|
Critic's Corner Discuss a review. Provide constructive feedback. Talk to the industry. |
|
|
|
Critic's Corner Discuss a review. Provide constructive feedback. Talk to the industry. |
|
In Reply to: How do YOU feel about MQA? posted by Doug Schneider on October 26, 2016 at 14:26:16:
I wasn't at RMAF so haven't had a chance to hear it. But based on what I've read, there's some very solid engineering here.
The idea of correcting the time domain performance of DAC filters for specific DACs is a good one. (I wonder if it will someday be possible to correct, based on a statistical analysis, for systematic errors such as nonlinearity and even the missing codes that afflicted some early converters. Also, the added time domain information brings to mind some other possibilities -- not just accurate level settings -- how can one speak of "fidelity" when one is guessing at the volume control? -- but compression control so that those with good equipment can decompress the commercial release -- and also so that meaningful loudness compensation can be applied when listening at lower levels.)
They also appear to have done a good job of folding the higher frequency information into the lower order bits in a compatible way. This will no doubt be useful for streaming. However, here is where some doubts creep in. The very widest dynamic range material, played at natural levels in the best circumstances, does need most of those bits, though arguably these conditions almost never prevail in commercial releases given that most recordings have a narrower dynamic range, are compressed, and are played at less than natural levels. (In this context, it has also to be remembered that the analog noise floor isn't a hard barrier -- we can hear about 10 dB down into it.)
Another concern I have is that there can be significant signal levels above 20 kHz, e.g.,
http://www.cco.caltech.edu/~boyk/spectra/fig1a.gif
What they are doing is apparently the equivalent of preemphasis and without more information it's impossible to know whether and when ultrasonic information will be discarded, or whether it even matters in light of psychoacoustic masking. (Masking was mentioned n one of the articles -- I wonder whether they are using a DCT compression scheme on the ultrasonics. It's an idea I'd toyed with many years ago, but in Red Book days there wasn't enough room to fit the extra signal in.)
Anyway, this has all to be seen in the light of practicality. I hope that time domain information (and other "Rosetta tone" information as well) will be appended to lossless files for downloads. For streaming, for now, MQA can only be an improvement and while I see a lot of understandable cynicism here, given the record of high definition audio protocols that went nowhere, the fact that Warner is encoding their catalog and that Tidal is speaking of adoption is positive -- though given the low cost of Internet bandwidth and the fact that most of us don't stream over cellular, I have to wonder whether a straightforward lossless high resolution scheme with the time domain correction appended wouldn't be a better approach for all but cellular streaming.
Which being said, in my experience, engineering is less of an obstacle to good sound than the difficulty of establishing high quality standards and practices. Formats that offer what from the general public's perspective are sonic improvements too subtle to care about are typically commercial failures. It's crucially important to get compatibility and economics right. MQA seems to have been very well thought out from both the perspectives of engineering and commercial practicality, and I hope it succeeds.
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- RE: How do YOU feel about MQA? - josh358 06:14:03 10/30/16 (0)
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2024, Audio Asylum, all rights reserved ]
