It seems like a pretty obvious question, but as I have been exploring the API I’ve just wondered what the purpose is of unloading banks, when I could just use release() to stop sound objects in the scene. Does the data from a bank remain in the scene even if nothing is playing? The way I have experimented was to make a gameobject and attach a bank loader component to it, then go in to an enemy footstep script and write a public ‘[EventRef]SoundBankLoader music;’ variable, I then call music.Invoke(); in my footstep function. This gives me a dialogue box I place my gameobject with bank component in. This worked, and was cool, but would like to know if there is more high level reasons to use this? or is there a better way to do it?
It allows you to have control over available resources. If you have large or a lot of banks, you may not want them all to be loaded all of the time.
You can have a dedicated bank that only contains sounds for a specific scene, that way if you are not using the scene you do not need to have it loaded.
Basically if you aren’t using the sounds in a bank then there is no need for it to be loaded. But it is always best to load banks well before they are required, to avoid any delays caused trying to play a sound that is being loaded.