MQA is Vaporware

[mansr]

32 minutes ago, John_Atkinson said:

Okay, I got different results with the Mytek Brooklyn. The front panel indicated that the 24/96 FLAC file was being rendered as 24/192k and a red rather than blue light illuminated next to the MQA indicator. This time, the spectrum with the 10kHz at 0dBFS was dirty, with an 86kHz component at -30dB. The waveform, however, was still visually clean.

 

John Atkinson

Technical Editor, Stereophile

Yikes.

 

32 minutes ago, John_Atkinson said:

Is that a single capture or an average?

[asdf1000]

1 minute ago, Ishmael Slapowitz said:

The article is misleading on so many levels.

 

First, they are trying a sleight of of hand to make those with only casual knowledge of digital audio think

that lossy encodes from 24 bit master files are somehow high resolution.

 

Second, Mastered For iTunes is nothing but utter marketing pablum. Encoding to AAC from the master files...they think

'this is a special favor they are doing? What else would you encode from.

 

Apple has been a major enemy of moving the quality fo recorded music forward since 2001.

 

Already mentioned, for now, it is lossy 'Mastered for iTunes' moving to Apple Music....

[Ishmael Slapowitz]

1 minute ago, Em2016 said:

 

Already mentioned, for now, it is lossy 'Mastered for iTunes' moving to Apple Music....

Oh yes I understood that. I mean that Apple plays fast and loose with it's definitions. Remember they sold 128 AAC as  iTunes downloads in 2001 or 2002 as "CD Quality".

[John_Atkinson]

1 minute ago, mansr said:

Yikes

 

The spuriae are all aliased products, with those in the audioband sufficiently low in level not be audible.. It does look as if MQA renderers differ in how they handled the upsampling.

 

1 minute ago, mansr said:

 

Is that a single capture or an average?

 

Single capture.

 

John Atkinson

Technical Editor, Stereophile

[asdf1000]

1 minute ago, Ishmael Slapowitz said:

Oh yes I understood that. I mean that Apple plays fast and loose with it's definitions. Remember they sold 128 AAC as  iTunes downloads in 2001 or 2002 as "CD Quality".

 

Understood. Those were the old days though. I imagine they (and the other Big Boys with deep pockets) are keeping an eye on Amazon...

 

[Paul R]

12 minutes ago, Rt66indierock said:

 

Notice how cryptic everyone is about what is actually mastered for iTunes. And from the Verge no evidence that Tidal has made enough waves to bother. 

 

The Mastered for iTunes rules are public and very clear.  

[crenca]

@John_Atkinson,& @mansr,

 

Going from memory here, but does not the PS Direct DAC series convert all input (PCM obviously) to DSD128 internally before output.  This is a possible explanation for the your results with it (i.e. low pass filter)?

 

 

[asdf1000]

26 minutes ago, crenca said:

@John_Atkinson,& @mansr,

 

Going from memory here, but does not the PS Direct DAC series convert all input (PCM obviously) to DSD128 internally before output.  This is a possible explanation for the your results with it (i.e. low pass filter)?

 

 

 

[mansr]

32 minutes ago, John_Atkinson said:

The spuriae are all aliased products, with those in the audioband sufficiently low in level not be audible..

At -50 dB they are around 40 dB higher than what a competent DAC should be capable of.

 

32 minutes ago, John_Atkinson said:

It does look as if MQA renderers differ in how they handled the upsampling.

It looks like there might be some clipping occurring. I can make a sample at -6 dBFS instead, which should avoid such problems. Tomorrow.

[mansr]

9 minutes ago, Em2016 said:

All inputs are converted to English in Times Roman, then that's converted to French in Helvetica, then inflection is applied which results in Russian in Baskerville, which is then converted to Japanese in Mincho.

[Ralf11]

3 minutes ago, mansr said:

All inputs are converted to English in Times Roman, then that's converted to French in Helvetica, then inflection is applied which results in Russian in Baskerville, which is then converted to Japanese in Mincho.

 

well there is le error, right ici

 

French fonts should be used - Elina, Roundhead, Decor, etc.

 

NOT Helvetica - Sacre Bleau !!

[The Computer Audiophile]

6 minutes ago, mansr said:

All inputs are converted to English in Times Roman, then that's converted to French in Helvetica, then inflection is applied which results in Russian in Baskerville, which is then converted to Japanese in Mincho.

