3 Fixed compilation on PowerPC.
5 Fixed compilation on some targets that lack lock-free 64-bit atomics.
7 Fixed a crash when parsing certain option values.
9 Fixed applying noexcept in the public headers with MSVC.
11 Fixed building for UWP with vcpkg.
13 Improved compatibility when compiling as C++20 or later.
15 Integrated fmtlib for some examples and utilities.
19 Updated library codebase to C++17.
21 Implemented the ALC_SOFT_system_events extension.
23 Implemented the AL_EXT_debug extension.
25 Implemented the AL_EXT_direct_context extension.
27 Implemented speaker configuration and headphones detection on CoreAudio.
29 Fixed a potential crash with some extension functions on 32-bit Windows.
31 Fixed a crash that can occur when stopping playback with the Oboe backend.
33 Fixed calculating the reverb room rolloff.
35 Fixed EAX occlusion, obstruction, and exclusion low-pass filter strength.
37 Fixed EAX distance factor calculations.
39 Fixed querying AL_EFFECTSLOT_EFFECT on auxiliary effect slots.
41 Fixed compilation on some macOS systems that lack libdispatch.
43 Fixed compilation as a subproject with MinGW.
45 Changed the context error state to be thread-local. This is technically out
46 of spec, but necessary to avoid race conditions with multi-threaded use.
48 Split the cubic resampler into 4-point spline and gaussian variants. The
49 latter prioritizing the suppression of aliasing distortion and harmonics,
50 the former not reducing high frequencies as much.
52 Improved timing precision of starting delayed sources.
54 Improved ring modulator quality.
56 Improved performance of convolution reverb.
58 Improved WASAPI device enumeration performance.
62 Added 'noexcept' to functions and function types when compiled as C++. As a
63 C API, OpenAL can't be expected to throw C++ exceptions, nor can it handle
64 them if they leave a callback.
66 Added an experimental config option for using WASAPI spatial audio output.
68 Added enumeration support to the PortAudio backend.
70 Added compatibility options to override the AL_VENDOR, AL_VERSION, and
73 Added an example to play LAF files.
75 Disabled real-time mixing by default for PipeWire playback.
77 Disabled the SndIO backend by default on non-BSD targets.
81 Implemented the AL_SOFT_UHJ_ex extension.
83 Implemented the AL_SOFT_buffer_length_query extension.
85 Implemented the AL_SOFT_source_start_delay extension.
87 Implemented the AL_EXT_STATIC_BUFFER extension.
89 Fixed compiling with certain older versions of GCC.
91 Fixed compiling as a submodule.
93 Fixed compiling with newer versions of Oboe.
95 Improved EAX effect version switching.
97 Improved the quality of the reverb modulator.
99 Improved performance of the cubic resampler.
101 Added a compatibility option to restore AL_SOFT_buffer_sub_data. The option
102 disables AL_EXT_SOURCE_RADIUS due to incompatibility.
104 Reduced CPU usage when EAX is initialized and FXSlot0 or FXSlot1 are not
107 Reduced memory usage for ADPCM buffer formats. They're no longer converted
108 to 16-bit samples on load.
112 Fixed CoreAudio capture support.
114 Fixed handling per-version EAX properties.
116 Fixed interpolating changes to the Super Stereo width source property.
118 Fixed detection of the update and buffer size from PipeWire.
120 Fixed resuming playback devices with OpenSL.
122 Fixed support for certain OpenAL implementations with the router.
124 Improved reverb environment transitions.
126 Improved performance of convolution reverb.
128 Improved quality and performance of the pitch shifter effect slightly.
130 Improved sub-sample precision for resampled sources.
132 Improved blending spatialized multi-channel sources that use the source
135 Improved mixing 2D ambisonic sources for higher-order 3D ambisonic mixing.
137 Improved quadraphonic and 7.1 surround sound output slightly.
139 Added config options for UHJ encoding/decoding quality. Including Super
142 Added a config option for specifying the speaker distance.
144 Added a compatibility config option for specifying the NFC distance
147 Added a config option for mixing on PipeWire's non-real-time thread.
149 Added support for virtual source nodes with PipeWire capture.
151 Added the ability for the WASAPI backend to use different playback rates.
153 Added support for SOFA files that define per-response delays in makemhr.
155 Changed the default fallback playback sample rate to 48khz. This doesn't
156 affect most backends, which can detect a default rate from the system.
