MQA fracas at RMAF 2018

[Jim Audiomisc]

Hi,
Come on, you know it's been fun. What else have you got to do ?

Did it not keep your brain ticking over ?.

Regards,
Shadders.

Well, TBH I *do* have other things to do. I'm still writing my 'biog' for one thing, and it is almost dinner time. And I know full well that almost anything I write will simply cause DZ to keep writing muddles which then get inflicted on others. Don't really want to waste time trying to deal with him TBH. Easier to let others make up their own minds given the tedious multitude of existing postings.

But if someone is interested I can as an example pick up one tiny point, purely for the non-DZ people reading this...

 

[Shadders]

Well, TBH I *do* have other things to do. I'm still writing my 'biog' for one thing, and it is almost dinner time. And I know full well that almost anything I write will simply cause DZ to keep writing muddles which then get inflicted on others. Don't really want to waste time trying to deal with him TBH. Easier to let others make up their own minds given the tedious multitude of existing postings.

But if someone is interested I can as an example pick up one tiny point, purely for the non-DZ people reading this...

Hi,
It is a forum - say what you like.

For me, i try and work out what the person is trying to say - although it may seem muddled, i think they have some point, and it is worth trying to understand it.

Regards,
Shadders.

 

[Jim Audiomisc]

Lets go over the basics, then. First, though I was never particularly good at mathematics I did have two undergraduate classes in control systems and a graduate course on system dynamics at MIT ('86 Aero/Astro) and took graduate classes in electro-optics and statistics at University of Massachusetts/Lowell School of Mechanical Engineering. So while I certainly can't claim to have hardcore signal processing background, I do understand the basics. I got mostly "A's"

First, any vibratory system, from photons to air molecules will exhibit "ringing" when it passes through a band limiting obstacle. This is axiomatic and not debated for at least a couple of hundred years. In optics, this effect is easily observed in any decent telescope as a series of rings around a star on a clear night and is called diffraction...

.

A genuine radar antenna engineer will, of course, know that the far field pattern of an antenna is linked to the aperture pattern by a 2D FT. They will also know that antenna engineers tend not to blythly use a top-hat illumination, but shape the aperture illumination generally with an 'edge taper' to deal with such "rings". Indeed, by using something like a close approximation to a Gaussian illumination you can get a far field pattern which is... a close approximation to a Gaussian. Not the Airy Pattern relationship.

Possibly, however, a radar engineer won't take into account the observed items having a resolvable size and that also tending to reduce any "rings".

I'll stop there and not add other points about this example. But the reality isn't as simplistic as presented by DZ and you'd need to understand the relevant *physics* as well as the *maths* to realise why his latter claims aren't that well founded on some of his comments. Sadly, having *taught* postgrad courses I know that a few students don't really understand even when they get a bit of paper.

No doubt he will know tell me that I don't know anything about the physics or maths. Again, if someone has a look at the some of my 'biog' pages on mm-wave optics, UKIRT, JCMT, etc, they can make up their own minds. I'll trust them to make their own choices.

http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html

The pages about Hawai'i, UKIRT and JCMT in the 1980s may be an amusing start. But beware the photos of me in a grass skirt and coconut shells. Very non-academic! ;->

 

[Yank]

At any rate, its' usefulness to the audiophile will be decided by the marketplace, which is what I argue for anyway.

If by "the marketplace" you mean "Record Company Suits" then I agree. It is doubtful that end users will be given a choice.

 

[Tin]

Your answer is outrageously patronizing on three levels:

1. You are certain of the immutable correctmess of your verbal argument to a point that you believe that your opponents' difference actually harms their dignity. That is the worst approach to a productive discussion.

Wrong, the worst would be name calling, like you did.

2. You think very little of your opponents intellect, since you kindly offer your solace to them for being "conned". Sad, in your opinion, but totally understandable, given the quality and cleverness of maketing used on them. Truly an insufferable, obnoxious position.

Once again, we're not opponents.
Also, everybody gets conned once in a while, and I see no reason to make a fuss about it. Either it happened to you, or your part of the MQA marketing machine. Take your pick.
Either way I'll admit that your Trumpian behavior hasn't impressed me, so I'll grant you that.

