[old tech]

So...have eaten a fair amount of popcorn watching this debate

Which is at times civilized/constructive and at other times arrogant/condescending/demeaning/etc. (a shame really!)

I've got no dog in this fight and the vast majority of my music library is at RBCD level (primarily FLAC). FWIW, I don't see the value in spending money for higher bit rates...but am always open to being convinced by a rational argument. That said, I will spend money for a better/different version of an artist's work.

All that aside, one complaint/dig against the M&M study that seems to pop up from the hi-rez advocates is that the source material used for their tests was not truly hi-rez. I'd be curious to hear this thread's take on that (i.e. if it's true, and if so, how it impacts the validity/results of the study).

If I missed it in a previous post, I apologize...thanks!

Yes there is some validity to that criticism but it sort of validated the study further. It came out later that many (but not all) the SACDs and DVD-As were not hi res, just upsampled from the original source. However none of the subjects identified the ones that that were not upsampled, which should have stood out if the higher sample rates/bit depths actually made a difference. The other, probably more powerful evidence was that these "non hi res" SACDs were for several years released in markets all over the world yet no one noticed, not one audiophile or golden eared subjective reviewer. In the end the "fake hi res" SACDs was confirmed by testing rather than listening.

There is also another criticism of M&M which transcends the test put forward by the likes of Waldrep and others, and that is that the term Hi Res is largely a marketing gimmick as true high res requires 24/96 all the way through the chain, from recording, storage to playback. Analog and CD masters can never be considered Hi Res just because it is in a high res bucket. Even so, I am unaware of any convincing test which has established listeners can hear a difference between true hi res with the same material downsampled to 16/44.

 

[Phronesis]

Just had a quick read of that paper and had a good laugh. The author starts off with a limited understanding of digital and analog recordings, eg he totally misunderstands sampling, eg the comment about inaccuracies at high frequencies with 44.1, he seems to have picked up a myth book somewhere, and his implication that analog cannot record ultrasonics, maybe LPs cannot but 15 ips reel to reel on 1" tape can.

Anyway, as for the test, I haven't read his methodology in detail but instead looked at his references and there they are, using research from people such as Oohashi who also claims something similar through bone conduction, yet his research has been thorougly discredited by peer review.

Has this paper been subject to peer review? Like Oohashi and others of that ilk, I suspect it would be thoroughly discredited as well.

Why don't these researchers who make the totally absurd claim that ultrasound in recording and playback affects our hearing of music ever look at the M&M paper which has been subjected to over 10 years of peer review scrutiny? I think we all know the answer to that.

I would guess that it was peer-reviewed. Here's the author's website: http://www.people.vcu.edu/~lenhardt/Pages/Main.html. On this topic, I'm guessing that he's generally more qualified than anyone in this forum, but regardless, a proper thorough evaluation of this paper by qualified people would be needed to judge it. I don't have those quals, but will take a look at the paper anyway.

 

[GearMe]

Yes there is some validity to that criticism but it sort of validated the study further. It came out later that many (but not all) the SACDs and DVD-As were not hi res, just upsampled from the original source. However none of the subjects identified the ones that that were not upsampled, which should have stood out if the higher sample rates/bit depths actually made a difference. The other, probably more powerful evidence was that these "non hi res" SACDs were for several years released in markets all over the world yet no one noticed, not one audiophile or golden eared subjective reviewer. In the end the "fake hi res" SACDs was confirmed by testing rather than listening.

There is also another criticism of M&M which transcends the test put forward by the likes of Waldrep and others, and that is that the term Hi Res is largely a marketing gimmick as true high res requires 24/96 all the way through the chain, from recording, storage to playback. Analog and CD masters can never be considered Hi Res just because it is in a high res bucket. Even so, I am unaware of any convincing test which has established listeners can hear a difference between true hi res with the same material downsampled to 16/44.

Thanks for the reply!

Makes sense...you brought up another related concern that I was going to ask about as well...basically the recording/playback chain is only as strong as its weakest link.

Follow on question...why is the hi-rez source material downsampled instead of the low-rez source material upsampled and then the test run by playing back through that resolution?

 

[james444]

... using research from people such as Oohashi who also claims something similar through bone conduction, yet his research has been thorougly discredited by peer review.

Has this paper been subject to peer review? Like Oohashi and others of that ilk, I suspect it would be thoroughly discredited as well.

From what I've been gathering, it's just Oohashi's conclusions regarding listener preference that have been discredited. But not his findings about physiological brain response to high-frequency audio. If you have more in-depth informaton, I'd appreciate a link to the source.

