MQA part the 3rd - t't't'timing...

[Werner]

A first MQA claim is to send hires through a CD rate channel. For frequencies below 48k this claim is more or less valid. Above is lost, and replaced with images.

A second claim is that MQA improves on the master. As such it cannot be faithful, strictly speaking.

A third claim is that this all happens tailor made to the recording and to your DAC at home. This has been proven false.

Then there is the provenance thing, also shown to be broken.

 

[gingermrkettle]

In this business, everything is "pay up."

I think the salient point might be what exactly the F is in FLAC. Because in comparison the M in MQA stands for Money.

 

[Rosewind]

In the blurb for a NAD C658 preamp the Danish text happily states,

"Bluesound kan afspille det tabsfrie 24-bit MQA-lydformat, som allerede er tilgængeligt på TIDAL"

"Bluesound is able to play back the lossless 24 bit MQA-sound format that is already available at TIDAL."

Lossless MQA?

The falsehood is out there.

https://www.hifiklubben.dk/nad-c-658-forforstaerker/nadc658gr/

 

[DimitryZ]

The thread that cannot live yet never dies

 

[Jim Audiomisc]

Haven't MQA withdrawn the claim it is 'lossless'? If so, the reported claim by NAD might fall foul of UK consumer law. No idea what the law might be in Denmark, though! 8-] And of course, the people in the UK who judge ad/promo claims have a policy that 'hyperbole' is to be expected.

 

[Rosewind]

I will ask at the local brach for proof that BlueOs enables lossless playback of MQA 24 bit files. That should be fun.

 

[Jim Audiomisc]

(ahem) They may reply: It losslessly plays back what's in the file. 8-]

 

[Rosewind]

 

[Rosewind]

I love the way fancy words spread once used to pimp up a sales pitch. This is from the promo text on the Kef website re. their KEF KC62 [origami]:

"P-Flex (aka Origami Surround)
The P-Flex driver surround has a unique pleated design inspired by Origami, the Japanese art of paper folding, engineered to better resist the acoustic pressure in the cabinet to allow the driver to move more precisely compared to traditional surround designs."

I'm now waiting for some high-endish manufacturer to present their Origami Amp. Hegel Origami? Quad Origami? Cambridge Origami ...

 

[mansr]

I hear Samsung have an origami phone now.

 

[firedog]

I will ask at the local brach for proof that BlueOs enables lossless playback of MQA 24 bit files. That should be fun.

They always reply that it is "perceptually lossless".

 

[Rosewind]

Nice try:

perceptual /pəˈsɛptjʊəl /
▸ adjective relating to the ability to interpret or become aware of something through the senses:
a patient with perceptual problems who cannot judge distances.
– DERIVATIVES
perceptually - adverb
------
Subjective - you perceive or interpret MQA as lossless. Yet it is not truly mathematically lossless. It is MQA who makes the claim and the listener who interprets.

 

[John Phillips]

They always reply that it is "perceptually lossless".

For the consumer the argument seems to boil down to two unprovable assertions that "subtle damage MQA processing inflicts on the audio file is not audible", paired with "subtle improvement from repairing the damage done by A/D kit in the studio is audible".

My own long listening experience is that after the quality of the performance, in a competent studio it's the mixing and mastering that matters vastly more than anything else. Nevertheless, it is clear that there is a population of audiophiles out there who, to my way of thinking, seem to obsess about subtleties while accepting "the elephant in the room". That is the way they enjoy the hobby - fair enough - but it isn't how everyone does.

However, I believe modern practice for the A/D process in a competent studio (96 kHz, 24 bit) simply leaves little or no potentially audible damage to be corrected if I get the MQA "repair" argument. So, there's just a now fixed catalogue of less modern recordings (which still matter, of course) where the MQA repair AIUI might possibly work for some.

I doubt I could easily hear either the damage or the repair. Even if I could I suspect the performance, mixing and mastering is still the overwhelming trump card for my way of enjoying the hobby (and for many others). And my professional engineering experience has always told me "Keep It Simple, Stupid". To me that means avoid additional processing and fix fundamental problems rather than affix a "sticking plaster" to symptoms.

 

[Jim Audiomisc]

It's about as easy to argue that 320k aac as provided by the BBC is 'lossless' if you take for granted a 'perceptually' rider.

 

[Jim Audiomisc]

My own long listening experience is that after the quality of the performance, in a competent studio it's the mixing and mastering that matters vastly more than anything else. Nevertheless, it is clear that there is a population of audiophiles out there who, to my way of thinking, seem to obsess about subtleties while accepting "the elephant in the room". That is the way they enjoy the hobby - fair enough - but it isn't how everyone does.

For me, it tends to bring to mind the way sellers of washing powders, breakfast cereals, etc, tend to have a box that is somewhat bigger than the volume of the solid content. This seems to be for two reasons:

1) People when seeing the boxes tend to judge 'how much I get' from the size of the box.

2) It occupies space on the shelf which otherwise might be used to offer a competitor's products.

I think software vendors may also use this trick at times.

A few days ago we had our shopping volunteer deliver this weeks' shopping. The'd got an alternative 'marge' because the usual one was AWOL. The alternative is a 'light' marge. Examination of the (very!) small print revealed that this is because the product is bulked out with a remarkable amount of water!

Made me think of how many bits-per-sample in 96k/24 and above are wasted on excess specifiying noise levels.

 

[nostromo]

Made me think of how many bits-per-sample in 96k/24 and above are wasted on excess specifiying noise levels.

Yes... That's what I thought upsampling was, once you've doubled the number of bits once there is nothing to be gained except lots of extra 'zeros'. Even the first doubling may not bring out anything useful except a larger a file. Unless the recorded sound was captured with the extra bits in the ADC, adding lots of bits later isn't much use beyond the first doubling of bits I think. I may be wrong as I'm prone to miss-conceptions all over the place.

 

[Jim Audiomisc]

Upsampling by a DAC can be useful in getting smoother output. It aids the low-pass filtering process by shoving out-of-band artifacts up to higher frequencies.

But for storing the results in a file it makes little sense in general.

However having a high sample rate initially makes reduction to fewer bits easier to do well as you can noise shape the process noise up into the ultrasonic.

 

[nostromo]

Thanks... I'm a little clearer now.

(I think I was getting confused with Fourier transforms... Time domain to frequency domain.)

 

[darrenyeats]

44kHz ADC/downsampling needs a very steep anti-aliasing filter ~20-22kHz. Combine that with not uncommon input energy in said transition band, leads to ringing.

The ringing is ultrasonic but could, in theory, have associated audible distortion products.

96kHz allows the transition band to be higher plus less steep. This leads to (a) shorter ringing (b) less ringing energy (c) higher ringing frequencies. A combination that substantially addresses the ringing for normal music signal.

The audible value, if any, is another argument. Just explaining the how and why of hi-res technically!

 

[davidsrsb]

The ringing is only a impulse or square wave artifact, it is not adding or boosting a ringing signal that was not already there.