I'm working on an MMORPG with Wwise in UE4. In the past we have been directly placing ambient sounds, which are custom sound actors, onto the maps in most situations. However, this method is not very ideal in terms of workflow. Not only we have to place all sounds manually but also we have to double-check the level and move those sounds whenever the level designers modify the map. I'm considering to change the integration workflow to a more actor-based way, meaning that we should place ambient sounds as a component to existing actors, such as torches, campfires, and portals, as much as possible. This method also has its cons. Like it is more complicated to apply the multi-position mode provided by Wwise. Also, we have to frequently ask the map design team to create classes or blueprints for those objects who need a common sound, otherwise we still have to place components manually on each instance of those actors in the level.
I believe there is no absolute solution for this question, but I guess there could be an industry standard or a guide that we can follow? Does anyone know how sound designers deal with the ambient sounds?
Thanks in advance.
🎉 Celebrating 25 Years of GameDev.net! 🎉
Not many can claim 25 years on the Internet! Join us in celebrating this milestone. Learn more about our history, and thank you for being a part of our community!
Should I place ambient sounds directly onto the map or onto the objects?
I think it depends on the specific design needs.
Placing wind sounds in a forest? Probably should just be direct ambient, likely tied to some kind of ambience zone/entity.
Placing splashing water for a waterfall or fountain? Absolutely tie it to the actor/object in question!
If you don't have zone actors for your ambient sounds ("play this ambient sound whenever the player is within this zone") then I would recommend building something like that for your game – it's quite helpful exactly for these kinds of situations.
And, yes, if the artists move the forest, they have to move the forest zone. But you can't really place the wind sound on the tree models – there'd be too many of them. And placing them on just one of the trees, is a recipe for accidents, where it either gets accidentally duplicated, or accidentally removed.
@hplus0603 Thank you Hplus. It's very helpful! You are right. It depends on what kinds of ambience the sound is. We do have zone actors. Now the only problem is how to deal with the “comb-filter effect”, which happens when two close actors play the same audio at the same time and it ruins the sound quality. We are going to find a better way to use the multi-position mode, otherwise we would still use the audio actors instead.
@glenn_zhang You could have a global “playing sounds” manager which registers when sounds start playing, by resource name or handle, and simply disallow starting a second copy of a sound that has started within the last 50 milliseconds or whatever. Assuming you integrate yourself, this is pretty easy to build with an unordered map and a mutex.
@hplus0603 Thank you! This is a great solution for this problem. I'll discuss it with the programmer.
