The sound quality of DSD?

[analogSurviver]

What leads you to believe that this has anything to do with ultrasonics though? There are about a million issues coming in before we even get to this point - trying to capture an uneven radiation pattern with microphones and then squeezing everything into two channels to be played back by two speakers placed in front of the listener poses a great number of challenges and is by no means lossless, not to mention dependent on playback system quality to boot. Seems you are worrying about a fly while ignoring the elephant in the room.

Aha - here is the "missing" AnalogSteph. Sorry, I apologize - I have been too short a time on ASR to remember all the names.

What has led me to this belief are analogue records and their playback. The playback equipment can have a humongous effect on the sound ultimately obtainable. There was not a single case where a narrower bandwidth device with equal/similar other characteristics of a more broadband competitor came on top - in over 40 years.

What leads me to this belief is the experience of doing recordings - mostly live. And what leads me to believe this even more is doing live binaural recordings - which is the most close approach to recording to be listened by humans. You are not going to hear anything closer to live than good binaural recording - and the more extended the response of the recording, the more realistic the acoustics of the venue are depicted.

 

[Blumlein 88]

Aha - here is the "missing" AnalogSteph. Sorry, I apologize - I have been too short a time on ASR to remember all the names.

What has led me to this belief are analogue records and their playback. The playback equipment can have a humongous effect on the sound ultimately obtainable. There was not a single case where a narrower bandwidth device with equal/similar other characteristics of a more broadband competitor came on top - in over 40 years.

What leads me to this belief is the experience of doing recordings - mostly live. And what leads me to believe this even more is doing live binaural recordings - which is the most close approach to recording to be listened by humans. You are not going to hear anything closer to live than good binaural recording - and the more extended the response of the recording, the more realistic the acoustics of the venue are depicted.

Where do you get the ideas about needing 100 khz?

 

[CDMC]

Isn't it only the local bats that get the extra satisfaction from the live performance?

And run from Class D and digital sources due to the harshness of the recordings and ultrasonic output. We will of course ignore that virtually no microphones pick up past 40khz, nor do any speakers reproduce that.

Not sure how people make the wild leap from the knowledge that we can’t hear or sense ultrasonic frequencies to it having some magical effect on music reproduction. I would love see someone go argue that video reproduction is inadequate because the monitor doesn’t reproduce ultraviolet, xray, and gamma rays.

 

[BDWoody]

And run from Class D and digital sources due to the harshness of the recordings and ultrasonic output. We will of course ignore that virtually no microphones pick up past 40khz, nor do any speakers reproduce that.

Not sure how people make the wild leap from the knowledge that we can’t hear or sense ultrasonic frequencies to it having some magical effect on music reproduction. I would love see someone go argue that video reproduction is inadequate because the monitor doesn’t reproduce ultraviolet, xray, and gamma rays.

I asked @Dougey_Jones in another thread what studies he was referencing when he was making similar claims re: the benefits of Ultrasonics, and he responded with these:

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0095464

https://www.frontiersin.org/articles/10.3389/fpsyg.2017.00093/full

https://www.researchgate.net/public...uency_Sounds_Affect_Brain_Activity_Hypersonic

in this post: https://www.audiosciencereview.com/...sl-su-9-balanced-dac-review.16150/post-535791


I'm not sure what to make of them...but it looks to this neuroscience layman that there is something happening, and these don't look like the typical half-assed 'studies' most like to cite. I would like to understand these a bit more...

 

[Killingbeans]

Now we're getting somewhere! So, ultrasonics affect us. But do we perceive it as information, or just as stimulus?

In other words, if we simply turn on a mosquito repeller in the background, will it give the same effect?

 

[Blumlein 88]

http://www.sengpielaudio.com/calculator-air.htm

3.28 db/meter absorbed at 100 khz. So 100 db loss in 100 feet.

 

[Blumlein 88]

I asked @Dougey_Jones in another thread what studies he was referencing when he was making similar claims re: the benefits of Ultrasonics, and he responded with these:

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0095464

https://www.frontiersin.org/articles/10.3389/fpsyg.2017.00093/full

https://www.researchgate.net/public...uency_Sounds_Affect_Brain_Activity_Hypersonic

in this post: https://www.audiosciencereview.com/...sl-su-9-balanced-dac-review.16150/post-535791


I'm not sure what to make of them...but it looks to this neuroscience layman that there is something happening, and these don't look like the typical half-assed 'studies' most like to cite. I would like to understand these a bit more...

