Sound Nodes

These nodes make up the bulk of playback nodes. Their main purpose is to define which of their children to activate.

Info

Containers will sometimes set a maximum number of (virtual) "voices". A voice in Wwise is an instance of a playing sound. Voices become virtual if their output volume is below a certain threshold, allowing Wwise to skip many computations to save resources. If a containers voice limit is reached, any additional playback requests will be ignored until it has capacity again.

Random Sequence Container

Wwise documentation

Picks one of its children in one of four different modes:

• continuous sequential plays all of its children in sequence
• continuous step keeps track of the last played child and plays the next one each time it's activated
• random sequential plays all of its children in random order (either truely random or shuffled)
• random step plays a random child each time it's activated

Layer Container

Wwise documentation

Groups children into layers and allows blending between layers based on game parameters (RTPCs). The blending can be customized through curves.

Switch Container

Wwise documentation

Select one of their children based on a switch or state.

Sounds

Sounds are the most common type of source nodes and are basically used for everything that's not music - be it screams, sword strikes, foot steps, or ocean waves. They are typically single-shot and not modified live (although they can be looped through properties).

Wwise stores its audio data in .wem files, typically encoded using Ogg Vorbis. Like most things Wwise, these files use a hash as their filename and are shipped along the soundbank in one of three different ways:

• embedded - the sound is stored in the DATA section of the soundbank. This is commonly done for small sounds (<200kB).
• streaming - to avoid long loading times for soundbanks, larger sounds like music tracks are stored outside the soundbank. You can find them next to the soundank under wem/XX/, where XX is the first two digits of the filename.
• prefetch streaming - for some longer sounds like dialog, frame-accurate playback is important. In these cases, Wwise will store a tiny snippet of just a few kB inside the soundbank. This snippet can be played back immediately while the rest of the sound is being loaded from the external wem folder.

There is also a rather detailed blog post exploring the pros and cons in detail.

Tip

To play back .wem files you can use foobar2000 with the vgmstream decoder plugin. Behind the curtains, Yonder uses the vgmstream CLI to convert your wems for playback, too.