IMHO, it's an entirely different question to ask whether an ultrasound stimulus will trigger a human brain response in the auditory cortex, or whether we can distinguish Hi Res from RBCD in a conscious listening test.

FWIW, here's the most comprehensive summary on the former question that I've found:

Spoiler: Perception of Ultrasonic Sounds

12. Perception of Ultrasonic Sounds
The frequency range of human hearing is normally defined as extending from 20 Hz to 20 kHz
(Newby & Popelka, 1985). This range is commonly referred to as the sonic range. The lower limit
of hearing (20 Hz) is defined by the lowest frequency at which the listener hears one continuous
sound. The upper limit of hearing (20 kHz) is defined by the highest frequency at which the
listener still has an auditory sensation, regardless of sound intensity. The highest frequency
perceptible differs greatly among individuals and is difficult to determine because some high
frequency sounds can cause a painful or a tactile sensation but not an auditory sensation. In
addition, some investigators have noted that the operational (normal) range of human hearing may
be extended to frequencies beyond 20 kHz when the ear is stimulated by bone conduction as
opposed to air conduction. Ultrasonic frequencies refer to frequencies above the range of air
conduction hearing (greater than 20 kHz), and the human ability to hear sounds in this frequency
range is normally referred to as ultrasonic hearing.
Although air-conducted sounds cannot be heard at frequencies above 20 kHz (Wever, 1949),
ultrasonic hearing as high as 100 kHz has been demonstrated through bone conduction stimulation.

150
Ultrasonic hearing has been found to be capable of supporting frequency discrimination and speech
detection in normal and older people with severe and profound hearing loss. One of the first
groups of investigators to report perception of ultrasonic frequencies by bone conduction demon-
strated this phenomenon by generating ultrasound waves in water (Deatherage et al., 1954). First,
auditory perception was noted when a listener’s jaw bone was placed in contact with a container
filled with water in which a transducer produced a signal of 50 kHz. The threshold for this signal
was approximately 2000 dynes/cm2, which equates to about 140 dB SPL. Second, auditory
perception was demonstrated through submersion of the listener in a container of water containing
the transducer. In this condition, the threshold was approximately 1000 dynes/cm2 at 50 kHz,
which equates to about 134 dB SPL. To support their claims that the vibrations were being
perceived by bone conduction, Deatherage et al. (1954) showed perception of a 7-kHz signal in
a body of water at a threshold of 12 dynes/cm2, which agreed well with previous bone conduction
threshold data obtained by Bekesy using direct mechanical stimulation.
Since this early report of auditory perception of ultrasonic stimuli delivered through bone
conduction, several investigators have pursued measurement of the human auditory system’s
sensitivity and discrimination ability in the ultrasonic range. For example, Corso (1963) evaluated
high frequency sensitivity to bone-conducted sounds by people with normal air conduction hear-
ing. Placing vibrators on the mastoid bone, Corso measured bone conduction thresholds for
frequencies between 6 and 95 kHz and reported good sensitivity for frequencies below 14 kHz and
poor or no sensitivity to sounds of frequencies between 20 and 95 kHz. In contrast, a later study
demonstrated that the sounds in the ultrasonic range can be heard by listeners (Lenhardt, Skellet,
Wang, & Clarke, 1991).
The mechanism through which ultrasonic sound is perceived by the listener is not known, although
several theories exist. These theories include
• Perception by the saccule within the vestibular system,
• Demodulation of the ultrasonic stimulus through the bones of the skull, which is then
perceived by the cochleae,
• Direct stimulation of the brain matter and cerebrospinal fluid, and
• Direct stimulation of the cochleae through the brain.
The first theory is that the bone-conducted sound is perceived by the saccule, one of the three
vestibular canals present in the inner ear, as demonstrated through the perception of ultrasound in
people with nonfunctional cochleae (Lenhardt et al., 1991). Figure 79 is a diagram of the cochlea
for review. In order to accept the saccule theory, the traditional pathways need to be eliminated
first (Dobie, Wiederhold, & Lenhardt, 1992). In support of this effort, several investigators have
demonstrated that ultrasonic stimulation through bone conduction cannot be masked through air
conduction which leads away from a cochlear-based process. Furthermore, ultrasonic signals
presented through bone conduction cannot be measured in the EAC (Staab et al., 1998).


151
A second theory about the perception of ultrasound is that the bone conduction process is suffi-
ciently nonlinear to demodulate the signal (Dobie et al., 1992; Lenhardt et al., 1991). Demodula-
tion refers to the perception of a frequency that is within the audible range that represents the
fluctuations of the carrier signal. According to this theory, the presentation of ultrasonic vibrations
to the skull does not allow for transfer of these vibrations through the middle ear to the cochlea but
to the cochlea directly. The cochlea then demodulates the signal into a range where it can be
heard.













