I actually hope it's not done in the core as it will potentially increase network traffic possibly to point of overload for some people that are not running hardwired links, ie multiple zones with wifi/powerline links.

If a "simple" app can decode MQA in software, one would like to think that the audio endpoints with their XMOS processors will be more than up to the task especially those that currently upsample/apodise. It also distributes the task across endpoints and doesn't burden the core with it - some already struggle with 192-96k downsample.

I was wondering if the update will be control devices being MQA aware i.e. putting an indicator on the album info that it is MQA encoded and audio endpoints having MQA decoder capability.

Hope that CES will answer many questions. I'm still not convinced that DSP speakers aren't the ultimate place to decode MQA - if you have an authenticated stream all the way to the speaker which will then be perfectly unpacked to >192k, surely that, for the time being at least, will be audio nirvana especially if delivered over a wifi link with no EMI injection through a coax/SpeakerLink connection.

Meridian owner since 1992
Prime & PSU, Focal Elear headphones, Roon (ROCK 8Gi5 NUC), Explorer 1 & 2, F80. 200/203, various Sonos (via Roon).