But what it’s done this way one doesn’t have to compete with anything else because everyone else uses apples (maybe oranges) whereas they use clams. 

[Ishmael Slapowitz]

Mastered For iTunes is such a charade. A farce in fact.

 

If they really gave a damn they would stream lossless ALAC. It would not even make a dent in their bottom line.

[mansr]

11 minutes ago, Ralf11 said:

well there is le error, right ici

 

French fonts should be used - Elina, Roundhead, Decor, etc.

 

NOT Helvetica - Sacre Bleau !!

It's Swiss French, you fool.

[Ralf11]

their food sux

[esldude]

7 hours ago, mansr said:

Here's another scope image. The yellow trace (left channel) is unchanged. The blue trace (right channel) is the same sine wave at -60 dBFS. There's some high-frequency (750 kHz) noise riding on it, but the same distortion is clearly recognisable.

 

The frequency of the sine wave is 10 kHz. Sample rate is 96 kHz.

 

 

Same signal, same DAC without MQA rendering:

Maybe you covered this later, but that undulation looks to be close to the 96 khz sample rate.  

 

EDIT: So I later in the thread saw where you indicated it was 86 khz.  Which fits nicely with 96 khz sample rate minus 10 khz.  Whether it is such an effect or something else I don't know.  

[jabbr]

1 hour ago, Ishmael Slapowitz said:

Mastered For iTunes is such a charade. A farce in fact.

 

If they really gave a damn they would stream lossless ALAC. It would not even make a dent in their bottom line.

No one can tell the difference, so who cares?

[MikeyFresh]

25 minutes ago, jabbr said:

No one can tell the difference, so who cares?

 

I wouldn't say no one, but I also don't wish to have an argument about that.

The mass market certainly doesn't seem to notice a difference between 256AAC and ALAC, and that's all Apple cares about.

[jabbr]

1 minute ago, MikeyFresh said:

 

I wouldn't say no one, but I also don't wish to have an argument about that.

The mass market certainly doesn't seem to notice a difference between 256AAC and ALAC, and that's all Apple cares about.

Ha ha I was being tongue in cheek. Apple can and does use the same justification for AAC vs ALAC as others might for CD vs high res  Slippery slope in my opinion. . 

[MikeyFresh]

26 minutes ago, jabbr said:

Slippery slope in my opinion.

 

Absolutely, although watch how fast Apple changes that stance if another heavyweight like Amazon (the 800 lb. gorilla in the streaming room) decides to offer lossless as @Em2016 suggested earlier today.

[asdf1000]

7 minutes ago, MikeyFresh said:

 

Absolutely, although watch how fast Apple changes that stance if another heavyweight like Amazon (the 300 lb. gorilla in the streaming room) decides to offer lossless as @Em2016 suggested earlier today.

 

My golfing buddies agree too... 😁

[mansr]

12 hours ago, John_Atkinson said:

Okay, I got different results with the Mytek Brooklyn. The front panel indicated that the 24/96 FLAC file was being rendered as 24/192k and a red rather than blue light illuminated next to the MQA indicator. This time, the spectrum with the 10kHz at 0dBFS was dirty, with an 86kHz component at -30dB. The waveform, however, was still visually clean.

Something here doesn't add up. The spectrum clearly shows a -30 dB component at 86 kHz, yet the waveform looks smooth. If I calculate and plot "sin(t) + sin(8.6 t) * 10^(-30/20)" the result looks like this:

 

Quite squiggly.

[mansr]

10 hours ago, mansr said:

It looks like there might be some clipping occurring. I can make a sample at -6 dBFS instead, which should avoid such problems. Tomorrow.

As promised.

sine-10k-3db-mqb.flac

[vortecjr]

I just streamed the file from my Mytek DAC to my analyzer. 96kHz content according to J-River, 192kHz output according to Mytek  DAC. The Mytek  DAC has a purple light next to the MQA logo. Here is the output of my analyzer:

 

    

[mansr]

44 minutes ago, vortecjr said:

I just streamed the file from my Mytek DAC to my analyzer. 96kHz content according to J-River, 192 output according to Benchmark DAC. The Benchmark DAC has a purple light next to the MQA logo. Here is the output of my analyzer:

 

Can you increase the bandwidth of the spectrum display to 100 kHz?