Thanks to Ronnie for a very clear and plausible explanation of the stages of decoding/unfolding. I know its speculative but also super helpful.

Suppose the 1st stage MQA decoding is done by software. Could it add whatever metadata is needed to set things up preemptively for DSP SE speakers downstream to perform 2nd and 3rd stage unfolding and also "correct" for their own DACS? Can the speakers do that?

If it is even possible then do you think we only need an endpoint that passes through at 24/96 and keeps the metadata? Wouldn't that make hardware decoding in the endpoint unnecessary?


Once upon a time