The fact that Atkinson showed up here

[Norton]

7 hours ago, John_Atkinson said:

 

Manufacturers who have been critical are not necessarily disinterested observers. 

 

I think  you've hit the nail on the head there John.  Linn's very public criticism of MQA for example, I mean it's not as if Linn and  Meridian are rival British manufacturers selling competing active systems into the high end "lifestyle" market.....

Personally I thought Linn's willingness to publicly slag off another UK manufacturer was irredeemably tacky and fatally undermined their values as a premium brand.

 

But at least Linn were open about it.  One might  also observe  that some critics who appear to be disinterested observers are not necessarily  unconnected with rival manufacturers.

 

 

 

[beerandmusic]

I can't believe that people even talk about MQA.

[Albrecht]

On ‎12‎/‎29‎/‎2017 at 9:27 AM, plissken said:

People know what 'trolls like Archimago' have contributed. What have you?

 

We're not talking about me. But I've built up a fair amount of comparative listening experiences, and therefore have some knowledge. Many more folks have built up more experiences and have more knowledge.

 

As a coward, and "anti-science" preacher: Archimago is "anti-contributor" as his contributions are disinformation. He draws erroneous,  pre-determined conclusions, then uses pseudo-science to claim to justify them. Just like the climate change deny-ers. In other words, bad science.

 

 

 

[Albrecht]

On ‎12‎/‎30‎/‎2017 at 1:05 AM, firedog said:

Define troll, please. How is he a troll?

 

Seems to me a serious, well intentioned individual who does his best to actually test and compare instead of just projecting so called "theories" about what is happening in audio reproduction. He also allows contrary views at his blog.

You can disagree with his methods. Doesn't make him a troll. 

Yeah, = his pre-determined conclusions, (then subsequent justifications), are attempts to bait audiophiles into an argument and fight: given that these "climate-change-denyers-like" conclusions are held by agenda driven, bad science, purveyors, and run contrary to 95.5% of experienced, comparative listeners.  

[Don Hills]

5 hours ago, Tony Lauck said:

... (This is something that mastering engineers know about, it's called "inter-sample peaks".)

 

 

Straw man. As you say, it's a well known phenomenon in the industry and any competent engineer will allow headroom for it. Ditto for DAC designers.

[fas42]

22 hours ago, Abtr said:

Can you elaborate on how this might work?

 

"Harshness" in a usual enemy in sound - and the ways to resolve can be as simple as the Treble control in car radios, which can attenuate distorted higher frequencies, and get rid of the subjective scratchiness - or more sophisticated like MQA. Recently I listened to some posted files, with and without MQA, on laptop speakers with no effort made to get best sound from the playback - and, yes, on semi-casual listening the MQA ones sounded "nicer", easier on the ear - but for me this was an "AM radio" effect; I wasn't getting closer to experiencing the musical message.

 

What MQA precisely does to create this subjective difference I haven't worked out yet - DiffMaker type analysis should unearth exactly where the variation is; something I've started looking at - to be continued, unless someone else has already precisely nailed it.

[fas42]

38 minutes ago, Don Hills said:

 

Straw man. As you say, it's a well known phenomenon in the industry and any competent engineer will allow headroom for it. Ditto for DAC designers.

 

Are there any of these "competent engineers"? I recently was looking at some recent popular music tracks, of known performers, which had been put through limiters to maximise loudness - and they were all stinkin' with inter-sample clipping. Which means how objectionable they sound is totally dependent on how the particular DAC reacts to this poor encoding.

[wdw]

12 hours ago, John_Atkinson said:

 

The Stereophile writers who have auditioned decoded MQA files and compared them with the PCM originals (where the provenance is known) have found that there is an improvement in sound quality. Sometimes the improvement is small, sometimes not so small, but there is never a degradation...

John Atkinson

Editor, Stereophile

 

John,  your comment abouts outlines my particular concerns about MQA....based on your earlier very positive writings as well as the effusive praise from Harley, I was keenly interested in hearing it.  A few years ago, here in Vancouver, BC Canada we had a music show and MQA was a primary exhibitor.  Based on their participation we happily attended with the sole intention of hearing this new format but found they would only play a file claimed to be MQA but directly refused, even upon repeated requests, to play the non-MQA equivalent.  I have posts of very similar behaviour at other shows.

It is this appearance of skulduggery that raises my suspicions.  It appears, from our/my perspective, that the Hi-Fi press have been marketed far too heavily while the general consumers is, generally, sent to sit on the sidelines.  I know their behaviour has become more open over time but it still doesn't pass the smell test.