Figure 79. Spatial relationship between the structures of the vestibular system,
the cochlea, and the celebrospinal fluid (CSF) spaces (Salt, 1996).
Several investigators have argued against the theory of the bone conduction pathway for ultra-
sonic perception following an examination of ultrasonic stimulation through the use of magneto-
encephalography (MEG) (Hosoi et al., 1998). In this procedure, areas of the brain are examined
through a scanning device to determine where neurons are activated in response to a particular
stimulus. Hosoi and colleagues (1998) stimulated listeners who had normal hearing with ultra-
sonic sounds by placing a vibrator on their sternocleidomastoid muscle (between the neck and
shoulder). They found brain activity in response to these stimuli in the auditory cortex. This was
true for people with normal hearing and for those with profound hearing loss. Imaizumi et al.
(2001) found the same results using positron emission tomography scans. Again, stimulation in
people with normal hearing or profound hearing loss resulted in activation of the auditory cortex.
This activation occurred through stimulation by air conduction, bone conduction, ultrasound and
vibro-tactile methods (Imaizumi et al., 2001). Regardless of the pathway or mechanisms behind
the phenomenon, it has been demonstrated that ultrasound can be perceived by the listener when
vibrations are applied directly to the human head or neck.
When an ultrasonic carrier signal, which is presented to the listener through bone conduction, is
amplitude modulated by a speech signal, the result is a clear perception of the speech stimuli and
not a sense of high-frequency vibration. In a study by Lenhardt and colleagues (1991), speech
recognition rates through this method for people with normal hearing were on the order of 83%

152
for the WIPI (word-identification through picture identification) task. For people with profound
hearing losses (pure tone averages of greater than 90 dB HL), performance in the same test was
between 20% and 30%. These results support the belief that ultrasonic hearing may be used as a
communications channel and thus as a non-surgical approach in the rehabilitation of profound
hearing losses. The results also imply that ultrasonic stimulation for speech communication may
be applicable for people with normal hearing.
An amplification device using ultrasonic stimulation through bone conduction called HiSonic was
developed by a group of investigators in Arizona (Staab et al., 1998). This device was developed
for use with people with profound hearing loss for whom standard hearing aids would not provide
sufficient amplification. Such a device would allow a non-surgical alternative to cochlear implants
as a hearing solution (see section 10). The device consists of a bone conduction vibrator that is
positioned on the mastoid bone of the wearer. In the Staab et al. (1998) study, pure tone stimuli in
the range of 500 to 2000 Hz were shifted in frequency to the ultrasonic range, and thresholds were
measured for listeners with normal hearing as well as those with profound hearing loss. For the
people with profound hearing loss, a comparison of thresholds measured with and without the
ultrasonic shift showed a clear advantage of using the HiSonic device in that thresholds obtained
with the device were considerably lower than those obtained without the device. Although there
was considerable variability among the performance of the individual listeners, benefit through use
of the device was demonstrated in 65% to 70% of the participants (Staab et al., 1998).
The third theory about a mechanism for ultrasonic perception is that the skull vibrations are
being transmitted directly to the brain and surrounding CSF, including direct stimulation of the
auditory cortex. Skull vibrations can be transmitted to the cochlear fluids and to the non-com-
pressible brain matter and the surrounding CSF. The resulting changes in fluid pressure can be
transmitted through the internal auditory meatus and cochlear aqueduct to the perilymph of the
scala tympani or through the vestibular aqueduct to the endolimphatic sac of the vestibular
system and further to the saccule (see figure 79). This mechanism of bone conduction was
proposed as an alternate pathway explaining the presence of auditory sensation during direct
stimulation of brain tissue and the CSF. However, the hypothesis of the vibration transmission
from the CSF to the cochlear fluids is contradicted by relatively high mechanical damping of
structures (neurons, blood vessels, connective tissue) occupying the aqueducts (Bystrzanowska,
1963, p. 19).
The fourth theory about the mechanism for ultrasonic perception is a mechanical conduction of
sound to the cochlea proposed by Freeman et al. (2000) and Sichel, Freeman, and Sohmer
(2002). This mechanism involves direct excitation of the CSF in the skull cavity. The CSF is
the watery fluid that occupies the spaces around the brain and the spinal cord providing shock
absorption protection to these organs. During some conditions, this fluid can enter the cochlea
from the subarachnoid space that is connected to the scala tympani through the cochlear aque-
duct (diameter ˜0.5 mm). The investigators in these two studies directly vibrated the brain matter
of rats, guinea pigs, and fat sand rats and demonstrated that direct stimulation of the brain


