number of internal buffers used by device
176 int device = 0; // 0 indicates the default or first available device
179 // Instantiate RtAudio and open a stream within a try/catch block
181 audio = new RtAudio();
183 catch (RtError &error) {
184 error.printMessage();
189 audio->openStream(device, channels, 0, 0, RTAUDIO_FLOAT32,
190 sampleRate, &bufferSize, nBuffers);
192 catch (RtError &error) {
193 error.printMessage();
194 // Perhaps try other parameters?
204 The RtAudio::openStream() method attempts to open a stream with a specified set of parameter values. In this case, we attempt to open a two channel playback stream with the default output device, 32-bit floating point data, a sample rate of 44100 Hz, a frame rate of 256 sample frames per read/write, and 4 internal device buffers. When device = 0, RtAudio first attempts to open the default audio device with the given parameters. If that attempt fails, RtAudio searches through the remaining available devices in an effort to find a device that will meet the given parameters. If all attempts are unsuccessful, an RtError is thrown. When a non-zero device value is specified, an attempt is made to open that device \e ONLY (device = 1 specifies the first identified device, as reported by RtAudio::getDeviceInfo()).
206 RtAudio provides four signed integer and two floating point data formats that can be specified using the RtAudioFormat parameter values mentioned earlier. If the opened device does not natively support the given format, RtAudio will automatically perform the necessary data format conversion.
208 The <I>bufferSize</I> parameter specifies the desired number of sample frames that will be written to and/or read from a device per write/read operation. The <I>nBuffers</I> parameter is used in setting the underlying device buffer parameters. Both the <I>bufferSize</I> and <I>nBuffers</I> parameters can be used to control stream latency though there is no guarantee that the passed values will be those used by a device (the <I>nBuffers</I> parameter is ignored when using the OS X CoreAudio, Linux Jack, and the Windows ASIO APIs). In general, lower values for both parameters will produce less latency but perhaps less robust performance. Both parameters can be specified with values of zero, in which case the smallest allowable values will be used. The <I>bufferSize</I> parameter is passed as a pointer and the actual value used by the stream is set during the device setup procedure. <I>bufferSize</I> values should be a power of two. Optimal and allowable buffer values tend to vary between systems and devices. Check the \ref apinotes section for general guidelines.
210 As noted earlier, the device capabilities reported by a driver or underlying audio API are not always accurate and/or may be dependent on a combination of device settings. Because of this, RtAudio does not attempt to query a device's capabilities or use previously reported values when opening a device. Instead, RtAudio simply attempts to set the given parameters on a specified device and then checks whether the setup is successful or not.
213 \section playbackb Playback (blocking functionality)
215 Once the device is open for playback, there are only a few final steps necessary for realtime audio output. We'll first provide an example (blocking functionality) and then discuss the details.
226 int sampleRate = 44100;
227 int bufferSize = 256; // 256 sample frames
228 int nBuffers = 4; // number of internal buffers used by device
230 int device = 0; // 0 indicates the default or first available device
233 // Open a stream during RtAudio instantiation
235 audio = new RtAudio(device, channels, 0, 0, RTAUDIO_FLOAT32,
236 sampleRate, &bufferSize, nBuffers);
238 catch (RtError &error) {
239 error.printMessage();
244 // Get a pointer to the stream buffer
245 buffer = (float *) audio->getStreamBuffer();
248 audio->startStream();
250 catch (RtError &error) {
251 error.printMessage();
255 // An example loop that runs for 40000 sample frames
257 while (count < 40000) {
258 // Generate your samples and fill the buffer with bufferSize sample frames of data
261 // Trigger the output of the data buffer
265 catch (RtError &error) {
266 error.printMessage();
274 // Stop and close the stream
276 audio->closeStream();
278 catch (RtError &error) {
279 error.printMessage();
289 The first thing to notice in this example is that we attempt to open a stream during class instantiation with an overloaded constructor. This constructor simply combines the functionality of the default constructor, used earlier, and the RtAudio::openStream() method. Again, we have specified a device value of 0, indicating that the default or first available device meeting the given parameters should be used. An attempt is made to open the stream with the specified <I>bufferSize</I> value. However, it is possible that the device will not accept this value, in which case the closest allowable size is used and returned via the pointer value. The constructor can fail if no available devices are found, or a memory allocation or device driver error occurs. Note that you should not call the RtAudio destructor if an exception is thrown during instantiation.
