- Update to 22.10:
* Added a quick comment to the SDL backend.
- Update to 22.09.01:
* Fix a regression where requesting a specific mastering voice format
would return a format overwritten by device defaults
- Update to 22.09:
* Update to SDL 2.24. This has been made a hard requirement, as it
fixes/adds way too much to ignore:
2.1 and 4.1 audio is now supported
* On Windows, we now prioritize DirectSound over WASAPI
* The PulseAudio samples hack has been removed
* The non-power-of-two samples hack for Emscripten/OSS has also been removed
* GetDeviceDetails(0) now uses SDL_GetDefaultAudioInfo, obsoleting our hack (forwarded request 1030904 from ukbeast89)