153
through a bone oscillator resulted in measurable ABRs. Progressive elimination of potential
bone conduction mechanisms (i.e., ossicular chain, meatus, and skull through craniotomy) did
not result in complete elimination of the ABR. This finding supports the notion that an auditory
sensation can arise from direct vibration of the brain and CSF. The difference in the inter-
pretation of the mechanism involved is in the identification of the location of perception. The
hypothesis that auditory stimulation of the brain and CSF is possible is also supported by some
results of human studies (Sohmer et al., 2000). The studies did not eliminate direct stimulation
of the auditory cortex as was theorized by others but theorized that the stimulation was directly
detected by the cochleae.
The use of ultrasonic stimulation is not without its contraindications. Several investigators have
noted tinnitus of several days’ duration as a side effect associated with the ultrasonic stimulation,
which could be an early sign of hearing damage (Corso, 1963; Deatherage et al., 1954). This
side effect is not to be taken lightly. In order to consider the use of ultrasonic stimulation for any
person, the safety of that provision must be carefully considered. To our knowledge, there are no
reports about the safety aspects of the provision of ultrasonic stimulation of the bones of the
skull. Until those data are made available, long-term use of ultrasonic stimulation should be
avoided.
In summary, there is evidence to suggest that ultrasonic frequencies can be perceived as sound
by the listener when they are transmitted through bone conduction vibration on the skull. The
pathway for such perceptions is not clear, but four theories have been proposed, ranging from
demodulation by the auditory mechanism to the detection of sound by the CSF within the skull
cavity. Caution should be used when one is attempting to implement ultrasonic stimulation since
there is little known regarding the safety limitations. There have been some reports of the onset
of tinnitus following stimulation by ultrasound.

(source: link; warning: huge document)

 

[bfreedma]

https://pdfs.semanticscholar.org/2702/810e3bfde17fc044cd7775b49e9c9a77ee8f.pdf

Thanks! Will read it when I have time to focus and digest the contents.

 

[castleofargh]

Thanks for the reply!

Makes sense...you brought up another related concern that I was going to ask about as well...basically the recording/playback chain is only as strong as its weakest link.

Follow on question...why is the hi-rez source material downsampled instead of the low-rez source material upsampled and then the test run by playing back through that resolution?

their method offered the most chances for a difference to occur. I often create a downsampled copy of the high res file, then upsample it back to the original resolution, and test the original against the upsampled 16/44. just in case the DAC would behave differently or make a noticeable noise or have a specific delay while switching resolution in one direction... I'd do that to strictly check the impact of the content and only the content. not the DAC and filter choice, not the fact that I'm streaming a different sample rate. my test is more limited and as such offers fewer opportunities to get a positive response.
what they did is have the high res signal go 2 ways, one as is, and one where it was converted to redbook. that as an attempt to really offer the complete experience of each formats as it would when playing them. their method gave more chances to get a difference than my own listening tests ever do.

 

[GearMe]

Gotcha...Thanks!

 

[old tech]

From what I've been gathering, it's just Oohashi's conclusions regarding listener preference that have been discredited. But not his findings about physiological brain response to high-frequency audio. If you have more in-depth informaton, I'd appreciate a link to the source.

IMHO, it's an entirely different question to ask whether an ultrasound stimulus will trigger a human brain response in the auditory cortex, or whether we can distinguish Hi Res from RBCD in a conscious listening test.

Thanks for that. It's been a while since I've looked into Oohashi's work and the peer review that followed it. IIRC, the main issue was that his test and findings were unable to be replicated and later, some flaws were found in the methodology.

Perhaps I am being a bit harsh on Professor Oohashi as he is respected in his field of audiology and has accepted the criticism of his study. As you rightly point out, he never used his findings to claim that ultrasonics influences perception of music, unlike Hendhardt who jumped to this conclusion without any evidence for it.

 

[old tech]

Thanks for the reply!

Makes sense...you brought up another related concern that I was going to ask about as well...basically the recording/playback chain is only as strong as its weakest link.

Follow on question...why is the hi-rez source material downsampled instead of the low-rez source material upsampled and then the test run by playing back through that resolution?