The Oohashi sourced material has failed to replicate more than once. Even the second citation mentions this.

The testing done in that second citation is quite interesting.

However the speaker they used only claims response to 22 khz. It is similar to the B&W Amir recently reviewed. The speaker was powered by a modest Teac receiver using its built in DAC. From what we have seen of testing such devices one wonders what its response was.

Still an interesting result and would be nice if it were replicated a couple or three times as the result is so unusual. How does ultrasonic sound effect the listener if the speaker never produces it?

 

[analogSurviver]

And run from Class D and digital sources due to the harshness of the recordings and ultrasonic output. We will of course ignore that virtually no microphones pick up past 40khz, nor do any speakers reproduce that.

Not sure how people make the wild leap from the knowledge that we can’t hear or sense ultrasonic frequencies to it having some magical effect on music reproduction. I would love see someone go argue that video reproduction is inadequate because the monitor doesn’t reproduce ultraviolet, xray, and gamma rays.

Not sure under which rock exactly you have been for the past 40 ( ! ) years ... - because late 70s-early eighties, Technics did have an entire chain, from phono cartridge to speaker, capable of clean performance past 100 kHz - latest version did exceed 120 kHz mark. At least 38 years ago - when the CD toddler still did not learn even how to walk...
There are tons and tons of speakers that reproduce past 40 kHz ( Infinity EMIT tweeter and a slew of its knock-offs - to mention but the most numerous group that in no way can be labeled as esoteric rarity ) - and at least some dozen microphones that are in normal studio usage for years, if not decades. You can ( or could... - did not check lately ) get back then CDs, later download samples from Earthworks' page - where they have recordings of the same music with their range of mics - with flat response to 25, 30, 40 and 50 kHz respectively. And, since microphone is an analog device, it can not and does not mimic brickwall filtering of PCM digital - but it has a far gentler rolloff above maximum specified flat response range. So, a 50 kHz flat response mic will still have usable response to about 70 kHz - if not more. There are 40 kHz flat DPA (Bruel &Kjaer ) mics, Neumanns, Schoeps, Gefell, Josephson, etc, etc. And I did give the link for the Audio Precision 100 kHz mic - didn't I ? Here is also the first mic to reach 100 kHz specifically intended for music recording - Sanken C100K
https://www.sanken-mic.com/en/product/product.cfm/3.1000400 It is being used for quite some time now ... - by those who actually do follow the progress in audio.

Class D amps have no place in any serious audio application - save their use for subwoofers, where all the ultrasonic garbage they put out is filtered purely mechanically. And the attempt of having clean response in digital past 100 kHz will - eventually - lead to totally clean response from digital to 50 kHz flat response and no spuriae say above -100 dB to whatever the highest frequency it can still put out - at a REASONABLE price. In devices most custumers will not even know that they offer such high performance - because by then, such performance should and would become the norm.

I wonder why people do not object lavish amounts of resources and money being put into say F1 racing. There is no way that kind of cars will ever be allowed on public roads in normal traffic... - yet people can see the things that did trickle down from motorsport can be usable in (better/costlier) everyday cars.

Why is advancing digital today to the levels achieved by analog 40 years ago such a boogaboo for some/most people ?

 

[analogSurviver]

The Oohashi sourced material has failed to replicate more than once. Even the second citation mentions this.

The testing done in that second citation is quite interesting.

However the speaker they used only claims response to 22 khz. It is similar to the B&W Amir recently reviewed. The speaker was powered by a modest Teac receiver using its built in DAC. From what we have seen of testing such devices one wonders what its response was.

Still an interesting result and would be nice if it were replicated a couple or three times as the result is so unusual. How does ultrasonic sound effect the listener if the speaker never produces it?

The problem I see here was the "sponsorship" - by TEAC. They most likely had to use exclusively TEAC equipment, which unfortunately did not include a speaker with well specified response to at least 40 kHz.

The other way around is also possible - microphone manufacturers lately started to claim significantly LOWER frequency response extension in their specs than of which their mics are actually capable of.
Just to have the peace from people complaining WHY must a mic have response past 20 kHz ...

 

[Killingbeans]

You can ( or could... - did not check lately ) get back then CDs, later download samples from Earthworks' page - where they have recordings of the same music with their range of mics - with flat response to 25, 30, 40 and 50 kHz respectively.

