Hello Brian and Dave, I'm doing my MSc Sound Design course and want to take "Sounds in VR" as my Final Project theme. Especially, I want to concentrate on problems of atmosphere/background noises. For, instance, it is not obvious how a stadium crowd noise should be created. It shouldn't be a stereo file (as you mentioned on the conferences), but it's quite hard to create such an environment with numerous localized mono sources as well. Is it better to implement some kind of procedural audio system? I would ask you if you have any thoughts in this topic that could help me in my research. Thanks
In a case like you mentioned above, ideally, you would want to set up an HRTF audio recorder in a live stadium. Speaking in UE4 terms, I would then set up a stereo based sound cue/actor in the center of the stadium. Design the loudness of the sound for that, with loudness getting higher in the direction you are travelling.
Not a lot of research has been done into this so far. It depends on whether you're trying to capture background noise for a static listener (audience member at a concert) or if you have someone that can move around. If locomotion is available then you have to be careful with background noises, since if you just capture and playback stereo/binaural tracks then distinct elements (birds, etc.) will follow the listener's head as they move/turn.
Thank you guys for your answers. I'm not sure if it's still related to Oculus discussion, but now I am thinking about more concrete project which is a hybrid system between localized and background noises in VR. For instance, if the player stands in the rain, he should be able to hear separate raindrops around him. But few steps away sound of the rain is just like a noise. This behavior stays the same as the player moves. So basically, rain drops are sound cues that move in space (and change in time) with the player. Background rain noise could consist of just a few mono files.
To be honest, I don't have much experience with game audio and don't know where to start from. Is it possible to create it with the resources of Unreal/Unite or should I involve Max/MSP as well? The tough question: is it possible (and necessary) to attach these sounds to the procedural rain drops? Do you find such an idea interesting/uncommon? If you know about any kind of similar systems, please let me know.
You most definitely do NOT want to attach a soundcue to each individual rain particle, for the simple fact of resources, namely CPU resources. For something like rain, look at the environments your character can go in, for example, under a roof, or inside a building, and create sound files for the environments. When the player transitions, the sounds transition.
There are things that we would like to do, and there are things that we can do applies heavily in this case.
Current WIPs using Unreal Engine 4:Agrona - Tales of an Era: Medieval Fantasy MORPG
portwain, I think you may want to look into a solution for fading between a 3d (mono) and 2d (stereo) sound. Both FMOD and Wwise have built in features to do this. For the rain scenario, the idea would be to use a 3d spatialized sound when the rain is in the distance. As you move closer to it, and eventually become surrounded by it, you fade from the 3d sound to a 2d sound. So when you're standing in the rain, you're playing a 2d stereo sound - the sound won't exactly match the impact of every single rain drop, but it doesn't have to. That's part of the smoke and mirrors of making games 😉