As an aside, Mark Waldrep who among many others state that true hi res can only be a music file that was at least 24/96 from recording all through the production chain does at least try to test his claims. He did an on-line listening test with about 80 or so respondents recently using his AIX studio files. Although it was not a test of scientific quality, the results were less than promising.
http://www.realhd-audio.com/?p=6274

 

[gregorio]

[1] I seem to be talking about scientific measurements.... Likewise, when we're talking science, we tend to look at things a bit differently.
[2] HOWEVER, here's a cute little microphone: http://www.pcb.com/products.aspx?m=378A14 It has a rated frequency response from 4 Hz to 70 kHz (+/- 2 dB). It also has an inherent noise of 50 dB, and a dynamic range of 173 dB.
[3] From those numbers, my guess is that, with a proper preamp, it would have no trouble recording a snare drum, or a cymbal, from an inch away (or from two meters away). And, at that range, it will record everything up past 70 kHz. (So doing so is not only theoretically possible; it's actually possible with currently available equipment.)
[4] It's not my problem if you don't happen to have one of these in your studio to take your measurements with...
[4a] HOWEVER, neither of those facts is relevant if we're discussing what's possible...

1. NO you are not talking about scientific measurements and NO, we do not "tend to look at things differently"! The laws of physics exist and are always applicable, unless you're in some alternate universe. What you're actually doing is making up nonsense and then passing it off as "science", so in fact you're actually perverting the science and are thereby insulting this forum!
2. Inherent noise of 50dB makes it almost unusable for recording musical instruments but regardless, it does NOT have "an inherent noise of 50dB" anyway! The 50dB figure is "A" weighted and due to rising thermal noise with freq (which is real science and in this universe!) the mic will have far higher self-noise in the ultrasonic range.
3. EXACTLY!! It's your "guess" and that guess is based on "inadvertent" errors which fortuitously just happen to support your agenda . How is that even slightly "science"? It's pretty much the exact opposite of science! Instead of making up nonsense claims and falsely passing it off as science, why don't you actually try just the very first step of science and get a musician to perform on cymbals, use this mic to record them and then see for yourself if it will record "everything up past 70kHz"??
4. I do have a measurement mic, in fact more than one.
4a. But you're NOT discussing what's possible! I'm discussing what exists and what's possible BUT you're discussing what's possible in some alternate universe where you get to choose which laws of physics apply!

I am perfectly willing to concede to you things like....
[1] - nobody ever does it that way
[2] - no recording studio you know currently has the equipment required to do it
[3] - it would probably sound bad anyway (or not sound especially good)
[4] - it wouldn't be what somebody sitting in the audience would hear
[5] I would again remind anybody coming in late that what we're discussing here is the possibility of recording cymbals and including spectral content at ultrasonic frequencies.
We are NOT discussing whether that would be audible or even whether it would be desirable... simply that it would be possible. (We are discussing SCIENCE and NOT good recording technique.....)

Continuing for those interested in the actual facts (in this universe!) ...
1. Ask yourself the obvious question, if this mic is so accurate/perfect why is it "nobody ever does it that way", why don't we always record with calibrated measurement mics? In short, there is no such thing as a perfect mic, they must always be compromised: You cannot have a mic with a very low noise floor, a very accurate freq response and resilience to high SPLs. In practise (in this universe) it's a trade-off, you improve one of these areas at the cost of another. Some specialist measurement mics are designed to measure very low SPLs and therefore have very poor resilience to high SPLs and typically poor freq accuracy. Other measurement mics are designed for highly accurate frequency measurement but at the cost of a very high noise floor (which isn't typically a problem because they are designed to be used with optimal/high level test signals). However, the performance of a musical instrument is not a test signal, there will be both loud and quiet elements: A snare drum for example will have a loud impact transient, immediately followed by the relatively quiet sound of the snares sympathetically vibrating against the bottom head and additionally, not every snare hit will be very loud, there will almost certainly be grace notes/flams which are far quieter, and, this effectively applies to all musical instruments. The reason that "nobody ever does it that way" is because these quieter elements would be below the noise floor of a measurement mic (such as the one KeithEmo cited). And, that's even in the audible freq band, let alone the ultrasonic band where the measurement mic will have an even higher noise floor! We always use music/studio mics rather than measurement mics to record music because music/studio mics are optimised for recording music in studios (duh)!
2. This statement is FALSE! Every commercial studio I know of has at least one measurement mic and these days, also studio mics which extend into the ultrasonic range.
3. It wouldn't "sound" at all, if it's below the noise floor of the mic!
4. That as well!
5. And I would remind anybody coming in late that what we're discussing here is the science applicable to recording commercial audio, the stuff that your audio reproduction system is reproducing. KeithEmo on the other hand is discussing certain bits of science and "inadvertently" ignoring the other relevant bits, inventing hypothetical scenarios which never exist and misrepresenting what is science and what is recording technique, which is particularly absurd as he clearly doesn't know anything about recording technique and refuses to try some recording tests and actually find out!