3. You kindly propose that what is at the root of the opponents direct listening experience is their poverty, compared to your wealth. If only they realized that their $1.5 DAC simply cant reveal the richness of LPCM, so they have to settle for second best, instead of saving their money to reach for the best. In an experiential hobby, if MQA/M2TECH provides a superior listening experience vs. LPCM/10k DAC, then it should and will be embraced by audiophiles as best sound for their budget. Irrespective of your helpful pitty at their lack of financial means.

Honestly, this is so wrong that I don't even know where to start.

- Listening to and enjoying music is not a 'experiential hobby'. It -can- be, but there it no need for it.
- 'embraced by audiophiles' is such a meaningless phrase. I have no idea what audiophiles will embrace, and neither have you.
- Where does the wealth part comes from? We're discussing a lossy solution to a non existent problem.

 

[DimitryZ]

A genuine radar antenna engineer will, of course, know that the far field pattern of an antenna is linked to the aperture pattern by a 2D FT. They will also know that antenna engineers tend not to blythly use a top-hat illumination, but shape the aperture illumination generally with an 'edge taper' to deal with such "rings". Indeed, by using something like a close approximation to a Gaussian illumination you can get a far field pattern which is... a close approximation to a Gaussian. Not the Airy Pattern relationship.

Possibly, however, a radar engineer won't take into account the observed items having a resolvable size and that also tending to reduce any "rings".

I'll stop there and not add other points about this example. But the reality isn't as simplistic as presented by DZ and you'd need to understand the relevant *physics* as well as the *maths* to realise why his latter claims aren't that well founded on some of his comments. Sadly, having *taught* postgrad courses I know that a few students don't really understand even when they get a bit of paper.

No doubt he will know tell me that I don't know anything about the physics or maths. Again, if someone has a look at the some of my 'biog' pages on mm-wave optics, UKIRT, JCMT, etc, they can make up their own minds. I'll trust them to make their own choices.

http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html

The pages about Hawai'i, UKIRT and JCMT in the 1980s may be an amusing start. But beware the photos of me in a grass skirt and coconut shells. Very non-academic! ;->

I am not a radar engineer, nor an EE, as I have stated multiple times. I am an aerospace engineer, which, if you really don't know, is like mechanical engineering with a dose of aerodynamics and orbital dynamics, as well as differently focused thermodynamics. My example of electromagnetic "ringing" is from the word of optics, which again, I can see with my eye, just like I can listen to MQA with my ears.

My work with radars is to analyze them for structural adequacy, surface deformation, fatigue reliability and UNDEX survivability for the Navy. My work with missiles is flight dynamics, body loading and structural analysis, aerothermal heating and nosecone survivability. My interest is IR and RF transparent material, composites and exotic metals. An exalted mathematician, such as yourself, can simply think of me as a glorified mechanic.

There are several ways of dealing with diffraction and aberation in optics and RF applications, including but not limited to the methods you describe. We have several PHDs specializing exactly in this area - but they are much nicer than you.

I hated most of my professors, since they were mostly pompous, pedantic asses...pontificating on the virtues of pure knowledge and lamenting the low quality of the students. Sounds familiar?

 

[Shadders]

I am not a radar engineer, nor an EE, as I have stated multiple times. I am an aerospace engineer, which, if you really don't know, is like mechanical engineering with a dose of aerodynamics and orbital dynamics, as well as differently focused thermodynamics. My example of electromagnetic "ringing" is from the word of optics, which again, I can see with my eye, just like I can listen to MQA with my ears.

My work with radars is to analyze them for structural adequacy, surface deformation, fatigue reliability and UNDEX survivability for the Navy. My work with missiles is flight dynamics, body loading and structural analysis, aerothermal heating and nosecone survivability. My interest is IR and RF transparent material, composites and exotic metals.

I hated most of my professors, since they were mostly pompous asses...pontificating on the virtues of pure knowledge and lamenting the low quality of the students. Sounds familiar?

Hi,
Try this - may be helpful.

https://www.dsprelated.com/freebooks/filters/

Regards,
Shadders.

 

[DimitryZ]

Hi,
Try this - may be helpful.

https://www.dsprelated.com/freebooks/filters/

Regards,
Shadders.

