The only details I can gather about the Mac implementation of FMOD is that it is based on Carbon and is linkable for CFM as well as Mach-O projects. Are there any details as to how the Mac version of FMOD is implemented? I suspect from the dual release on CFM and Mach-O that it only uses the Carbon Sound Manager exclusively. Is there a possibility in the future of a CoreAudio implementation of FMOD? There are lots of built in Audio Units in MacOS 10.2, such as a 3D sound unit, reverb, panning, echo, etc.. You could implement FMOD Sound FX through custom Audio Units, and arrange your final mixing through AUGraphs. Looking over the FMOD API (I just started looking at it today) it seems that CoreAudio is remarkably compatible with the FMOD API in structure. Plus they just released a CoreAudio guide pdf a while ago (finally). How about it eh? I think it would be totally worth the effort. The Mach-O version of Carbon Sound Manager(and the CFM version when run in MacOS X) really uses CoreAudio as its implementation, so using FMOD on MacOSX is like using a sound API, which uses another sound API, which is basically a wrapper for another sound API that does all the work.
I think this is only useful if the Mac OS supports audio hardware through their API.
It seems the effects in FMod are done in software where not available in hardware, so why redirect to CoreAudio and lose debugging facilities, and introduce minor sound differences?
- Mentos answered 16 years ago
CoreAudio allows you to write your own Audio Units, enabling you to write all of your effects exactly the way you want them at the CoreAudio level. You should still be able to debug them at this level as well. You benefit from a more direct path to the underlying sound architecture of MacOSX, giving you efficiency and lower latency.
Please login first to submit.