[1] Personally, when I visit a very small venue, I don't enjoy sitting three feet from the band... and one reason is that the cymbals are unpleasantly loud and bright.
And I wouldn't even argue if someone were to suggest that rolling off the cymbals sounds better to most people. Those are the kind of artistic decisions that the recording engineer is expected to make.
[2] And, by the way, HAS anyone ever actually tested whether the sound of breaking glass sounds more realistic when you include frequencies up to 50 kHz?
[3] A lot of what I keep hearing on this forum lately boils down to: "We already know that people can't hear or be affected by ultrasonic frequencies, ever, under any normal listening conditions, so there's no point in doing a test to find out if they can or not." From a scientific point of view that's a real stretch.
[4] What I strongly believe is that there isn't enough evidence to make a legitimate claim either way...
[5] And, AS AN INDIVIDUAL, it always bugs me when people make strong and far-reaching claims based on flimsy and inadequate evidence.

1. Firstly, how are you going to sit 3 feet from the band, they're all going to be standing/sitting in a tiny circle around you are they? In this universe, the audience is typically going to be sitting in front of the band (with the drumkit at the back of the band) and therefore many meters away from the cymbals. At many meters away from the cymbals there's going to be a great deal less ultrasonic content than the only 6% they're producing in the first place, rolling off the very high and ultrasonic content of closely mic'ed cymbals is not an artistic decision, it's a technical decision based on the laws of physics. Not rolling-off those freqs would be an artistic decision but as human perception partly relies on the attenuation of high freqs at distance, you'd effectively end up with the cymbals sounding much closer (more present) than the rest of the drumkit.
2. Again, you're joking right? You seem to have this bizarre notion that sound engineers pick the worst mic they can find and never test or experiment with anything. This notion is ridiculous and the exact opposite of the actual facts/truth! The recording of breaking glass is so common in film and TV that Foley teams always have a large crate of glass in their store rooms and the recording of breaking glass has been tested to death with just about every mic imaginable, in just about every position by thousands of different engineers all over the planet for decades. Personally I only have mics that go up to 40kHz, so not 50kHz but still plenty of ultrasonic content and no, it does NOT sound more realistic, either for me or for the countless other engineers, sound designers, Foley artists, Directors, etc. In fact, generally less so because generally the glass is breaking more than just a few inches from the sound POV and therefore has attenuated high freqs (and ultrasonic freqs).
3. A lot of what you keep saying on this forum boils down to: "I have absolutely no idea what extensive testing has been carried out by thousands/tens of thousands of engineers over the course of decades, so I'm just going to make-up an "inadvertent" misrepresentation that it's never been tested and everyone who disagrees with me is just guessing. I on the other hand have never tested either and refuse to do so but my guesses are worth more than the actual facts. According to me, that's SCIENCE!".
4. Firstly, there's clearly a big difference between the amount of evidence which exists and the amount of evidence you personally know about. Time and again audiophiles state "we don't know this or that", when in fact it's perfectly well known, often for many decades or even centuries, it's a fallacy based entirely on their own IGNORANCE of the facts. Just because they don't know doesn't mean that we (science/mankind as a whole) don't know. There's a mass of evidence, by (as mentioned) the engineers who work with the content every day, industry bodies such as the AES, EBU, ITU (and others), state organisations such as the BBC, NHK (and many others) and some published scientific papers as well. Secondly, this is NOT the "What KeithEmo Strongly Believes" forum!!
5. Then why do you keep doing it??? Why don't you AS AN INDIVIDUAL actually learn some of the facts/evidence, why don't you take YOUR OWN ADVICE and try recording some drumkit solos (and breaking glass) with measurement and other mics and until you do, why don't you STOP making far-reaching claims based on ABSOLUTELY NO EVIDENCE WHATSOEVER? How come you're not "always bugged" by yourself?

[1] We say that "for DACs 192k is NOT a product differentiator" (it was ten years ago - when few DACs supported it - but not now). ...
As a company who sells hardware, we do our best to include in our products the features that the market has a demand for, and specifically the features that our customers ask for. And, at the moment, a lot of our customers specifically ask for us to support high-res sample rates.
[2] So, why has Emotiva, who I work for, never conducted and published any tests about whether people really can hear the difference between high-resolution files and plain old CDs?