158 Changed the default resampler to cubic.
160 Changed the default HRTF size from 32 to 64 points.
164 Fixed PipeWire version check.
166 Fixed building with PipeWire versions before 0.3.33.
170 Fixed CoreAudio capture.
172 Fixed air absorption strength.
174 Fixed handling 5.1 devices on Windows that use Rear channels instead of
177 Fixed some compilation issues on MinGW.
179 Fixed ALSA not being used on some systems without PipeWire and PulseAudio.
181 Fixed OpenSL capturing noise.
183 Fixed Oboe capture failing with some buffer sizes.
185 Added checks for the runtime PipeWire version. The same or newer version
186 than is used for building will be needed at runtime for the backend to
189 Separated 3D7.1 into its own speaker configuration.
193 Implemented the ALC_SOFT_reopen_device extension. This allows for moving
194 devices to different outputs without losing object state.
196 Implemented the ALC_SOFT_output_mode extension.
198 Implemented the AL_SOFT_callback_buffer extension.
200 Implemented the AL_SOFT_UHJ extension. This supports native UHJ buffer
201 formats and Super Stereo processing.
203 Implemented the legacy EAX extensions. Enabled by default only on Windows.
205 Improved sound positioning stability when a source is near the listener.
207 Improved the default 5.1 output decoder.
209 Improved the high frequency response for the HRTF second-order ambisonic
212 Improved SoundIO capture behavior.
214 Fixed UHJ output on NEON-capable CPUs.
216 Fixed redundant effect updates when setting an effect property to the
219 Fixed WASAPI capture using really low sample rates, and sources with very
220 high pitch shifts when using a bsinc resampler.
222 Added a PipeWire backend.
224 Added enumeration for the JACK and CoreAudio backends.
226 Added optional support for RTKit to get real-time priority. Only used as a
227 backup when pthread_setschedparam fails.
229 Added an option for JACK playback to render directly in the real-time
230 processing callback. For lower playback latency, on by default.
232 Added an option for custom JACK devices.
234 Added utilities to encode and decode UHJ audio files. Files are decoded to
235 the .amb format, and are encoded from libsndfile-compatible formats.
237 Added an in-progress extension to hold sources in a playing state when a
238 device disconnects. Allows devices to be reset or reopened and have sources
239 resume from where they left off.
241 Lowered the priority of the JACK backend. To avoid it getting picked when
242 PipeWire is providing JACK compatibility, since the JACK backend is less
243 robust with auto-configuration.
247 Improved alext.h's detection of standard types.
249 Improved slightly the local source position when the listener and source
252 Improved click/pop prevention for sounds that stop prematurely.
254 Fixed compilation for Windows ARM targets with MSVC.
256 Fixed ARM NEON detection on Windows.
258 Fixed CoreAudio capture when the requested sample rate doesn't match the
259 system configuration.
261 Fixed OpenSL capture desyncing from the internal capture buffer.
263 Fixed sources missing a batch update when applied after quickly restarting
266 Fixed missing source stop events when stopping a paused source.
268 Added capture support to the experimental Oboe backend.
272 Updated library codebase to C++14.
274 Implemented the AL_SOFT_effect_target extension.
276 Implemented the AL_SOFT_events extension.
278 Implemented the ALC_SOFT_loopback_bformat extension.
280 Improved memory use for mixing voices.
282 Improved detection of NEON capabilities.
284 Improved handling of PulseAudio devices that lack manual start control.
286 Improved mixing performance with PulseAudio.
288 Improved high-frequency scaling quality for the HRTF B-Format decoder.
290 Improved makemhr's HRIR delay calculation.
292 Improved WASAPI capture of mono formats with multichannel input.
294 Reimplemented the modulation stage for reverb.
296 Enabled real-time mixing priority by default, for backends that use the
297 setting. It can still be disabled in the config file.
