Core Processing Sequence
This is main processing for M.A.G.E. Project. Being the core code it will use provided data - sounds, specs, scene - and libraries to output the final audio file.
Whiteboard
There is host yet to post the specification onto...
In short it is three phased process.
1. INFO GATHER
- Parses project (XML) file to get specs data.
- Parses ASE (ASCII type MAX scene file) to gather data on scene objects
- Discard any non-related data
EDIT: MAGE is going to create ambisonic data, so matrix will be problem delegated to ambisonic decoder.
2. ACTIVE PROCESS (main loop)
- Load in mono audio-data. Do pre-process filtering.
- Do doppler by dynamic resampling.
- Do distance attenuation.
- (?) Mid-process filters.
- Perform ambisonic panning (by plugins)
- Post to main output at correct offset.
3. AMBISONIC OUTPUT
- Output audio data as ambisonic output stream (thru JACK or as file dump)
- Clear memory and shutdown
4. (Optional) DECODER
- Use one of many available decoders
- Decode ambisonic data to requred matrix (stereo, 5.1, quad, etc)
5. (Optional) POST-PROCESSOR
- Receive decoded multi-channel data
- Mix any overlay (non-3D) audio data (if present)
- Apply post-processing filters.
- Mux/Interleave and output to file.
More detailed info will be available when there will be a host for website.