Likewise, I have a link for your enjoyment - your best friends at Xivero make a very nice apodization software and appear to be fully backing the overall "elimination of the pre-ringing" approach in the entire recording/playback chain.

https://www.xivero.com/xipodizer/

XIPODIZER
Remove pre-ringing caused by steep linear phase antialiasing filters

From their website:

_______________________________________________________________
Standard resolution CD quality records with a sample rate of 44.1kHz ask for steep low-pass filters at around 22kHz to avoid any aliasing during the analog to digital conversion process. The same principle applies during playback, where the digital to analog converter must use a low-pass filter to reconstruct the analog signal.

Filters used to suppress aliasing and to reconstruct the analog signal are usually “linear phase”, which implies that they create pre-ringing and post-ringing (pls. see Figure 1) by limiting the bandwidth of the input signal.
The ringing amplitude is significant because the spectral energy of music above 20kHz is still quite high.

There is quite a debate in the Hi-Fi community that the unnatural and none-causal impulse response of a linear-phase filter, plagued by pre-ringing, causes artifacts that sound inferior.

In contrast to High Resolution Audio we do not have the possibility to apply the filtering outside of the audio spectrum but we can use algorithms like “Apodization” that are quite efficiently used within different fields of optics like microscopy and astronomy.
_________________________________________________________________

And, more SHOCKINGLY:

__________________________________________________________________
APODIZATION IN THE WHOLE RECORDING AND PLAYBACK CHAIN

Figure 5 represents the full recording and playback chain, including the analog to digital and digital to analog processing but no apodization.
The resulting output signal (pls. see figure 6) exhibits strong pre- and post-ringing.

In figure 7 the recorded signal is apodized by applying the XiPodizer to reduce the effect of pre-ringing.

The XiPodizer is very effective in suppressing the pre-ringing (pls. see figure 8) by compromising a bit of the available audio bandwidth.
__________________________________________________________________

Is this FRAUD? Should I report them? I think I am being CONNED!

 

[Shadders]

Likewise, I have a link for your enjoyment - your best friends at Xivero make a very nice apodization software and appear to be fully backing the overall "elimination of the pre-ringing" approach in the entire recording/playback chain.

https://www.xivero.com/xipodizer/

XIPODIZER
Remove pre-ringing caused by steep linear phase antialiasing filters

Hi,
Yes, but CD's are correctly decimated/downsampled and there is no ringing on the CD. You have stated there is, so please provide an example, such as the artist, title, pressing date etc

The package will only reduce the ringing for your specific DAC depending on the filter implemented in your DAC.

I checked the Hifi News high resolution spectrums - and the HF content at 20kHz is -80dB. So, even if there is ringing, the peak amplitude of the ringing is at a peak of -92dB. This is at the noise floor of a CD recording all bar 4dB. Essentially it could be considered in the dither noise.

Many of the latest DAC IC's have minimum phase filters as a selection - they have slow roll off and fast roll off linear phase too. You should be able to select at a push of a button which filter you want.

We do not need MQA.

Regards,
Shadders.

 

[DimitryZ]

Here we go again, now XIVIERO is no good! I though their anti-MQA paper was just great - everything there was top notch. Now, they jump on the apodization bandwagon and embrace the whole "Ringing CD" thing, they are in the doghouse? There is no pleasing you, is there?

You mean to say "I don't need MQA". I completely agree - you should listen to strictly LPCM. Have you heard, that listening to MQA, according to latest double blind tests, will actually make you loose 10 points of you IQ and get an ulcer?

 

[Tin]

Here we go again, now XIVIERO is no good! There is no pleasing you, is there?

Actually, it seems that Shadders is very easy to please; he has 'liked' a lot of the comments by Jim, me and several others. If that isn't a token of being pleased, I don't know what is.

The only one who constantly tries to start a fight is you. Hating Open Source activists, hating Zealots for trying to kick out the Romans, hating your professors, hating it when people offer you a way out.

 

[Shadders]

Here we go again, now XIVIERO is no good! There is no pleasing you, is there?

You mean to say "I don't need MQA". I completely agree - you should listen to strictly LPCM. Have you heard, that listening to MQA, according to latest double blind tests, will actually make you loose 10 points of you IQ and get an ulcer?

Hi,
I merely pointed out that it will only assist those people who have a slow/fast roll off linear phase filter in their DAC.

I showed that the energy at 22.05kHz is so low, and that the peak impulse value is 12dB below that. This means that the other artefacts for the ringing are even lower - in the noise (guaranteed).

What this all means is that ringing from a DAC filter is NOT an issue - even when it does occur.