299 Enabled dual-band processing for the built-in quad and 7.1 output decoders.
301 Fixed a potential crash when deleting an effect slot immediately after the
302 last source using it stops.
304 Fixed building with the static runtime on MSVC.
306 Fixed using source stereo angles outside of -pi...+pi.
308 Fixed the buffer processed event count for sources that start with empty
311 Fixed trying to open an unopenable WASAPI device causing all devices to
314 Fixed stale devices when re-enumerating WASAPI devices.
316 Fixed using unicode paths with the log file on Windows.
318 Fixed DirectSound capture reporting bad sample counts or erroring when
321 Added an in-progress extension for a callback-driven buffer type.
323 Added an in-progress extension for higher-order B-Format buffers.
325 Added an in-progress extension for convolution reverb.
327 Added an experimental Oboe backend for Android playback. This requires the
328 Oboe sources at build time, so that it's built as a static library included
331 Added an option for auto-connecting JACK ports.
333 Added greater-than-stereo support to the SoundIO backend.
335 Modified the mixer to be fully asynchronous with the external API, and
336 should now be real-time safe. Although alcRenderSamplesSOFT is not due to
337 locking to check the device handle validity.
339 Modified the UHJ encoder to use an all-pass FIR filter that's less harmful
340 to non-filtered signal phase.
342 Converted examples from SDL_sound to libsndfile. To avoid issues when
343 combining SDL2 and SDL_sound.
345 Worked around a 32-bit GCC/MinGW bug with TLS destructors. See:
346 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83562
348 Reduced the maximum number of source sends from 16 to 6.
350 Removed the QSA backend. It's been broken for who knows how long.
352 Got rid of the compile-time native-tools targets, using cmake and global
353 initialization instead. This should make cross-compiling less troublesome.
357 Implemented the AL_SOFT_direct_channels_remix extension. This extends
358 AL_DIRECT_CHANNELS_SOFT to optionally remix input channels that don't have
359 a matching output channel.
361 Implemented the AL_SOFT_bformat_ex extension. This extends B-Format buffer
362 support for N3D or SN3D scaling, or ACN channel ordering.
364 Fixed a potential voice leak when a source is started and stopped or
365 restarted in quick succession.
367 Fixed a potential device reset failure with JACK.
369 Improved handling of unsupported channel configurations with WASAPI. Such
370 setups will now try to output at least a stereo mix.
372 Improved clarity a bit for the HRTF second-order ambisonic decoder.
374 Improved detection of compatible layouts for SOFA files in makemhr and
377 Added the ability to resample HRTFs on load. MHR files no longer need to
378 match the device sample rate to be usable.
380 Added an option to limit the HRTF's filter length.
384 Converted the library codebase to C++11. A lot of hacks and custom
385 structures have been replaced with standard or cleaner implementations.
387 Partially implemented the Vocal Morpher effect.
389 Fixed the bsinc SSE resamplers on non-GCC compilers.
391 Fixed OpenSL capture.
393 Fixed support for extended capture formats with OpenSL.
395 Fixed handling of WASAPI not reporting a default device.
397 Fixed performance problems relating to semaphores on macOS.
399 Modified the bsinc12 resampler's transition band to better avoid aliasing
402 Modified alcResetDeviceSOFT to attempt recovery of disconnected devices.
404 Modified the virtual speaker layout for HRTF B-Format decoding.
406 Modified the PulseAudio backend to use a custom processing loop.
408 Renamed the makehrtf utility to makemhr.
410 Improved the efficiency of the bsinc resamplers when up-sampling.
412 Improved the quality of the bsinc resamplers slightly.
414 Improved the efficiency of the HRTF filters.
416 Improved the HRTF B-Format decoder coefficient generation.
418 Improved reverb feedback fading to be more consistent with pan fading.
420 Improved handling of sources that end prematurely, avoiding loud clicks.
422 Improved the performance of some reverb processing loops.
424 Added fast_bsinc12 and 24 resamplers that improve efficiency at the cost of
425 some quality. Notably, down-sampling has less smooth pitch ramping.