291 Assuming the constructor is successful, it is necessary to get a pointer to the buffer, provided by RtAudio, for use in feeding data to/from the opened stream. Note that the user should <I>NOT</I> attempt to deallocate the stream buffer memory ... memory management for the stream buffer will be automatically controlled by RtAudio. After starting the stream with RtAudio::startStream(), one simply fills that buffer, which is of length equal to the returned <I>bufferSize</I> value, with interleaved audio data (in the specified format) for playback. Finally, a call to the RtAudio::tickStream() routine triggers a blocking write call for the stream.
293 In general, one should call the RtAudio::stopStream() and RtAudio::closeStream() methods after finishing with a stream. However, both methods will implicitly be called during object destruction if necessary.
296 \section playbackc Playback (callback functionality)
298 The primary difference in using RtAudio with callback functionality involves the creation of a user-defined callback function. Here is an example that produces a sawtooth waveform for playback.
305 // Two-channel sawtooth wave generator.
306 int sawtooth(char *buffer, int bufferSize, void *data)
309 double *my_buffer = (double *) buffer;
310 double *my_data = (double *) data;
312 // Write interleaved audio data.
313 for (i=0; i<bufferSize; i++) {
314 for (j=0; j<2; j++) {
315 *my_buffer++ = my_data[j];
317 my_data[j] += 0.005 * (j+1+(j*0.1));
318 if (my_data[j] >= 1.0) my_data[j] -= 2.0;
328 int sampleRate = 44100;
329 int bufferSize = 256; // 256 sample frames
330 int nBuffers = 4; // number of internal buffers used by device
331 int device = 0; // 0 indicates the default or first available device
336 // Open a stream during RtAudio instantiation
338 audio = new RtAudio(device, channels, 0, 0, RTAUDIO_FLOAT64,
339 sampleRate, &bufferSize, nBuffers);
341 catch (RtError &error) {
342 error.printMessage();
347 // Set the stream callback function
348 audio->setStreamCallback(&sawtooth, (void *)data);
351 audio->startStream();
353 catch (RtError &error) {
354 error.printMessage();
358 std::cout << "\nPlaying ... press <enter> to quit.\n";
362 // Stop and close the stream
364 audio->closeStream();
366 catch (RtError &error) {
367 error.printMessage();
377 After opening the device in exactly the same way as the previous example (except with a data format change), we must set our callback function for the stream using RtAudio::setStreamCallback(). When the underlying audio API uses blocking calls (OSS, ALSA, SGI, and Windows DirectSound), this method will spawn a new process (or thread) that automatically calls the callback function when more data is needed. Callback-based audio APIs (OS X CoreAudio Linux Jack, and ASIO) implement their own event notification schemes. Note that the callback function is called only when the stream is "running" (between calls to the RtAudio::startStream() and RtAudio::stopStream() methods). The last argument to RtAudio::setStreamCallback() is a pointer to arbitrary data that you wish to access from within your callback function.
379 In this example, we stop the stream with an explicit call to RtAudio::stopStream(). When using callback functionality, it is also possible to stop a stream by returning a non-zero value from the callback function.
381 Once set with RtAudio::setStreamCallback, the callback process exists for the life of the stream (until the stream is closed with RtAudio::closeStream() or the RtAudio instance is deleted). It is possible to disassociate a callback function and cancel its process for an open stream using the RtAudio::cancelStreamCallback() method. The stream can then be used with blocking functionality or a new callback can be associated with it.
384 \section recording Recording
386 Using RtAudio for audio input is almost identical to the way it is used for playback. Here's the blocking playback example rewritten for recording:
397 int sampleRate = 44100;
398 int bufferSize = 256; // 256 sample frames
399 int nBuffers = 4; // number of internal buffers used by device
401 int device = 0; // 0 indicates the default or first available device
404 // Instantiate RtAudio and open a stream.
406 audio = new RtAudio(&stream, 0, 0, device, channels,
407 RTAUDIO_FLOAT32, sampleRate, &bufferSize, nBuffers);
409 catch (RtError &error) {
410 error.printMessage();
415 // Get a pointer to the stream buffer
416 buffer = (float *) audio->getStreamBuffer();
419 audio->startStream();
421 catch (RtError &error) {
422 error.printMessage();
426 // An example loop that runs for about 40000 sample frames
428 while (count < 40000) {
430 // Read a buffer of data
434 catch (RtError &error) {
435 error.printMessage();
439 // Process the input samples (bufferSize sample frames) that were read
449 catch (RtError &error) {
450 error.printMessage();
460 In this example, the stream was opened for recording with a non-zero <I>inputChannels</I> value. The only other difference between this example and that for playback involves the order of data processing in the loop, where it is necessary to first read a buffer of input data before manipulating it.