Regards,
Shadders.

 

[DimitryZ]

Hi,
I merely pointed out that it will only assist those people who have a slow/fast roll off linear phase filter in their DAC.

I showed that the energy at 22.05kHz is so low, and that the peak impulse value is 12dB below that. This means that the other artefacts for the ringing are even lower - in the noise (guaranteed).

What this all means is that ringing from a DAC filter is NOT an issue - even when it does occur.

Regards,
Shadders.

Yes, that's what folks at CA say as well - too low to hear, no worries. And yet people seem to hear either the effect itself or the combination of ultrasonic and high frequency modulation, or something else.

 

[Shadders]

Yes, that's what Miska says as well - too low to hear, no worries. And yet people seem to hear either the effect itself or the combination of ultrasonic and high frequency modulation, or something else.

"Miska" means "bowl" in Russian...I wonder if he/she knows that.

Hi,
It is possible that the placebo effect is occurring ?.

The Xivero will slowly roll off your high end frequencies - and this is preferred (apparently) by humans. So, maybe people don't like the high frequencies ?

All we can do is ensure a linear system - which means linear phase filters.

Regards,
Shadders.

 

[DimitryZ]

Hi,
It is possible that the placebo effect is occurring ?.

The Xivero will slowly roll off your high end frequencies - and this is preferred (apparently) by humans. So, maybe people don't like the high frequencies ?

All we can do is ensure a linear system - which means linear phase filters.

Regards,
Shadders.

I assume you include yourself in this "humans" category?

But seriously, I think we made tremendous progress today. I believe I am sensing that you agree there is a reason the DAC manufacturers are including numerous reconstruction filters for their customers to choose for themselves. This would indicate that at least some humans can actually hear the differences between different DAC filters - the sound they produce is within a margin of preference.

And if this is the case (humans can hear different DAC filters), it would stand to reason that they would also be able to hear ADC filters. This would of course lead to all kinds of confusing and radical developments and open a bevy of possibilities that listeners could find this or that approach to sound reproduction that they prefer, instead of relying on measurements - that assured them at least 30 years ago that digital audio has reached perfection and anything they hear that is not perfect is bias and "placebo effect."

 

[Shadders]

I assume you include yourself in this "humans" category?

But seriously, I think we made tremendous progress today. I believe I am sensing that you agree there is a reason the DAC manufacturers are including numerous reconstruction filters for their customers to choose for themselves. This would indicate that at least some humans can actually hear the differences between different DAC filters - the sound they produce is within a margin of preference.

And if this is the case (humans can hear different DAC filters), it would stand to reason that they would also be able to hear ADC filters. This would of course lead to all kinds of confusing and radical developments and open a bevy of possibilities that listeners could find this or that approach to sound reproduction that they prefer, instead of relying on measurements - that assured them at least 30 years ago that digital audio has reached perfection and anything they hear that is not perfect is bias and "placebo effect."

Hi,
Possibly - not sure why they have done so - but maybe MQA has prompted them to offer more options. The latest TI DAC IC also has a mini DSP core, so you may be able to implement your own. Not sure if the core is floating point or fixed point processing.

The differences in the filters are that they are either linear phase or minimum phase, which have different roll off's from 0.40fs to 0.47fs. So for the 0.40fs, this is the roll off where more of the higher frequencies (earlier) are attenuated - which maybe the people prefer ?

For ADC's, their sample rate is much higher such that any signal being processed is far below the filter cut off frequency. That is, there are no filter artefacts to be heard. Then again, the delta-sigma ADC's are always being refined, but for audio - not sure if any changes can be heard.

Regards,
Shadders.

 

[misterdog]

I have an audiophile friend, who also records a lot of music using purist tecording technique. He tells me that his initial file sounds great, but are audibility degraded witj every stage of his workflow.

Did he fall for that old chestnut about 8 track cassette being as close to the master tape as standing on it, and that the ferric oxide used on the tape was applied by the musicians themselves.

He could have had all the ADC errors in his 'studio' corrected by MQA - surely.

Did you not feel the need to enlighten him, and show him the path to true righteousness.

 

[misterdog]

I got mostly "Arse"

We can see why.

 

[flatpopely]

<moderating>

I’m going to sit on my hands for sometime while I think what to do with this thread.
It’s getting very cyclical and personal.