427 Added support for SOFA input files with makemhr.
429 Added a build option to use pre-built native tools. For cross-compiling,
430 use with caution and ensure the native tools' binaries are kept up-to-date.
432 Added an adjust-latency config option for the PulseAudio backend.
434 Added basic support for multi-field HRTFs.
436 Added an option for mixing first- or second-order B-Format with HRTF
437 output. This can improve HRTF performance given a number of sources.
439 Added an RC file for proper DLL version information.
441 Disabled some old KDE workarounds by default. Specifically, PulseAudio
442 streams can now be moved (KDE may try to move them after opening).
446 Implemented capture support for the SoundIO backend.
448 Fixed source buffer queues potentially not playing properly when a queue
451 Fixed possible unexpected failures when generating auxiliary effect slots.
453 Fixed a crash with certain reverb or device settings.
455 Fixed OpenSL capture.
457 Improved output limiter response, better ensuring the sample amplitude is
462 Implemented the ALC_SOFT_device_clock extension.
464 Implemented the Pitch Shifter, Frequency Shifter, and Autowah effects.
466 Fixed compiling on FreeBSD systems that use freebsd-lib 9.1.
468 Fixed compiling on NetBSD.
470 Fixed the reverb effect's density scale and panning parameters.
472 Fixed use of the WASAPI backend with certain games, which caused odd COM
473 initialization errors.
475 Increased the number of virtual channels for decoding Ambisonics to HRTF
478 Changed 32-bit x86 builds to use SSE2 math by default for performance.
479 Build-time options are available to use just SSE1 or x87 instead.
481 Replaced the 4-point Sinc resampler with a more efficient cubic resampler.
483 Renamed the MMDevAPI backend to WASAPI.
485 Added support for 24-bit, dual-ear HRTF data sets. The built-in data set
486 has been updated to 24-bit.
488 Added a 24- to 48-point band-limited Sinc resampler.
490 Added an SDL2 playback backend. Disabled by default to avoid a dependency
493 Improved the performance and quality of the Chorus and Flanger effects.
495 Improved the efficiency of the band-limited Sinc resampler.
497 Improved the Sinc resampler's transition band to avoid over-attenuating
500 Improved the performance of some filter operations.
502 Improved the efficiency of object ID lookups.
504 Improved the efficienty of internal voice/source synchronization.
506 Improved AL call error logging with contextualized messages.
508 Removed the reverb effect's modulation stage. Due to the lack of reference
509 for its intended behavior and strength.
513 Fixed resetting the FPU rounding mode after certain function calls on
516 Fixed use of SSE intrinsics when building with Clang on Windows.
518 Fixed a crash with the JACK backend when using JACK1.
520 Fixed use of pthread_setnane_np on NetBSD.
522 Fixed building on FreeBSD with an older freebsd-lib.
524 OSS now links with libossaudio if found at build time (for NetBSD).
528 Fixed an issue where resuming a source might not restart playing it.
530 Fixed PulseAudio playback when the configured stream length is much less
531 than the requested length.
533 Fixed MMDevAPI capture with sample rates not matching the backing device.
535 Fixed int32 output for the Wave Writer.
537 Fixed enumeration of OSS devices that are missing device files.
539 Added correct retrieval of the executable's path on FreeBSD.
541 Added a config option to specify the dithering depth.
543 Added a 5.1 decoder preset that excludes front-center output.
547 Implemented the AL_EXT_STEREO_ANGLES and AL_EXT_SOURCE_RADIUS extensions.
549 Implemented the AL_SOFT_gain_clamp_ex, AL_SOFT_source_resampler,
550 AL_SOFT_source_spatialize, and ALC_SOFT_output_limiter extensions.
552 Implemented 3D processing for some effects. Currently implemented for
553 Reverb, Compressor, Equalizer, and Ring Modulator.
555 Implemented 2-channel UHJ output encoding. This needs to be enabled with a
556 config option to be used.
