Home Critic's Corner

Discuss a review. Provide constructive feedback. Talk to the industry.

I do not agree with the author's conclusions

>> "But if the peak amplitude of the limited-word-length ultrasonic data is reduced to that in the original recording, you do preserve those data's full resolution above the original analog noise floor." <<

It seems to me that this statement implies that it is impossible for the human ear/brain to extract information below the "analog noise floor" (however *that* is defined). I disagree with this premise. That seems tantamount to saying that the human ear/brain cannot hear below the ambient noise level in the room, or that the dynamic range of LP is only ~60 dB - significantly worse than digital.

The overwhelming experiential evidence of millions of people is that the resolution of a good LP playback system is *extremely* difficult to match, even with 24-bit digital systems. And literally billions of people have experienced the "cocktail party effect" where a listener can discern a quiet conversation even in a crowded, noisy environment - effectively hearing well *below* the "noise floor" in both cases.

MQA's graphs are only given in "noise per root Hz" or other apparently arbitrary and non-standard methods of measurement. This has the result of obscuring the loss of resolution that results from the MQA encode/decode process. The main debate seems to be centered around the degree to which this is audible. Is there a perfect 1:1 correlation between the measured "noise floor" in "noise per root Hz" and human ear/brain audibility thresholds? Not as far as I am aware. (If there were, perhaps MP3 would be indistinguishable from Redbook CD quality digital audio.)

Furthermore I personally have heard a noticeable loss of resolution with MQA-processed files. Perhaps the most notable example of this is the "whisper" overdub on Jim Morrison's lead vocal in The Doors track "Riders on the Storm". In my system the whisper is clearly audible in the original high-res 96/24 file, yet difficult to make out in the MQA version (tested with both the Meridian Explorer2 and Mytek Brooklyn D/A converters, both of which provide full hardware decoding of MQA files).

For me personally MQA holds little appeal. I believe that others have created digital filters with audible performance equal to or better than the MQA digital filter. For the truly hard-core, try using a filterless ("NOS") D/A converter or stick with DSD, both of which can take the concepts promoted by MQA to the next level.

Yet I also recognize that others audiophiles will have other priorities. Perhaps they primarily listen to streaming and don't have the bandwidth required to stream true high-res (such as offered by Qobuz Sublime+). Or perhaps they prefer the MQA digital filter to the one built into their current D/A converter and are willing to buy a new D/A converter to gain access to the MQA digital filter. There is no "right" or "wrong" here.

As a designer of hardware, I believe that Mike Moffat of Schiit has the wisest perspective - content is king. The reason to design, manufacture, or purchase hardware should be based on it supporting the formats which have the content you enjoy. For example, if one is a big fan of obscure music prior to the era of the LP, it would probably be wise to own a turntable capable of replaying 78s, as only the most popular music of that era has been transferred to either LP or digital formats.

Conversely if there ever is a large demand for MQA-equipped products, a hardware manufacturer would not be wise to ignore this. Obviously those companies that have developed custom digital filter solutions and simple methods of firmware upgrades will have an advantage here, as AudioQuest has demonstrated with their latest DragonFly Black and Red USB DACs. Without knowing more details of their internal architecture, I would assume that manufacturers such as dCS, Playback Design, PS Audio, and many others would have the same advantage that Ayre has - there appears to be little impediment to easily adding MQA decoding as part of a simple firmware update.

As always, only my own personal opinions and not necessarily those of my employer or fellow employees.

This post is made possible by the generous support of people like you and our sponsors:
  Analog Engineering Associates  

Follow Ups Full Thread
Follow Ups
  • I do not agree with the author's conclusions - Charles Hansen 19:24:26 06/07/17 (0)


Post a Message!

Forgot Password?
Moniker (Username):
Password (Optional):
  Remember my Moniker & Password  (What's this?)    Eat Me
E-Mail (Optional):
Message:   (Posts are subject to Content Rules)
Optional Link URL:
Optional Link Title:
Optional Image URL:
Upload Image:
E-mail Replies:  Automagically notify you when someone responds.