But a CD holds no information above 22kHz?

 

[Blumlein 88]

snip....
The other way around is also possible - microphone manufacturers lately started to claim significantly LOWER frequency response extension in their specs than of which their mics are actually capable of.
Just to have the peace from people complaining WHY must a mic have response past 20 kHz ...

Never once seen anyone complain about mics with more than 20 khz response. The idea mic makers would under-rate bandwidth for this reason is preposterous.

It is true microphones usually don't have steep cut offs. Many condenser mics have response of some reasonable sort up to 30-35 khz.

 

[Wes]

My bat colony loves this thread!

 

[andymok]

Not sure under which rock exactly you have been for the past 40 ( ! ) years ... - because late 70s-early eighties, Technics did have an entire chain, from phono cartridge to speaker, capable of clean performance past 100 kHz - latest version did exceed 120 kHz mark. At least 38 years ago - when the CD toddler still did not learn even how to walk...
There are tons and tons of speakers that reproduce past 40 kHz ( Infinity EMIT tweeter and a slew of its knock-offs - to mention but the most numerous group that in no way can be labeled as esoteric rarity ) - and at least some dozen microphones that are in normal studio usage for years, if not decades. You can ( or could... - did not check lately ) get back then CDs, later download samples from Earthworks' page - where they have recordings of the same music with their range of mics - with flat response to 25, 30, 40 and 50 kHz respectively. And, since microphone is an analog device, it can not and does not mimic brickwall filtering of PCM digital - but it has a far gentler rolloff above maximum specified flat response range. So, a 50 kHz flat response mic will still have usable response to about 70 kHz - if not more. There are 40 kHz flat DPA (Bruel &Kjaer ) mics, Neumanns, Schoeps, Gefell, Josephson, etc, etc. And I did give the link for the Audio Precision 100 kHz mic - didn't I ? Here is also the first mic to reach 100 kHz specifically intended for music recording - Sanken C100K
https://www.sanken-mic.com/en/product/product.cfm/3.1000400 It is being used for quite some time now ... - by those who actually do follow the progress in audio.

Class D amps have no place in any serious audio application - save their use for subwoofers, where all the ultrasonic garbage they put out is filtered purely mechanically. And the attempt of having clean response in digital past 100 kHz will - eventually - lead to totally clean response from digital to 50 kHz flat response and no spuriae say above -100 dB to whatever the highest frequency it can still put out - at a REASONABLE price. In devices most custumers will not even know that they offer such high performance - because by then, such performance should and would become the norm.

I wonder why people do not object lavish amounts of resources and money being put into say F1 racing. There is no way that kind of cars will ever be allowed on public roads in normal traffic... - yet people can see the things that did trickle down from motorsport can be usable in (better/costlier) everyday cars.

Why is advancing digital today to the levels achieved by analog 40 years ago such a boogaboo for some/most people ?

I see Michael Bishop mentioned transients. May I ask what is the importance/significance of bandwidth to transients that we hear/perceive, in what way?

 

[analogSurviver]

But a CD holds no information above 22kHz?

THAT is THE biggest "clincher"; even on as bandwidth restricted medium as CD, the difference(s) among mics that are practically identical in 20-20kHz band, are CLEARLY audible.
Those CDs date from about a decade ago - I still have them somewhere.

Now imagine how would they sound on an frequency wise unrestricted recording ... - or at least one good to 100 kHz.

 

[Blumlein 88]

THAT is THE biggest "clincher"; even on as bandwidth restricted medium as CD, the difference(s) among mics that are practically identical in 20-20kHz band, are CLEARLY audible.
Those CDs date from about a decade ago - I still have them somewhere.

Now imagine how would they sound on an frequency wise unrestricted recording ... - or at least one good to 100 kHz.

It's a clincher, just maybe not the one you think it is.

 

[analogSurviver]

I see Michael Bishop mentioned transients. May I ask what is the importance/significance of bandwidth to transients that we hear/perceive, in what way?

Correct. Simple - the higher the frequency response, the better the transient.

