Audiokinetic Wwise Knowledge Base

Category » Wwise

There were 76 articles found in this category:

61. Is there a way in Wwise to retrieve a list of available states, RTPC’s, events, etc?
    Beyond querying SoundFrame connected directly to Wwise, you would need to parse generated text and .BNK files to retrieve this information.
62. If AK::SoundEngine::LoadBank can load a bank from memory, will it load and copy to an internal heap, or will it use the bank "inplace" and if so will it modify the memory ?
    The LoadBank() that loads from memory does it by reading data in-place, without modifying it. Audio Data is registered in-place and not copied or moved. Meta Data is parsed and related objects are instantiated in the sound engine pool. However this represents only a small percentage of the tota ...
63. On the PC Platform, does Wwise cache commonly used decompressed Vorbis into memory, or does it retain the compressed files and decompress on the fly?
    No, we do not cache any decompressed Vorbis data into memory. All Vorbis decompression is done on the fly.
64. Thread safety and the sound engine API's functions and callback notifications
    Whenever you call a function on the sound engine's API, you actually push a command into a queue. Even AK::SoundEngine::RenderAudio() simply pushes a message "process previous commands" into the queue. These commands are consumed and processed by another thread, the audio thread, which is synchr ...
65. What is the difference between a State and a Switch, and when should I use one or the other?
    States are global, whereas Switches are bound to game objects. Usually, use a State whenever the state of the game needs to have an effect on all sound instances. For example, the game would set the State globally on the sound engine depending on what mode it is in, so events triggered by the ga ...
66. Source vs Voice Starvation
    "Source" (I/O) starvation occurs because the I/O throughput could not meet the requirements of the sound engine for a specific streaming sound. "Voice" (CPU) starvation occurs because the sound engine did not have time to refill the output (sink) buffer. This generally occurs when CPU usage is n ...
67. What is the Music Engine?
    The music engine is the high-level engine that handles complex, high-level scheduling of music segments of the interactive music hierarchy.
68. Looping a music segment
    The only way to loop a music Segment is by making it a child of a Playlist container. It is then possible to specify transition rules between segments of the Playlist (like playing the pre-entry/post-exit or not).
69. Initializing the music engine
    In order to successfully post events of the Interactive Music Hierarchy, you must initialize the music engine separately, using AK::MusicEngine::Init(), after you successfully initialized the sound engine. AK::MusicEngine::Term() should be called before AK::SoundEngine::Term().
70. Streaming property on sounds
    It is possible to turn on/off streaming for a sound on a per-platform basis, by unlinking the Stream checkbox in the General Settings tab of the Sound Property Editor on a chosen platform.
71. Is event preparing/unpreparing (AK::SoundEngine::PrepareEvent()) ref-counted?
    Yes, AK::SoundEngine::PrepareEvent() keeps a ref count of prepared events. This means that if AK::SoundEngine::PrepareEvent() has been called 3 times to prepare a given event, it must be called 3 times to unprepare it before it is actually unloaded. This is useful if the same event can be prepa ...
72. Unpreparing an event asynchronously and then re-preparing the same event before it finishes the unprepare
    What happens if you unprepare an event asynchronously and then call prepare event on it (synchronously or asynchronously) before it finishes unpreparing? All of the prepare/unprepare requests are put at the end of a queue that is processed in a separate thread, in the order they were queued. Not ...
73. Differences between targets for AK static libraries
    All of the static libraries shipped by Audiokinetic are built using the following attirubutes: Debug librairies : Monitoring and profiling available, no optimizations Profile libraries : Monitoring and profiling available, optimizations enabled Release libraries : No monitoring or profiling, op ...
74. Source starvation and voice starvation explained
    Source starvation means that somehow the input stream starves probably because the throughput obtained from the stream manager is insufficient for some reason (e.g. due to long seek times on DVD etc.) Voice starvation means that the sound engine was not able to fill the hardware voice fast enou ...
75. Can Wwise automatically determine the size, shape and enclosure of a room and set the reverb accordingly?
    No. The game has to feed the sound engine with the parameters to be applied based on the geometry of rooms.
76. How do we reference events in game?
    You can either use the event name (strings) directly or you can use Event IDs. The overhead cost of using strings is pretty low (The sound engine does not keep strings internally, it is using a hash function to quickly compute strings into unique IDs ). But for game that ends up having many tho ...