Table of Contents

HDR and the Wwise Voice Pipeline

As previously mentioned, the HDR system works with logical volumes set in the Wwise authoring tool, and ignores the actual amplitude of audio data. The levels considered by the HDR system's logic correspond to those at the input of the HDR bus, as can be seen in the Voice Monitor when inspecting the HDR bus in Input bus mode. These levels depend on the voice volumes, which are the sum of contributions from the Actor-Mixer Hierarchy, control busses of the Master-Mixer Hierarchy, actions, RTPC and distance attenuation, as well as on gains of each mixing bus located upstream of the HDR bus in the signal flow. When multiple signal paths lead to the HDR bus, for example when using auxiliary sends, the path with maximum gain is used by the HDR logic.

At each audio frame, the sound engine begins by computing the volume of all voices as it is seen by the HDR bus. It then executes the HDR system's logic, which returns the global HDR gain/attenuation according to the position of the HDR window, and reapplies this gain into each voice. Once this is done, it processes voices as usual: voices are evaluated against the volume threshold to decide if they should be virtual, data is produced, and volumes are applied.

Make-Up Gain and Source Normalization

The HDR system's logic ignores two volume properties supplied by Wwise: source normalization and make-up gain(s). These volume controls are primarily used to normalize the audio assets independently of their logical volume, as you could have done in your wave editor prior to importing them into Wwise. For example, you can use the localization make-up gain to even out volume differences between two languages, but the HDR system should have the same behavior across all languages. Thus, it is important to have volume controls that are ignored by the HDR logic, and source normalization and make-up gains play this role.

Note that the virtual voices system, as well as the Voice Monitor calculations, does not take into account the make-up gain and the source normalization. However, virtual voices do take the HDR attenuation into account. Thus, voices can become virtual when they are below the HDR window.

The make-up gain slider in the Conversion tab of the Property Editor can also be used with HDR for aesthetic purposes. Since it is transparent from the HDR system's point of view, it can be seen as a post-HDR volume (in terms of signal flow), allowing you to change the volume of sounds independently of the HDR window. Normally, any sound playing alone above the HDR threshold comes out of the system at the same level. With the make-up gain, you can effectively make it louder. For example, you might want the player's guns of your first-person shooter to be much louder than the rest, while benefiting from the HDR system's dynamic mixing capabilities. In this case, give them a boost using the make-up gain.