463 \section duplex Duplex Mode
465 Finally, it is easy to use RtAudio for simultaneous audio input/output, or duplex operation. In this example, we use a callback function and simply scale the input data before sending it back to the output.
473 // Pass-through function.
474 int scale(char *buffer, int bufferSize, void *)
476 // Note: do nothing here for pass through.
477 double *my_buffer = (double *) buffer;
479 // Scale input data for output.
480 for (int i=0; i<bufferSize; i++) {
481 // Do for two channels.
492 int sampleRate = 44100;
493 int bufferSize = 256; // 256 sample frames
494 int nBuffers = 4; // number of internal buffers used by device
495 int device = 0; // 0 indicates the default or first available device
499 // Open a stream during RtAudio instantiation
501 audio = new RtAudio(device, channels, device, channels, RTAUDIO_FLOAT64,
502 sampleRate, &bufferSize, nBuffers);
504 catch (RtError &error) {
505 error.printMessage();
510 // Set the stream callback function
511 audio->setStreamCallback(&scale, NULL);
514 audio->startStream();
516 catch (RtError &error) {
517 error.printMessage();
521 std::cout << "\nRunning duplex ... press <enter> to quit.\n";
525 // Stop and close the stream
527 audio->closeStream();
529 catch (RtError &error) {
530 error.printMessage();
540 When an RtAudio stream is running in duplex mode (nonzero input <I>AND</I> output channels), the audio write (playback) operation always occurs before the audio read (record) operation. This sequence allows the use of a single buffer to store both output and input data.
542 As we see with this example, the write-read sequence of operations does not preclude the use of RtAudio in situations where input data is first processed and then output through a duplex stream. When the stream buffer is first allocated, it is initialized with zeros, which produces no audible result when output to the device. In this example, anything recorded by the audio stream input will be scaled and played out during the next round of audio processing.
544 Note that duplex operation can also be achieved by opening one output stream instance and one input stream instance using the same or different devices. However, there may be timing problems when attempting to use two different devices, due to possible device clock variations, unless a common external "sync" is provided. This becomes even more difficult to achieve using two separate callback streams because it is not possible to <I>explicitly</I> control the calling order of the callback functions.
547 \section multi Using Simultaneous Multiple APIs
549 Because support for each audio API is encapsulated in a specific RtApi subclass, it is possible to compile and instantiate multiple API-specific subclasses on a given operating system. For example, one can compile both the RtApiDs and RtApiAsio classes on Windows operating systems by providing the appropriate preprocessor definitions, include files, and libraries for each. In a run-time situation, one might first attempt to determine whether any ASIO device drivers exist. This can be done by specifying the api argument RtAudio::WINDOWS_ASIO when attempting to create an instance of RtAudio. If an RtError is thrown (indicating no available drivers), then an instance of RtAudio with the api argument RtAudio::WINDOWS_DS can be created. Alternately, if no api argument is specified, RtAudio will first look for ASIO drivers and then DirectSound drivers (on Linux systems, the default API search order is Jack, Alsa, and finally OSS). In theory, it should also be possible to have separate instances of RtAudio open at the same time with different underlying audio API support, though this has not been tested. It is difficult to know how well different audio APIs can simultaneously coexist on a given operating system. In particular, it is most unlikely that the same device could be simultaneously controlled with two different audio APIs.