Best,  

 

 

[esldude]

1 hour ago, fas42 said:

 

Are there any of these "competent engineers"? I recently was looking at some recent popular music tracks, of known performers, which had been put through limiters to maximise loudness - and they were all stinkin' with inter-sample clipping. Which means how objectionable they sound is totally dependent on how the particular DAC reacts to this poor encoding.

How do you differentiate between intersample clipping and plain old clipping.  Many releases I see have plain clipping (and as a result probably some intersample clipping too).  

[JSeymour]

4 hours ago, Albrecht said:

 

"We're not talking about me. But I've built up a fair amount of comparative listening experiences, and therefore have some knowledge. Many more folks have built up more experiences and have more knowledge.

 

As a coward, and "anti-science" preacher: Archimago is "anti-contributor" as his contributions are disinformation. He draws erroneous,  pre-determined conclusions, then uses pseudo-science to claim to justify them. Just like the climate change deny-ers. In other words, bad science."

 

The above paragraph is absolute BS.  He posts honestly, back by testing.  What is your predetermined agenda? I have not listened to MQA, but there is no need for me to.  I want the best sounding PCM I can find.  No DRM, no lossy, and I own it.

 

 

[fas42]

2 hours ago, esldude said:

How do you differentiate between intersample clipping and plain old clipping.  Many releases I see have plain clipping (and as a result probably some intersample clipping too).  

 

Plain clipping is where there's a flat top, between two samples - visually, it's obvious that the waveform has exceeded maximum value. Intersample is where you could quite easily convince yourself that clipping never happens, from a visual inspection. And Audacity agrees - at the sample rate of the track, the program splatters red warning markers for clipping throughout, in the first instance; in the second instance, seemingly from limiter processing, Audacity never complains. It's only when you resample, sometimes necessary to 4x, that the clipping points sprout like weeds.

[Don Hills]

3 hours ago, fas42 said:

 

Are there any of these "competent engineers"? ...

 

I was thinking in the context of high fidelity. I don't consider such distorted music as high fidelity. (Which is a problem for me, because most of the current music that I like is produced in this style.)

[semente]

8 hours ago, fas42 said:

 

Plain clipping is where there's a flat top, between two samples - visually, it's obvious that the waveform has exceeded maximum value. Intersample is where you could quite easily convince yourself that clipping never happens, from a visual inspection. And Audacity agrees - at the sample rate of the track, the program splatters red warning markers for clipping throughout, in the first instance; in the second instance, seemingly from limiter processing, Audacity never complains. It's only when you resample, sometimes necessary to 4x, that the clipping points sprout like weeds.

 

Isn't this clipping a result of improper upsampling settings or something?

[Tony Lauck]

18 hours ago, esldude said:

How do you differentiate between intersample clipping and plain old clipping.  Many releases I see have plain clipping (and as a result probably some intersample clipping too).  

Easily done with audio workstation software such as Audacity, Izotope RX, or Soundforge.

 

There is no way to tell whether or not a DAC will produce intersample clipping on a recording that has no clipping except by listening (or measuring the inpulse response).  For listening, you can use a digital volume control in the computer, such as HQPlayer or a digital volume control in the DAC (if it's been done correctly) and see if this eliminates the harshness.  (You will need to make corresponding analog volume control changes to get a fair comparison.)  If you are a decent recording engineer you will be able to recognize clipping when it occurs on most signal peaks. 

 

Example of the theory:  An unclipped square wave at 8 kHz has harmonics at 24 kHz, 40kHz, etc...  If these are stripped off as the result of a filter for the 44.1 Khz sampling rate the result will be a sine wave at 8 kHz.  The peak amplitude of the sine wave divided by the peak amplitude of the square wave  will be a ratio of 4 / pi, about 2.1 dB.   For worst case there are pathological waveforms for which the peaks can be arbitrarily large if the theoretical filter (perfect sinc) is used.  (I've constructed some where the peaks are more than 10 dB above 0 dBfs, but not something likely to occur in any decent music. (Pop music mastered for loudness by such as Lucy do not count as decent in my book, neither the music, nor the engineering.)

 

It is easy to see these effects when doing any sort of EQ.  One would expect to lose headroom when the amplitude response of a filter includes a boost, but it is even possible to lose headroom if the filter has a cut. (Witness the squarewave discussion above.)

 

HQPlayer will show clips when upsampling or if doing digital room correction as I do.  I use about 4 dB of digital reduction in HQPlayer before sending the signal to my DAC.  (I don't lose any resolution at this point, because the DAC takes a 32 bit signal and I run it with digital volume control.)  I have carefully calibrated the analog gain in my system so there is only a few dB of headroom when the digital volume control is set to -0 dB.  It is not possible to clip the active monitors at full volume setting, and this is about 10 dB louder than what is good for my ears, while still having completely quiet sound out of the speakers when playing 24 bit dither noise.

