Originally Posted by UltraMega
Accurately rendering the audio is super easy for even a basic computer. Lots of games have the doppler effect. Just about any racing game does. Even the first Crysis games has the doppler effect. Your asking for something we already have.
That said, the level of audio accuracy is mostly up to the developer.
Well, a lot of developers don't spend the appropriate amount of time to make audio work properly, and especially in games where accurate audio is important it can be very frustrating for players. Having a system that figures all of that out - like how a sound changes in frequency over distance, what objects it bounces off of, etc. that can be packaged in such a way that developers need to spend a minimal amount of time thinking about it would probably be for the best especially if it could reliably produce accurate positional audio.
Another issue that I find (and you can see this especially in games like BFV) is that you only really hear things that are happening fairly close to you, for instance on some maps in BFV they are at most a mile or two from one end to the other, and you really cannot hear super loud things like bombs going off or a tank firing once they are out of your general vicinity. Now, obviously you wouldn't want the 10 tanks on the map to be playing the same audio file over and over for every player, that would be fairly ridiculous. But if you had a system for automatically mixing all that audio and giving it the proper distance effects etc. it would probably work out really well and at least be more accurate (in terms of playing all the sound files). *I also realize that a lot of the issues with distanced audio are either up to the developers discretion or a packet issue to help reduce server stress by not sending so many bits of information to every player about audio, although I'm not entirely sure how much more stress this places on a server especially if you can already see those players in the distance why not hear them?
Anyways that turned into a bit of a rant. Basically I want consistently reliable and accurate audio and I think that the RT cores on these Turing cards could help to facilitate that a lot more practically than they can do actual ray-tracing (which to be honest unless you go full-out with is not much of a visual improvement over standard rasterized lighting in current games)