552 \section methods Summary of Methods
554 The following is a short summary of public methods (not including constructors and the destructor) provided by RtAudio:
557 <LI>RtAudio::openStream(): opens a stream with the specified parameters.</LI>
558 <LI>RtAudio::setStreamCallback(): sets a user-defined callback function for the stream.</LI>
559 <LI>RtAudio::cancelStreamCallback(): cancels a callback process and function for the stream.</LI>
560 <LI>RtAudio::getDeviceCount(): returns the number of audio devices available.</LI>
561 <LI>RtAudio::getDeviceInfo(): returns an RtAudioDeviceInfo structure for a specified device.</LI>
562 <LI>RtAudio::getStreamBuffer(): returns a pointer to the stream buffer.</LI>
563 <LI>RtAudio::tickStream(): triggers processing of input/output data for the stream (blocking).</LI>
564 <LI>RtAudio::closeStream(): closes the stream (implicitly called during object destruction).</LI>
565 <LI>RtAudio::startStream(): (re)starts the stream, typically after it has been stopped with either stopStream() or abortStream() or after first opening the stream.</LI>
566 <LI>RtAudio::stopStream(): stops the stream, allowing any remaining samples in the queue to be played out and/or read in. This does not implicitly call RtAudio::closeStream().</LI>
567 <LI>RtAudio::abortStream(): stops the stream, discarding any remaining samples in the queue. This does not implicitly call closeStream().</LI>
571 \section compiling Compiling
573 In order to compile RtAudio for a specific OS and audio API, it is necessary to supply the appropriate preprocessor definition and library within the compiler statement:
576 <TABLE BORDER=2 COLS=5 WIDTH="100%">
578 <TD WIDTH="5%"><B>OS:</B></TD>
579 <TD WIDTH="5%"><B>Audio API:</B></TD>
580 <TD WIDTH="5%"><B>C++ Class:</B></TD>
581 <TD WIDTH="5%"><B>Preprocessor Definition:</B></TD>
582 <TD WIDTH="5%"><B>Library or Framework:</B></TD>
583 <TD><B>Example Compiler Statement:</B></TD>
589 <TD>__LINUX_ALSA__</TD>
590 <TD><TT>asound, pthread</TT></TD>
591 <TD><TT>g++ -Wall -D__LINUX_ALSA__ -o probe probe.cpp RtAudio.cpp -lasound -lpthread</TT></TD>
595 <TD>Jack Audio Server</TD>
597 <TD>__LINUX_JACK__</TD>
598 <TD><TT>jack, pthread</TT></TD>
599 <TD><TT>g++ -Wall -D__LINUX_JACK__ -o probe probe.cpp RtAudio.cpp `pkg-config --cflags --libs jack` -lpthread</TT></TD>
605 <TD>__LINUX_OSS__</TD>
606 <TD><TT>pthread</TT></TD>
607 <TD><TT>g++ -Wall -D__LINUX_OSS__ -o probe probe.cpp RtAudio.cpp -lpthread</TT></TD>
610 <TD>Macintosh OS X</TD>
613 <TD>__MACOSX_CORE__</TD>
614 <TD><TT>pthread, stdc++, CoreAudio</TT></TD>
615 <TD><TT>g++ -Wall -D__MACOSX_CORE__ -o probe probe.cpp RtAudio.cpp -framework CoreAudio -lpthread</TT></TD>
622 <TD><TT>audio, pthread</TT></TD>
623 <TD><TT>CC -Wall -D__IRIX_AL__ -o probe probe.cpp RtAudio.cpp -laudio -lpthread</TT></TD>
627 <TD>Direct Sound</TD>
629 <TD>__WINDOWS_DS__</TD>
630 <TD><TT>dsound.lib (ver. 5.0 or higher), multithreaded</TT></TD>
631 <TD><I>compiler specific</I></TD>
637 <TD>__WINDOWS_ASIO__</TD>
638 <TD><I>various ASIO header and source files</I></TD>
639 <TD><I>compiler specific</I></TD>
644 The example compiler statements above could be used to compile the <TT>probe.cpp</TT> example file, assuming that <TT>probe.cpp</TT>, <TT>RtAudio.h</TT>, <tt>RtError.h</tt>, and <TT>RtAudio.cpp</TT> all exist in the same directory.
646 \section debug Debugging
648 If you are having problems getting RtAudio to run on your system, try passing the preprocessor definition <TT>__RTAUDIO_DEBUG__</TT> to the compiler (or uncomment the definition at the bottom of RtAudio.h). A variety of warning messages will be displayed that may help in determining the problem. Also try using the programs included in the <tt>test</tt> directory. The program <tt>info</tt> displays the queried capabilities of all hardware devices found.
650 \section apinotes API Notes
652 RtAudio is designed to provide a common API across the various supported operating systems and audio libraries. Despite that, some issues should be mentioned with regard to each.