 

[Tony Lauck]

7 hours ago, semente said:

 

Isn't this clipping a result of improper upsampling settings or something?

 

DAC designers have to make a tradeoff.  Assuming the digital sampling has extra guard bits they can afford to provide headroom to prevent clipping.  However, unless the output of the upsampling is then reduced down to the actual resolution of the converter circuitry there will be clipping.  Here's the conflict:  if they provide little reduction than "hot" music will sound distorted, but the measured S/N of the DAC will be good.  If they provide more reduction then "hot" music will be clean, but the measured S/N of the DAC will be inferior.  Pro audio equipment often allows for the converter to be set up so that these tradeoffs are under user control.  (For example the Mytek Stereo 192-DSD allows for setting the amount of analog headroom over a 4 dB range, as I recall.)

[Sal1950]

On 12/31/2017 at 7:41 AM, John_Atkinson said:

The Stereophile writers who have auditioned decoded MQA files and compared them with the PCM originals (where the provenance is known) have found that there is an improvement in sound quality. Sometimes the improvement is small, sometimes not so small, but there is never a degradation.

Has the listening been done under a blind controlled  condition?

If the difference is obvious how do you determine that it's really better and not just different? It's easy for folks to be fooled by euphonic coloration, look at the current fad for vinyl.

Either way, in the end your willing in time to lose the distribution of totally lossless original PCM files to the DRM like control of MQA?

That will be a sad ending, your mags position is sticking knives in the backs of audiophiles for the sake of some marketing $ dollars.

[John_Atkinson]

2 minutes ago, Sal1950 said:

Has the listening been done under a blind controlled  condition?

 

If you read my article on listening to MQA files, you will note that some of my comparisons were performed single-blind.

 

2 minutes ago, Sal1950 said:

That will be a sad ending, your mags position is sticking knives in the backs of audiophiles for the sake of some marketing $ dollars.

 

What marketing dollars? As I have pointed out before, while MQA has advertised in The Absolute Sound, they haven't spent a cent advertising in Stereophile.

 

BTW, I note you quoting the late Peter Aczel in your posts - not the most appropriate role model for those who desire integrity in audio journalism. :-)

 

John Atkinson

Editor, Stereophile

 

 

[John_Atkinson]

1 hour ago, Tony Lauck said:

DAC designers have to make a tradeoff.  Assuming the digital sampling has extra guard bits they can afford to provide headroom to prevent clipping.  However, unless the output of the upsampling is then reduced down to the actual resolution of the converter circuitry there will be clipping. 

 

Indeed. I discuss the prevalence of inter-sample overs in my review of MBL's N31 DAC in the February 2018 issue of Stereophile, with examples. Interestingly, MBL has achieved 3.5dB overload margin in their filters without sacrificing low-level resolution.

 

John Atkinson

Editor, Stereophile

[Ralf11]

what were Aczel's  integrity issues?

[esldude]

1 hour ago, Tony Lauck said:

Easily done with audio workstation software such as Audacity, Izotope RX, or Soundforge.

 

There is no way to tell whether or not a DAC will produce intersample clipping on a recording that has no clipping except by listening (or measuring the inpulse response).  For listening, you can use a digital volume control in the computer, such as HQPlayer or a digital volume control in the DAC (if it's been done correctly) and see if this eliminates the harshness.  (You will need to make corresponding analog volume control changes to get a fair comparison.)  If you are a decent recording engineer you will be able to recognize clipping when it occurs on most signal peaks. 

 

Example of the theory:  An unclipped square wave at 8 kHz has harmonics at 24 kHz, 40kHz, etc...  If these are stripped off as the result of a filter for the 44.1 Khz sampling rate the result will be a sine wave at 8 kHz.  The peak amplitude of the sine wave divided by the peak amplitude of the square wave  will be a ratio of 4 / pi, about 2.1 dB.   For worst case there are pathological waveforms for which the peaks can be arbitrarily large if the theoretical filter (perfect sinc) is used.  (I've constructed some where the peaks are more than 10 dB above 0 dBfs, but not something likely to occur in any decent music. (Pop music mastered for loudness by such as Lucy do not count as decent in my book, neither the music, nor the engineering.)

 

It is easy to see these effects when doing any sort of EQ.  One would expect to lose headroom when the amplitude response of a filter includes a boost, but it is even possible to lose headroom if the filter has a cut. (Witness the squarewave discussion above.)