1. Just to be clear what you're effectively saying and the truth of the matter:
Ten years ago snake oil salesmen made a big marketing push for 192kHz audio files and 192/24 DACs. That marketing was successful, many/most consumers believed the BS and now demand 192kHz. So all the chip makers now only make 192kHz chips and you're just satisfying the market demand (for snake oil) that ten years ago you helped create. Of course, now that 192kHz is no longer a "product differentiator" and even very cheap DACs now include it as standard, the audiophile snake oil industry has to come-up with some new BS to act as a "product differentiator" (and justify their 10-100 times price premium), hence the next round of even greater snake oil; 32bit, 384kHz and 768kHz, just as 192kHz was the next round of greater snake oil over 96kHz.
2. Because either you'd have to fake the tests and run the risk of being found out and having potentially disastrous publicity or provide accurate tests demonstrating that you and the rest of the audiophile industry have been BS'ing and selling snake oil for well over a decade. That's a no win scenario and why neither your company nor any other audiophile company ever conducts or publishes such tests. To be certain, if there were an audible difference/improvement, the audiophile world would be awash with the test results!!

G

 

[gregorio]

[1] And second, the further away, the less assymetric the waveform will be; any brass instrument can, in general, produce only over pressure ( it can not suck air inside itself/person playing it ) Real close miking will reveal there is almost no underpressure at the brass instrument nozzle - the output pressure is almost all in the positive pressure region. That's WHY frequency response to DC is required - to preserve this offset also in the recording. I know it takes TWICE the amplitude in recording it would take if the lows below say 15 Hz or so would be rolled off ... but the realism would be lost.
Yes, some better players can play their brass also by sucking
[2] OK - worst case scenario attenuation of 30 and 40 kHz for 3/4/5 metres ? Brass instrument "exhaust" to mic distance , direct path, no obstacles whatsoever. How many dBs ? Whatever the correct answer, it can not get low enough to be possible to neglect.

I'm not sure any of this is worth responding to, it's all just complete nonsense but just in case:

1. There's no such thing as a brass instrument "nozzle", presumably he means "mouthpiece". The sound is produced by the player blowing air and vibrating his/her lips against the mouthpiece. This sound, as with ALL sound, is a sound "wave" which has both positive and negative pressure, otherwise it wouldn't be a soundwave! DC is not a sound wave and why frequency response to DC is NOT required. The lowest brass instrument is the contra-bass Tuba (often just call the Bass Tuba or just Tuba) and the lowest note in it's range is "C", 3 octaves below "middle C" (although this note is extremely rarely ever written), which has a frequency of 31Hz. This is more than an octave above the 15Hz analogsurvivor is FALSELY claiming. So there is no "realism" (or anything else) below 31Hz in the first place and therefore it cannot be lost! Also, you cannot play a brass instrument by sucking, though there are a handful of experimental compositions which require just blowing or sucking air (without the lips vibrating), thereby just producing the sound of air/breathing rather than actual notes.
2. Clearly if it's below the noise floor of the mic, then it's not only possible to "neglect" it, it's unavoidable!

1.) Non availability of the equipment that can ACTUALLY record above 20 kHz in a competent way in MOST studios/by MOST recording engineers.
2.) Non willingness by MOST recording studios/engineers to adopt it - because it, by default and by definition, would require turning the current predominant use of multimiking upside down, in favour of less complicated microphone recording techniques.

1. That's false, ALL of the studios I know do have mics and recording equipment capable of recording above 20kHz (in a "competent way").
2. Multi-mic'ing has NO baring whatsoever here. We can just as easily record say 30 channels of 96kHz or 192kHz as 30 channels of 44.1kHz and ALL the professional DAW systems I'm aware of have had that capability for more than 15 years. Nothing at all would turn "upside down" and indeed most 96kHz and 192kHz original recordings ARE multi-mic'ed!

G

 

[KeithEmo]

That is a very interesting article... and it seems to me to make quite a few very valid points.
Well worth reading.

I'm still generally on break, but wanted to share this, in case it hasn't been shared before (I don't remember):

https://tapeop.com/blog/2018/08/30/subconscious-auditory-effects/

It may have some relevance to the current topic of ultrasonics, and perhaps worth keeping in mind that the boundary between sonic/ultrasonic varies a lot between people. 18 kHz may be consciously perceivable for one person, only subconsciously perceived by another, and neither consciously nor subconsciously perceived by another.

 

[KeithEmo]

If you really want to make a general test of "whether SACDs sound audibly different than CDs"....

You really need to use more than one SACD, more than one CD player, and more than one sample of each medium.
You also need at least a few samples which have been confirmed to contain what you're testing for.

