This issue is discussed in the patent application. The touchup channel can overload if the HF information becomes too complex to compress into the available bit rate. A large buffer on the touchup channel can help prevent overloads if the complex section is not too long in duration. The success of this system is very dependent upon the complexity of the HF information and the size of the buffer. The patent application does not specify a buffer size nor the probability of an overload. It basically just states that this can happen and if it does, the HF channel reverts to a lossy mode.
Please note that noise is the most “complex” signal. A random noise signal will not compress. In contrast, musical tones tend to compress well.
Cymbals and other high-frequency percussion instruments may cause the most problems.
Thanks for info. Now this really shed some light why it is not considered a truly lossless codec. It depends on the complexity of music program it can handle. It would be interesting to know if there’s test to determine how MQA ‘behave’ across different music patterns or when subjected to a 10kHz square wave response test.
There is more to MQA than that patent document; and there is a world of difference between a patent doc and a real-world implementation where experienced recording engineers are allowed freedom to control the parameters of the encoding. Ultimately, we need to hear a lot more material and a lot more mainstream material to judge.
In my opinion this is the least interesting part of MQA, frankly. You get a size improvement for files encoded in higher bit rates. You get a massive size penalty for 16/44 files. The music origami does nothing to improve the sound as far as I know.
To me the key contribution of MQA would be in improving the creation of the master, either by careful mastering of original files or careful production of new recordings.
As others have stated, we shall be able to see the merits when enough material becomes available. Until then it’s a theoretical exercise.
@joel, I think we all understand to some degree what it is and what it does, the big question is where is it? there have been a number of false starts over the last year or so and still today we have very little selection of MQA material available.
They only have two recording studios (one in LA and one in Norway) listed on their website and no record labels. I would love to know the license and royalty terms for all of the entities from recording to distribution to playback.
If this 83+ is not enough for you there are many other articles full of questions and answers that are linked in this thread. I’m not sure what is left to be answered. We just need content.
Yes there are lots of articles on theory and commentary out there, but it reminds me of the old Wendy’s marketing theme “Where’s the Beef?”
I’ve bought the Mytek Brooklyn DAC with the pretty little blue MQA glow. Problem is aside from a few 2L purchased MQA titles, there is nothing and even less talk about anything coming. Where’s the Beef?
Furthermore the very little MQA titles I have to listen to do not leave me with jaw on the ground. Is it better than 24x192 or DSD-64 or whatever? Some MQA sounded good, some ‘eh’ just like my experience with the other formats. I think it all comes down to the production process. Without being able to do real side-by-side listening of different music genres beyond choir and quartets, and without real jaw dropping performance, the whole MQA experience has been nothing short of disappointing so far.
