Here's that link to Libaudioverse on GitHub

Hi Ctoth and welcome to the forum!
This is great I will try and test it out today or tomorrow. From what you described it sounds very promising.
Don't worry after you have about 5 or 10 posts on this forum you will be able to post multiple links.
Btw, how is camlorn? I haven't talked to him since he was driven off this forum.
Didn't you also make accessible output 2? Or is that somebody else.

hmm, nice. not bad at all. It actually worked on my system without any notable issues. I hope camlorn records some sort of demo like he did with camlorn_audio, as that was fascinating.

oh, well it would seem that it has been deleted from his blog. But I'll post up a temporary Dropbox link in an hour or so when I get to school, because it is stored on my school computer. But I'm serious that thing is so freaking cool that I'm still fascinated with it even though I downloaded it a year ago. Best 3d sound I've ever heard that wasn't actual reality. Literally you can tell! When objects are behind and in front, above  in below; I even flinched the first time I listened to it because it sounded like the song being played was truly about to hit me in the face. I'll definitely post that link for you

That demo is still up, here.
I haven't made another one because Libaudioverse is substantially the same, though less tinny in real-world configurations.  The difference with Libaudioverse is primarily performance, stability, and a much more sane API.  I think OpenALSoft has better vertical, but I'm not able to look at or use his algorithms: doing so would make Libaudioverse permanently GPL and strip me of my ability to offer commercial licenses.  I've about matched the reverb quality, and I support (and have the potential to support) a bunch of other stuff related to synthesis, i.e. instrument design.  Keep in mind that this is a pre-release Libaudioverse, we're still at least weeks away from me declaring 0.8.
And to answer the other question I've been getting: Libaudioverse is the only piece of this that doesn't run on Mac and Linux to my knowledge, but it will in the future.  Not the immediate future, though.

There isn't one anymore.  It's on the project page for Camlorn_audio.  As I recall, the old post was lost in my migration from one VPS provider to another after the first managed to lose all the data for all their customers, but it might still be floating around...somewhere.

Wow, I would like to take a look at the sourcecode for this!
I am not 100% sure how the scripting language works, but it looks pretty clear from just glancing through it.

The turning with the keyboard is a little too quick though, I have a little problem figuring out where I'm facing.

Just wondering, why use YAML rather than python? Is it because it is safer? I didn't know that there was a pre-built parser for a declarative language like that, it's pretty cool!

Also, why not create a single .exe file rather than lay all the .pyd files out there? I know accessible_output2 and libaudioverse along with sounds and the map files need to be outside the .exe, but that should be it.
Awesome work!

And, since I can answer more questions that were posted while I'm posting:
Yaml is a data format, not a full language.  It acts as a serialization format that happens to be human-editable, like json or xml but more convenient.  Q and I both use it because it's more convenient than typing everything out as Python syntax and because you can save in that format, too.
.pyd files are actually C Python extension dlls with the extension changed to indicate this fact.  To my knowledge, they have to be outside the executable and in the same directory as the executable for Python to find them.  There might be some sort of hackish way to get them into the exe, but why bother?
You get to take any gameplay questions up with Q, I'm only here because this pertains to Libaudioverse directly and I need to monitor for horrible breakage, given that this is the first full piece of software to reach even a somewhat wide audience.  If there is not horrible breakage in Libaudioverse somewhere, then I simply haven't waited long enough yet (or this game doesn't use it yet, I suppose that's possible too).

Here is the result.
Maximum:33%
Around 25%
Part of the experiment: https://dl.dropboxusercontent.com/u/102 … rackle.mp3

Perhaps, drivers or audio software installed in my computer might be causing an extra lug since I have the same experience using DAW software. Also, NVDA uses around 18% CPU in my computer. So, the crackling might be happening for only my environment.

@yukionozawa
I don't know what's going on in your situation, it's somewhat odd.  I'd not expect the game to be taking an entire core, but I'm pretty sure it's not audio backend trouble.  The only thing I can think of that might be causing it is if your Ram has low bandwidth, as Libaudioverse very definitely doesn't fit entirely in the cache.  But I'm not sure that shows up as increased CPU usage.
@kyleman123
I can't figure out what's going on in your situation without more info.  The Python traceback is unhelpful, as the crash is happening in the C++ piece.  I've improved Libaudioverse's logging to try to catch it, but the current version of the game doesn't have that change in it at the moment.  Initialization involves something like 15 functions and 500 lines of code, it's the single largest code path that is exposed as one function.
Running from an external shouldn't have an impact on it, at least not at the stage of initialization.  I'm  betting it's crashing in audio backend initialization.  What is your OS?

I am having the same problem as Kyleman123
I can't open the game on my windows computer

I tried to run the game the other night and ran into the same problem. Pretty sure the traceback was the same, except the access violation address was all zeros. I got it to run exactly once, but never again.