At its most basic level MQA is software upscaling at the first "unfold" and additional hardware (if present) upscaling filter at the second "unfold". I think it’d be more dynamic if it was actually upscaling everything to a super high bit rate (like a chord m scaler). In its current form, it’s just managing the upscaling if the content is available from the publisher at the native high sampling rates. There’s just not much native content in the super high sample rates.
I don’t listen to Tidal or even have a MQA device, but conceptually it makes sense for recreating atmospheric transients in music. The actual resulting sound of their upscaling filters is what really should be questioned and to date, I’ve not seen any qualitative A/B listening test among MQA’s upscaling filter and other software (HQplayer, PGGB, etc) and hardware upscaling (Chord M Scaler) solutions.
Where they get hammered is from their business practices. Namely, downsampling content which was originally a native higher bitrate than the tier the customer paid for—in essence, giving the customer what they paid for. As they sell tiered plans that makes sense, but for the audiophile, the main consumers of the hi-res music, it breaks the cardinal law of "do no harm" to the music.
Ironically, the audiophile (who has bought in to the concept) would likely go for the high tier plan and have the hardware decoder, thus making it a moot point.