The fastest audio device I have had the privilege to hear has been the prototype phono cartridge that later went into production as Benz Micro Switzerland Ruby. It had a really unusable output - 0.03mV/5cm/sec - and no phono input could manage to eek an usable S/N (particularly hum... ) out of it. But , its rise time has been 3 ( in a word : three ) microseconds.
Listening to music off normal analogue records with that prototype and then switching to any "normal" cartridge had about the same effect as switching between the same radio station - from FM to AM.
The speakers used have been Magnaplanar ??? - the big ones with the full height length ribbon tweeter and electronics with MHz bandwidth.
It was the only system I ever heard/felt/perceive capable of conveying the difference of speed of sound propagation in solids and gas; you could feel the whack on tympany with your feet/body BEFORE you could hear it ... - just as in real life. The sound namely travels much faster in solids (concrete floor, wood, carpet in this case ) than in the air.

 

[PaulD]

Correct. Simple - the higher the frequency response, the better the transient.

The fastest audio device I have had the privilege to hear has been the prototype phono cartridge that later went into production as Benz Micro Switzerland Ruby. It had a really unusable output - 0.03mV/5cm/sec - and no phono input could manage to eek an usable S/N (particularly hum... ) out of it. But , its rise time has been 3 ( in a word : three ) microseconds.
Listening to music off normal analogue records with that prototype and then switching to any "normal" cartridge had about the same effect as switching between the same radio station - from FM to AM.
The speakers used have been Magnaplanar ??? - the big ones with the full height length ribbon tweeter and electronics with MHz bandwidth.
It was the only system I ever heard/felt/perceive capable of conveying the difference of speed of sound propagation in solids and gas; you could feel the whack on tympany with your feet/body BEFORE you could hear it ... - just as in real life. The sound namely travels much faster in solids (concrete floor, wood, carpet in this case ) than in the air.

Ah huh... So, confirmation bias then? The listening test would need better controlled to be worth reporting as authoritative.

I have listened to a timpani, and kit drums (and glocks and a bunch of other transient instruments), played in a studio and in a concert hall, up close and in the audience, when recording. Then after positioning mics I have listened in the control room, and the transients sound the same, minus any microphone colouration. Then it has been run through a decent ADA process through the recorder and back into the control room, and it sounds the same, coming back from the disk. Yes yes with fancy BS mics as well as just sensible mics (my favourites are the relatively unpopular DPA4060s etc., flat to well past hearing). No one needs MHz bandwidth in audio, and people who have claimed that have failed to prove it in blind tests.

We can measure everything that can be heard.
We can measure well beyond the limits of hearing.

 

[analogSurviver]

Ah huh... So, confirmation bias then? The listening test would need better controls to be worth reporting as authoritative.

I have listened to a timpani, and kit drums (and glocks and a bunch of other transient instruments), played in a studio and in a concert hall, up close and in the audience, when recording. Then after positioning mics I have listened in the control room, and the transients sound the same, minus any microphone colouration. Then it has been run through a decent ADA process through the recorder and back into the control room, and it sounds the same, coming back from the disk. Yes yes with fancy BS mics as well as just sensible mics (my favourites are the relatively unpopular DPA4060s etc., flat to well past hearing). No one needs MHz bandwidth in audio, and people who have claimed that have failed to prove it in blind tests.

We can measure everything that can be heard.
We can measure well beyond the limits of hearing.

Haha - we share our favourite mics - DPA4060s ! In fact, I use exclusively 406x family, and am considering getting yet another pair of updated version, called Core 406x. Improvement across the board, from feedback from users most notably in the bass - which, although decent, is not exactly stellar on - now called vintage - 406x series.
Now, please check out your own 4060s for just how far above 20 kHz they can still deliver ... - and you are most likely to run in digital high frequency noise, before being able to find just what is the limit of the microphone.

For the MHz bandwidth to be effective/audible in audio , ENTIRE chain ( down to the last capacitor in the equipment ) has to support it. In most cases, this can not be achieved - and the last place that is likely to have that is a recording studio.

 

[BDWoody]

Now imagine how would they sound on an frequency wise unrestricted recording ... - or at least one good to 100 kHz

Actual controlled tests to provide any support?

Otherwise, Imagining is all I imagine anyone could do.

 

[Mnyb]

Does anything of this really has anything to do about DSD, high bandwith can be achived with PCM as well ? 192/24 pcm for example

Afaik freuency response and impulse response are eqvavilent , so if we cant hear past 20k we cant percive transient faster than that either , its two sides of the same coin ? is it not ?
(please dont respond with 1/44,1k argument and "timing" , to show basic missunderstanding of sampling please )