654 \subsection linux Linux:
656 RtAudio for Linux was developed under Redhat distributions 7.0 - Fedora. Three different audio APIs are supported on Linux platforms: OSS, <A href="http://www.alsa-project.org/">ALSA</A>, and <A href="http://jackit.sourceforge.net/">Jack</A>. The OSS API has existed for at least 6 years and the Linux kernel is distributed with free versions of OSS audio drivers. Therefore, a generic Linux system is most likely to have OSS support (though the availability and quality of OSS drivers for new hardware is decreasing). The ALSA API, although relatively new, is now part of the Linux development kernel and offers significantly better functionality than the OSS API. RtAudio provides support for the 1.0 and higher versions of ALSA. Jack, which is still in development, is a low-latency audio server, written primarily for the GNU/Linux operating system. It can connect a number of different applications to an audio device, as well as allow them to share audio between themselves. Input/output latency on the order of 15 milliseconds can typically be achieved using any of the Linux APIs by fine-tuning the RtAudio buffer parameters (without kernel modifications). Latencies on the order of 5 milliseconds or less can be achieved using a low-latency kernel patch and increasing FIFO scheduling priority. The pthread library, which is used for callback functionality, is a standard component of all Linux distributions.
658 The ALSA library includes OSS emulation support. That means that you can run programs compiled for the OSS API even when using the ALSA drivers and library. It should be noted however that OSS emulation under ALSA is not perfect. Specifically, channel number queries seem to consistently produce invalid results. While OSS emulation is successful for the majority of RtAudio tests, it is recommended that the native ALSA implementation of RtAudio be used on systems that have ALSA drivers installed.
660 The ALSA implementation of RtAudio makes no use of the ALSA "plug" interface. All necessary data format conversions, channel compensation, de-interleaving, and byte-swapping is handled by internal RtAudio routines.
662 The Jack API is based on a callback scheme. RtAudio provides blocking functionality, in addition to callback functionality, within the context of that behavior. It should be noted, however, that the best performance is achieved when using RtAudio's callback functionality with the Jack API. At the moment, only one RtAudio instance can be connected to the Jack server. Because RtAudio does not provide a mechanism for allowing the user to specify particular channels (or ports) of a device, it simply opens the first <I>N</I> enumerated Jack ports for input/output.
664 \subsection macosx Macintosh OS X (CoreAudio):
666 The Apple CoreAudio API is based on a callback scheme. RtAudio provides blocking functionality, in addition to callback functionality, within the context of that behavior. CoreAudio is designed to use a separate callback procedure for each of its audio devices. A single RtAudio duplex stream using two different devices is supported, though it cannot be guaranteed to always behave correctly because we cannot synchronize these two callbacks. This same functionality might be achieved with better synchrony by creating separate instances of RtAudio for each device and making use of RtAudio blocking calls (i.e. RtAudio::tickStream()). The <I>numberOfBuffers</I> parameter to the RtAudio::openStream() function has no affect in this implementation.
668 It is not possible to have multiple instances of RtAudio accessing the same CoreAudio device.
670 \subsection irix Irix (SGI):
672 The Irix version of RtAudio was written and tested on an SGI Indy running Irix version 6.5.4 and the newer "al" audio library. RtAudio does not compile under Irix version 6.3, mainly because the C++ compiler is too old. Despite the relatively slow speed of the Indy, RtAudio was found to behave quite well and input/output latency was very good. No problems were found with respect to using the pthread library.
674 \subsection windowsds Windows (DirectSound):
676 In order to compile RtAudio under Windows for the DirectSound API, you must have the header and source files for DirectSound version 5.0 or higher. As far as I know, there is no DirectSoundCapture support for Windows NT. Audio output latency with DirectSound can be reasonably good, especially since RtAudio version 3.0.2. Input audio latency still tends to be bad but better since version 3.0.2. RtAudio was originally developed with Visual C++ version 6.0 but has been tested with .NET.
678 The DirectSound version of RtAudio can be compiled with or without the UNICODE preprocessor definition.
680 \subsection windowsasio Windows (ASIO):
682 The Steinberg ASIO audio API is based on a callback scheme. In addition, the API allows only a single device driver to be loaded and accessed at a time. ASIO device drivers must be supplied by audio hardware manufacturers, though ASIO emulation is possible on top of systems with DirectSound drivers. The <I>numberOfBuffers</I> parameter to the RtAudio::openStream() function has no affect in this implementation.
