Determine if multiple tracks are needed to meet requirements

[tdaede]

With a single track, sending a subset of the audio to a client requires the audio track to be disassembled and reassembled. With multiple tracks, the client can select the tracks it needs. Determine if the single track implementation is enough to meet the requirements.

[sunghee-hwang]

Even with multi tracks, parsers need to get sample by sample of each track to form IA bitstream with a selected layer. A main difference is that timed metadata of a single track has the boundary information among layers while multi tracks has it as sample size info on each track. Another option may be that file parsers output OBU streams with full layers, then OBU parsers output depacketized IA bitstream with the selected layer to the client.

[sunghee-hwang]

With a single track, sending a subset of the audio to a client requires the audio track to be disassembled and reassembled. With multiple tracks, the client can select the tracks it needs. Determine if the single track implementation is enough to meet the requirements.

As now we have an OBU for each Substream, the client can select Substreams it needs after parsing IA_Static_Meta. So, the current single track meets both R-11 & R-12.