|
Digital Drive Upsamplers, DACs, jitter, shakes and analogue withdrawals, this is it. |
For Sale Ads |
Use this form to submit comments directly to the Asylum moderators for this forum. We're particularly interested in truly outstanding posts that might be added to our FAQs.You may also use this form to provide feedback or to call attention to messages that may be in violation of our content rules.
Original Message
Transient optimized chain
Posted by Dave_K on February 20, 2017 at 10:43:04:
MQA is currently being used as a compressed audio format to stream Warner titles that were previously mastered for a hi-res PCM release e.g. DVD-A or download. Universal is said to follow.
For existing recordings, MQA have claimed that if the ADC is known, its identity or characteristics can be encoded, and then the decoder can utilize that information to select a reconstruction filter. I guess you would have to assume the same ADC is used for all the tracks of a multi-track recording. But I haven't seen them discuss the details of this process publicly or give examples, so I don't know what to think about its potential benefits, if any. Also, given how quickly Warner encoded their catalog, I think there is a very high probability that the encoding was generic.
Those are just practical issues though. My real complaint about MQA is theoretical. In their AES paper, Stuart and Craven proposed a Gaussian as an ideal system impulse response. In image processing, a Gaussian filter is also known as a blur filter or unsharpen filter. It is used to reduce sharpness or to artistically blur images. It will have the same result on a music waveform, smoothing over transients and blunting attack. How is that transient optimized? In the paper, Stuart & Craven note their ideal is equivalent to "only" 30m of air. Why is 30 meters ideal and not zero meters? From an engineer's POV, their idea just seems dumb.