558 Implemented dual-band processing for high-quality ambisonic decoding.
560 Implemented distance-compensation for surround sound output.
562 Implemented near-field emulation and compensation with ambisonic rendering.
563 Currently only applies when using the high-quality ambisonic decoder or
564 ambisonic output, with appropriate config options.
566 Implemented an output limiter to reduce the amount of distortion from
569 Implemented dithering for 8-bit and 16-bit output.
571 Implemented a config option to select a preferred HRTF.
573 Implemented a run-time check for NEON extensions using /proc/cpuinfo.
575 Implemented experimental capture support for the OpenSL backend.
577 Fixed building on compilers with NEON support but don't default to having
580 Fixed support for JACK on Windows.
582 Fixed starting a source while alcSuspendContext is in effect.
584 Fixed detection of headsets as headphones, with MMDevAPI.
586 Added support for AmbDec config files, for custom ambisonic decoder
587 configurations. Version 3 files only.
589 Added backend-specific options to alsoft-config.
591 Added first-, second-, and third-order ambisonic output formats. Currently
592 only works with backends that don't rely on channel labels, like JACK,
595 Added a build option to embed the default HRTFs into the lib.
597 Added AmbDec presets to enable high-quality ambisonic decoding.
599 Added an AmbDec preset for 3D7.1 speaker setups.
601 Added documentation regarding Ambisonics, 3D7.1, AmbDec config files, and
602 the provided ambdec presets.
604 Added the ability for MMDevAPI to open devices given a Device ID or GUID
607 Added an option to the example apps to open a specific device.
609 Increased the maximum auxiliary send limit to 16 (up from 4). Requires
610 requesting them with the ALC_MAX_AUXILIARY_SENDS context creation
613 Increased the default auxiliary effect slot count to 64 (up from 4).
615 Reduced the default period count to 3 (down from 4).
617 Slightly improved automatic naming for enumerated HRTFs.
619 Improved B-Format decoding with HRTF output.
621 Improved internal property handling for better batching behavior.
623 Improved performance of certain filter uses.
625 Removed support for the AL_SOFT_buffer_samples and AL_SOFT_buffer_sub_data
626 extensions. Due to conflicts with AL_EXT_SOURCE_RADIUS.
630 Implemented device enumeration for OSSv4.
632 Fixed building on OSX.
634 Fixed building on non-Windows systems without POSIX-2008.
636 Fixed Dedicated Dialog and Dedicated LFE effect output.
638 Added a build option to override the share install dir.
640 Added a build option to static-link libgcc for MinGW.
644 Fixed building with JACK and without PulseAudio.
646 Fixed building on FreeBSD.
648 Fixed the ALSA backend's allow-resampler option.
650 Fixed handling of inexact ALSA period counts.
652 Altered device naming scheme on Windows backends to better match other
655 Updated the CoreAudio backend to use the AudioComponent API. This clears up
656 deprecation warnings for OSX 10.11, although requires OSX 10.6 or newer.
660 Implemented a JACK playback backend.
662 Implemented the AL_EXT_BFORMAT and AL_EXT_MULAW_BFORMAT extensions.
664 Implemented the ALC_SOFT_HRTF extension.
666 Implemented C, SSE3, and SSE4.1 based 4- and 8-point Sinc resamplers.
668 Implemented a C and SSE based band-limited Sinc resampler. This does 12- to
669 24-point Sinc resampling, and performs anti-aliasing.
671 Implemented B-Format output support for the wave file writer. This creates
672 FuMa-style first-order Ambisonics wave files (AMB format).
674 Implemented a stereo-mode config option for treating stereo modes as either
675 speakers or headphones.
