It is time for an update on this, as I promised I would share my views on MQA. Now is probably a good time to do so:
I understand that many of you want to know "will Vinnie Rossi support MQA with current / future dacs? What is Vinnie's take on MQA?"
The simple answer is "I am still not sure at this time"
because I have too many questions that I need answered by MQA. It is my responsibility to do all the due diligence required before deciding to adapt MQA in my products, and I am still working on this part if it. I am waiting to hear back from MQA (I have signed an NDA and am supposed to be sent more information, and I would of course need to clear it with them before I can post about any specific details here). If this is all that you want to know, you can STOP reading here.
Below are only some of my many questions that I would like MQA to answer for me before I can make the decision to get on board, or not. 1)
There is plenty of ambiguity regarding things like software vs hardware decoding, and if software decoding can now (or ever will) do "full unfolding" of MQA's "audio origami" encoding, and how it would compare to their hardware based decoding (in which I believe a chip on board the dac unit would run firmware that does the decoding, and an LED could be lit verifying what you are hearing is authentic MQA).2)
There are also many questions about how MQA is "correcting" the "time smearing" of the d/a conversion process. The pre-ringing of a d/a converter's impulse response that they mention in their videos that causes the time smearing should not be an issue if your dac implementation uses minimal phase filtering (BTW - this is the default filter of LIO's DSD/PCM dac, so there is no pre-ringing or smearing of the impulse response. Linear phase filters do have pre-ringing - but not minimal phase filters). So what is the problem with the d/a conversion then? 3)
Is MQA working with various d/a chip manufacturers (e.g. TI, AKM, Analog Devices, ESS Sabre, etc) and telling them what they are dong wrong and how it should be "corrected" via MQA's hardware based decoding / correction (and the same question for all the A/D converters used in the studios when they recorded the music in the first place)? If they have a collection of, say, 20,000 albums to encode to MQA and correct in the process, will they find the information on the A/D converters used in the original digital recorder used by the recording studio? They would need to know what was used (define the problem
encoding the files with MQA to fix the problem
introduced by the A/D process that they speak of. At least this is how I see it, but not every A/D converter is the same, so I hope it is not just a manner of applying the same algorithm to every album. If it is not, then it is going to be a tremendous task for them! 4)
I know they don't like to call this "EQ'ing" or "DSP" - but it certainly sounds like DSP to me at this time. And is the hardware MQA implementation simply doing what can be done via playback software (decoding, decompressing, authenticating) - for those who don't own playback software that does this?
My questions above, and plenty more that I have for MQA, are what I hope to have answered so I can post answers directly from MQA's mouth
, and not add to all the confusion. We may think we know the answers already, but unless it has been confirmed with MQA and is current information
(MQA seems to have changed its stance here and there since it was first brought to the market. For example - playback software decoding). If you have been reading all the links that I have provided in this thread, you can see that there is still much confusion - even from those who are doing their independent measurement comparisons and the like. So at this point
, my only criticism of MQA is the lack of detailed information on their website to answer all these (and more)
questions. I believe it is important to know what is happening to the music and what we are getting. How is it being processed, changed, manipulated... whatever you want to call it. All the confusion and speculation on the various forums and audio sites could be avoided if everything was already spelled out in a clear fashion. I understand that not everyone understands the technicalities of digital audio and the like, but it would be helpful if MQA would post links to white papers for those of us (especially manufacturers) who are interested in this. I'll be sure to post more when I have concrete information to share instead of just more questions.
Finally - you will note above that I have not even began to discuss one of the most (if not *the* most) important MQA topic: "How does it sound?" I have only been using Roon / TIDAL (software decoding), so please keep this in mind. I still have plenty more listening to do, but my initial impressions vary just like others who have posted about it:
- I have been impressed with some albums, and less so with others (more detail on that later...)
- I do love the concept of TIDAL streaming MQA for those of us who want an enhanced listening experience via streaming.
- I certainly would not want to pay to download my music collection all over again as MQA encoded files - at least not until
we see where this all goes. It's just too early to tell, but it sure it exciting and if it is raising more awareness to the industry regarding
how QUALITY MATTERS, then I certainly dig that aspect of it.
- Also, I currently only stream music to explore new artists. If I really like the music, I buy it
(via download on a number of different sites, or I buy the CD, or I buy the LP). I still want to own it. I understand the industry has been quickly moving to a "pay for access" model, and I do want to have access to as much as possible. Who doesn't? But for me, having access alone is not enough because there is uncertainty (e.g. What if Jay-Z / TIDAL calls it quits?).