 

HQPlayer will show clips when upsampling or if doing digital room correction as I do.  I use about 4 dB of digital reduction in HQPlayer before sending the signal to my DAC.  (I don't lose any resolution at this point, because the DAC takes a 32 bit signal and I run it with digital volume control.)  I have carefully calibrated the analog gain in my system so there is only a few dB of headroom when the digital volume control is set to -0 dB.  It is not possible to clip the active monitors at full volume setting, and this is about 10 dB louder than what is good for my ears, while still having completely quiet sound out of the speakers when playing 24 bit dither noise.

 

I think Frank's (fas42) answer was better.  If you oversample a file by 4x or 8x it will uncover nearly all intersample overs. You still have to figure out if some DACs handle those overs or not, but not having them in the first place is better all the way around.  Some recording interfaces with variable gain on the output do allow you to work around it. 

[fas42]

9 hours ago, semente said:

 

Isn't this clipping a result of improper upsampling settings or something?

 

Tony pointed out one aspect of it - on the other side, in the mastering, the problem is created with applying limiting, in an audio workstation, to maximise hotness of the track: there is no, or almost no, headroom with the peak samples, spread throughout the track. Forgetting upsampling, the reconstituted analogue waveform, from filtering, is now hitting over-value points much of the time - and the SQ may suffer, very considerably.

 

I did this exercise only a month of so ago: the released version of the current pop track was sounding irky, all through the clip. Looked at the waveform; limited, with peak values at maximum; resampling confirmed inter-sample clipping - all I then did was attenuate the released version of the track, about 1 dB or so - just enough to eliminate that inter-sample behaviour, and resave as a new version. So, the new was identical to the old, with a slight level reduction, that's all - but it made a major difference to the subjective quality on the laptop speakers; the circuitry was far more in its comfort zone, and sounded such.

 

Edit: I hadn't read all the latest posts, before writing this - I'm just repeating what others have said, in different words.

[John_Atkinson]

5 minutes ago, Ralf11 said:

what were Aczel's  integrity issues?

 

Well documented. Involved accepting money for subscriptions, including from this would-be reader, but then folding his magazine and as far as I am aware using the funds to start a speaker company. (He favorably reviewed the speaker in his magazine without revealing that he had an ownership interest in the company that made it.) When that company was wound up, he relaunched the magazine but refused to honor paid-for subs to the earlier incarnation.

 

John Atkinson

Editor, Stereophile

[sandyk]

8 minutes ago, fas42 said:

 

Tony pointed out one aspect of it - on the other side, in the mastering, the problem is created with applying limiting, in an audio workstation, to maximise hotness of the track: there is no, or almost no, headroom with the peak samples, spread throughout the track. Forgetting upsampling, the reconstituted analogue waveform, from filtering, is now hitting over-value points much of the time - and the SQ may suffer, very considerably.

 

I did this exercise only a month of so ago: the released version of the current pop track was sounding irky, all through the clip. Looked at the waveform; limited, with peak values at maximum; resampling confirmed inter-sample clipping - all I then did was attenuate the released version of the track, about 1 dB or so - just enough to eliminate that inter-sample behaviour, and resave as a new version. So, the new was identical to the old, with a slight level reduction, that's all - but it made a major difference to the subjective quality on the laptop speakers; the circuitry was far more in its comfort zone, and sounded such.

Frank

 The solution sounds way too simple.

 Perhaps some other members can try doing the same to see if this also works for them, or whether this problem is more specifically related to the actual gear that you are using ?

Regards

Alex

[fas42]

1 minute ago, sandyk said:

Frank

 The solution sounds way too simple.

 Perhaps some other members can try doing the same to see if this also works for them, or whether this problem is more specifically related to the actual gear that you are using ?

Regards

Alex

 

Yes, trivially easy - but effective, for this older laptop I'm using. Its a HP model, specifically sold at the time with promo stickers for multimedia use. IOW, the audio circuitry and speakers were a touch better than the PC ordinaires of that year, .

 

Of course this is not going to solve the dynamic range issues, and all the other ills of heavy handed mastering - but at least you are giving the playback circuitry a little bit of, correct, headroom - which may make a worthwhile improvement.

[Fair Hedon]

36 minutes ago, John_Atkinson said:

 

If you read my article on listening to MQA files, you will note that some of my comparisons were performed single-blind.

 

 

What marketing dollars? As I have pointed out before, while MQA has advertised in The Absolute Sound, they haven't spent a cent advertising in Stereophile.

 

BTW, I note you quoting the late Peter Aczel in your posts - not the most appropriate role model for those who desire integrity in audio journalism. :-)

 

John Atkinson

Editor, Stereophile

 

 

Totally and utterly disingenuous. MQA partners in fact HAVE advertised in Stereophile. You made a market, and are now reaping the financial benefits via ad dollars.