684 A number of ASIO source and header files are required for use with RtAudio. Specifically, an RtAudio project must include the following files: <TT>asio.h,cpp; asiodrivers.h,cpp; asiolist.h,cpp; asiodrvr.h; asiosys.h; ginclude.h; iasiodrv.h; iasiothiscallresolver.h,cpp</TT>. The Visual C++ projects found in <TT>/tests/Windows/</TT> compile both ASIO and DirectSound support.
686 The Steinberg provided <TT>asiolist</TT> class does not compile when the preprocessor definition UNICODE is defined. Note that this could be an issue when using RtAudio with Qt, though Qt programs appear to compile without the UNICODE definition (try <tt>DEFINES -= UNICODE</tt> in your .pro file). RtAudio with ASIO support has been tested using the MinGW compiler under Windows XP, as well as in the Visual Studio environment.
688 \section wishlist Possible Future Changes
690 There are a few issues that still need to be addressed in future versions of RtAudio, including:
693 <li>Provide a mechanism so the user can "pre-fill" audio output buffers to allow precise measurement of an acoustic response;</li>
694 <li>Allow the user to read / write non-interleaved data to / from the audio buffer;</li>
695 <li>Further support in Windows OS for multi-channel (>2) input / output. This is currently only possible with ASIO interface (in large part due to limitations with the DirectSound API). But perhaps a port to the WinMM API should be investigated?</li>
696 <li>Investigate the possibility of allowing the user to select specific channels of a soundcard. For example, if an audio device supports 8 channels and the user wishes to send data out channels 7-8 only, it is currently necessary to open all 8 channels and write the two channels of output data to the correct positions in each audio frame of an interleaved data buffer.</li>
699 \section acknowledge Acknowledgements
701 Many thanks to the following people for providing bug fixes and improvements:
703 <LI>Robin Davies (Windows DS and ASIO)</LI>
704 <LI>Ryan Williams (Windows non-MS compiler ASIO support)</LI>
705 <LI>Ed Wildgoose (Linux ALSA and Jack)</LI>
708 The RtAudio API incorporates many of the concepts developed in the <A
709 href="http://www.portaudio.com/">PortAudio</A> project by Phil Burk
710 and Ross Bencina. Early development also incorporated ideas from Bill
712 href="http://www-ccrma.stanford.edu/software/snd/sndlib/">sndlib</A>.
714 href="http://www-ccrma.stanford.edu/groups/soundwire/">SoundWire
715 group</A> provided valuable feedback during the API proposal stages.
717 The early 2.0 version of RtAudio was slowly developed over the course
718 of many months while in residence at the <A
719 href="http://www.iua.upf.es/">Institut Universitari de L'Audiovisual
720 (IUA)</A> in Barcelona, Spain and the <A
721 href="http://www.acoustics.hut.fi/">Laboratory of Acoustics and Audio
722 Signal Processing</A> at the Helsinki University of Technology,
723 Finland. Much subsequent development happened while working at the <A
724 href="http://www-ccrma.stanford.edu/">Center for Computer Research in
725 Music and Acoustics (CCRMA)</A> at <A
726 href="http://www.stanford.edu/">Stanford University</A>. The most
727 recent version of RtAudio was finished while working as an assistant
728 professor of <a href="http://www.music.mcgill.ca/musictech/">Music
729 Technology</a> at <a href="http://www.mcgill.ca/">McGill
730 University</a>. This work was supported in part by the United States
731 Air Force Office of Scientific Research (grant \#F49620-99-1-0293).
733 \section license License
735 RtAudio: a realtime audio i/o C++ classes<BR>
736 Copyright (c) 2001-2005 Gary P. Scavone
738 Permission is hereby granted, free of charge, to any person
739 obtaining a copy of this software and associated documentation files
740 (the "Software"), to deal in the Software without restriction,
741 including without limitation the rights to use, copy, modify, merge,
742 publish, distribute, sublicense, and/or sell copies of the Software,
743 and to permit persons to whom the Software is furnished to do so,
744 subject to the following conditions:
746 The above copyright notice and this permission notice shall be
747 included in all copies or substantial portions of the Software.
749 Any person wishing to distribute modifications to the Software is
750 requested to send the modifications to the original developer so that
751 they can be incorporated into the canonical version.
753 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
754 EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
755 MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
756 IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR
757 ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
758 CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
759 WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.