677 Implemented per-device configuration options.
679 Fixed handling of PulseAudio and MMDevAPI devices that have identical
682 Fixed a potential lockup when stopping playback of suspended PulseAudio devices.
684 Fixed logging of Unicode characters on Windows.
686 Fixed 5.1 surround sound channels. By default it will now use the side
687 channels for the surround output. A configuration using rear channels is
690 Fixed the QSA backend potentially altering the capture format.
692 Fixed detecting MMDevAPI's default device.
694 Fixed returning the default capture device name.
696 Fixed mixing property calculations when deferring context updates.
698 Altered the behavior of alcSuspendContext and alcProcessContext to better
699 match certain Windows drivers.
701 Altered the panning algorithm, utilizing Ambisonics for better side and
702 back positioning cues with surround sound output.
704 Improved support for certain older Windows apps.
706 Improved the alffplay example to support surround sound streams.
708 Improved support for building as a sub-project.
710 Added an HRTF playback example.
712 Added a tone generator output test.
714 Added a toolchain to help with cross-compiling to Android.
718 Implemented EFX Chorus, Flanger, Distortion, Equalizer, and Compressor
721 Implemented high-pass and band-pass EFX filters.
723 Implemented the high-pass filter for the EAXReverb effect.
725 Implemented SSE2 and SSE4.1 linear resamplers.
727 Implemented Neon-enhanced non-HRTF mixers.
729 Implemented a QSA backend, for QNX.
731 Implemented the ALC_SOFT_pause_device, AL_SOFT_deferred_updates,
732 AL_SOFT_block_alignment, AL_SOFT_MSADPCM, and AL_SOFT_source_length
735 Fixed resetting mmdevapi backend devices.
737 Fixed clamping when converting 32-bit float samples to integer.
739 Fixed modulation range in the Modulator effect.
741 Several fixes for the OpenSL playback backend.
743 Fixed device specifier names that have Unicode characters on Windows.
745 Added support for filenames and paths with Unicode (UTF-8) characters on
748 Added support for alsoft.conf config files found in XDG Base Directory
749 Specification locations (XDG_CONFIG_DIRS and XDG_CONFIG_HOME, or their
750 defaults) on non-Windows systems.
752 Added a GUI configuration utility (requires Qt 4.8).
754 Added support for environment variable expansion in config options (not
755 keys or section names).
757 Added an example that uses SDL2 and ffmpeg.
759 Modified examples to use SDL_sound.
761 Modified CMake config option names for better sorting.
763 HRTF data sets specified in the hrtf_tables config option may now be
764 relative or absolute filenames.
766 Made the default HRTF data set an external file, and added a data set for
767 48khz playback in addition to 44.1khz.
769 Added support for C11 atomic methods.
771 Improved support for some non-GNU build systems.
775 Fixed a regression with retrieving the source's AL_GAIN property.
779 Fixed device enumeration with the OSS backend.
781 Reorganized internal mixing logic, so unneeded steps can potentially be
782 skipped for better performance.
784 Removed the lookup table for calculating the mixing pans. The panning is
785 now calculated directly for better precision.
787 Improved the panning of stereo source channels when using stereo output.
789 Improved source filter quality on send paths.
791 Added a config option to allow PulseAudio to move streams between devices.
793 The PulseAudio backend will now attempt to spawn a server by default.
795 Added a workaround for a DirectSound bug relating to float32 output.
797 Added SSE-based mixers, for HRTF and non-HRTF mixing.
799 Added support for the new AL_SOFT_source_latency extension.
801 Improved ALSA capture by avoiding an extra buffer when using sizes
802 supported by the underlying device.
804 Improved the makehrtf utility to support new options and input formats.
806 Modified the CFLAGS declared in the pkg-config file so the "AL/" portion of
807 the header includes can optionally be omitted.
809 Added a couple example code programs to show how to apply reverb, and
812 The configuration sample is now installed into the share/openal/ directory
813 instead of /etc/openal.
815 The configuration sample now gets installed by default.