Therefore, at a minimum, you must confirm three things....
1) you must confirm that the master used for both was the same
2) you must confirm that the entire sginal path of both productions took full advantage of the medium
3) you must confirm that your player is taking full advantage of both types of media playback
(if the analog signal path of your SACD player only extends to 20 kHz, then that is the limit of your test)

If you want to compare the high frequency response, or the ultrasonic response, of two different recordings...
Then you must first confirm that the master recording that both were produced from actually contained those frequencies.
Then you need to confirm that the rest of the signal chain hasn't failed to reproduce what you're testing for.
Then you need to confirm that the actual output of your SACd player is delivering both signal accurately.
(And you also need to confirm that your speakers, or headphones, aren't limiting the respons eeither.)

You can't tell whether 30 kHz is audible unless you start with a test signal that CONTAINS 30 kHz.
If you want to test whether the "ability" of an SACD to deliver ultrasonic frequencies makes a difference...
You must first confirm that the test samples you choose actually do so...

Proving that specific SACDs don't sound better doesn't prove whether ANY SACD CAN sound better or not.
It could trun out that most commercial SACDs are mastered badly and fail to live up to the potential of the medium.
(I would suggest that most CDs in fact fail to live up to the potential of the CD medium.)

NOTE that this isn't all that hard.
First you confirm, using proper test equipment, that the actual outputs of the two are in fact different.
Then you confirm, using proper test equipment, that your reproduction chain is delivering that difference to your ears.

I should also point out the obverse.....
If you simply want to test if there is a difference that is audible TO YOU ON YOUR SYSTEM....
Then performing the test using your system and your ears is perfectly adequate....

Do you have an SACD player? All you need is an SACD recorded in native DSD and with a redbook layer that is the same mastering. I can point you to the one I used if you want. Aside from that, it's just a preamp and switcher.

 

[KeithEmo]

I should also point out one more complicating factor....

Because of the restrictions REQUIRED by the SACD production license....
Commercial SACD players are ONLY permitted to deliver a full quality digital version of content on an SACD via HDMI.
So no legal SACD player will give you a full quality digital output you can simply send to a PCM DAC.
(They are restricted to only delivering it via "a secure connection" - and HDMI is the only type that qualifies.)
(And VERY FEW stand-alone DACs have HDMi audio inputs.)
Therefore, you are going to be limited to using the ANALOG outputs of your SACD player...
(Some SACD players may offer a digital output, but, if they're legal, then it will have been down-sampled, at a minimum.)
And so to the capabilities of its internal DACs and analog signal path...

Don't have any of those things handy. I have a DAC that may do 96khz, but I sincerely doubt its analog ouputs are any good up past 20khz. Also, we'd have to independently establish the noise floor of each piece of gear at / above 30khz, etc. Not to mention find some transducers that are actually good > 30khz, again, probably not trivial. I have some Blue Sky monitors and I have some ER4XRs and (although they are wonderful) I don't think either one would be worth a crap for this test.

I mean, I can tell you right now none of my gear that I already own is likely up to the task. There's no question that ultrasonics are worthless in the context of things I own and things I actually listen to. But that's not the question I am interested in.

 

[KeithEmo]

Quite so....

It may be that most listeners don't hear a difference - but that a few do.
Ot it may be that most listeners would simply find that it gives them a headache...
Or even that they hear a difference but don't find either to be better...
However, if the results aren't totally clear....
It would be very simple to include a switchable filter to allow people who want to to apply a limit....

I'm also perfectly comfortable if the recording engineer were to decide that choice for me.
However, I would prefer if we both have that choice, rather than being limited based on someohne else's assumptions.

I also really dislike the current situation - which amounts to "appeal to authority"...
From what I can tell, we are far from "having a consensus among all good recording engineers".

- @gregorio insists that "as a recoding engineer, he knows better"...

- But Cookie Marenco, who also seems to be a well regarded recording engineer,
is dead certain of the exact opposite...

- And Mark Waldrep, who also seems to have a decent reputaion, seems to believe that high-resolution recordings
do actually have the ability to sound significantly better, but that many of the recording currently sold as high-res
simply aren't.

And, while several of the individuals listed "have their own agenda", it's also absurd to suggest that "everyone who has their own agenda must by either lying or deluded". (If you want to live with that assumption, then every recording engineer who earns a salary has incentive to claim that he's the only decent engineer on the planet, and any engineer who does things a little differently is either a liar or an idiot, and every studio has a reason to suggest that only they have the quipment that really sounds good, and everybody who chose different equipment must be deluded.)

Recent article on ambient ultrasonics:

https://www.livescience.com/62533-ultrasonic-ultrasound-health-hearing-tinnitus.html

Doesn't seem that ultrasonics are necessarily something we'd want to capture in reproduced music. Seems plausible to me that at least some people might enjoy music more without ultrasonics.

(Yes, I know, supposed to be on break. I'm taking a break from being on break.)