No it's not, as you would know with even a smidgen of experience in either perceptual or audio research. Regardless, even if the sounds are only "felt" it will affect our overall perception of the music; can I assume you understand that concept?
Redbook 44.1 kHz standard: theoretically sufficient timbral resolution?
- Thread starter Al M.
- Start date
You speak as if you are familiar with research on the topic. A few paper references would give your claims some credibility here.
Have you read the Wiki references? Here is the last one from ASA Journal: A Note on the Audibility of Intense Ultrasonic Sound, by Bruce H. Deatherage, Lloyd A. Jeffress and Hugh C. Blodgett
"During the summer of 1952 we heard a high pitched tone while
working in the water in the beam of a transducer transmitting
at 50 kc. The sound could be heard only by bone conduction, never
by air, even with reasonably high intensities.."
So this is not at all supportive hearing ultrasonics using our audio systems. It goes on to say:
"The pitch of the sound heard under ultrasonic stimulation
is the highest pitch a given subject is capable of hearing. Placing
some part of the head directly in contact with a transducer, if
stimulation begins with a frequency well within the sonic range,
it will be found, when the frequency is slowly raised until the
pitch ceases to change, that for different subjects the cessation of
pitch change will occur at different frequencies. Further changes
upward in the frequency, and into the ultrasonic region, will leave
the pitch unchanged at that highest level. "
Surely this is no definition of high-fidelity if each one of us maxes out at a different spurious frequency.
"Thresholds were determined in two different ways: (1) bringing
a small part of the jaw bone in contact with the surface of water
in a bucket holding the transducer. At 50 kc, the threshold
obtained was approximately 2000 dynes/cm 2. (2) Submerging a
swimmer completely in the water containing the sound field.
Here the threshold was about 1000 dynes/cm 2 at 50 kc. At 7 kc
the threshold was approximately 12 dynes/cm 2, a figure which
agrees with yon B6k6sy's findings for bone conduction."
Translating into English
, the sensitivity to 50 Khz was 100 to 200 times lower than 7 Khz. With so little ultrasonic existing in music already, there is little to no chance of hearing this even if you played the loudspeaker in the water and stuck your head into the same .And this cautionary note:
"Those who demonstrate the phenomenon to themselves are
cautioned to exercise some care in controlling intensities. Pain will
seldom if ever be experienced with strong bone-conducted stimulation
in the ultrasonic region over short periods of time, but it will
produce tinnitus of several days' duration. The tinnitus may well
be a sign of damage."
Wiki fidelity has improved in recent times but that page is in need of serious review and editing.
Translating into English
lol
Take a look at the frequency spectrum of most 24/96 or better recordings. Yes, the volume at high frequencies is down compared to the midrange; that doesn't mean it's not perceived nor that it doesn't affect the sound we hear. And why would anyone think that we all hear the same!! Obviously tomelex hears differently than many others whose musical enjoyment depends at least in part on accurate reproduction of the highest frequencies they can perceive. Equally obviously each of our hearing abilities decline with age, both bone and air conduction, both high frequency "cut-off" and absolute volume threshold at any frequency; I don't think that's controversial.
As far as references go, the Internet can be a terrible place to find them if they are scientific publications, most of which only offer that information to subscribers. But a lot of it is just common sense; for example, most people, audiophiles or not, "sense" (but don't hear?) infrasonics. But since they are "infrasonic", should we ignore all frequencies under 20-25 Hz in our recording and playback systems?
As far as references go, the Internet can be a terrible place to find them if they are scientific publications, most of which only offer that information to subscribers. But a lot of it is just common sense; for example, most people, audiophiles or not, "sense" (but don't hear?) infrasonics. But since they are "infrasonic", should we ignore all frequencies under 20-25 Hz in our recording and playback systems?
There are engineers who work on the assumption that humans cannot hear ultrasonic sounds e.g. https://en.wikipedia.org/wiki/Sound_from_ultrasound
So to create a directional sound beam they propagate ultrasonic energy at very high levels on the assumption that the listener will not hear the ultrasonic part at all, only the nonlinear distortion components at a much quieter level.
There is also a device to discourage teenage loitering based on the assumption that only young people can hear frequencies above 17 kHz, and a commercial phone ringtone whose pitch is so high that adults cannot hear it.
Sorry, but I am frustrated at the way all of audio has to be reduced from science to 'craft-based' messing about and that people give credence to it, as though some hitherto unknown discovery has been made. It is obvious that digital audio in the form of modern integrated circuits is not amenable to craft-based messing about, so the tinkerers spot that they must play the 'classic' card and bring back the DAC chips that had separate filtering that they can mess about with. Make a few claims about unsubstantiated "problems" with the mathematically-perfect theory et voila: seeds of doubt and confusion propagated successfully amongst the audiophiles, and the tinkerers carve themselves their own small niche in the audio world based on their ability to solder resistors and capacitors together and possibly calculate/simulate/measure phase shift and frequency response of basic filters. Then it is simply a question of picking a filter that isn't as extreme (or some might say perfect) as the conventional reconstruction filter, but is a cinch to make from a few basic components. Joe claims 0.5 is the best sonic result, while Fred picks 0.7 - all values 'work'. Everyone can have their own special filter and claim it to be so much better than those "mid fi" ones those boring scientists came up with using theory alone.
Don't know if you're referring to my previous post ironically or something, but objectivity is of no use if there is no agreement on the objective criteria by which something can be judged. There can never be any such agreement as long as credence is given to any, and all, assertions dreamt up by wannabe high end designers who want their own little slice of the action.
Sorry to digress the OP.
I do wonder if this would also be applicable to headphones with aspects of the design that is in contact around the ear and the headband.
Keith Howard showed that many headphone designs actually transfer a signal from one side to the other due to headband resonance (it is audible in the tests he did); not noticed with a simple sine wave but more complex such as pink noise.
So in theory that should also transfer not just via headband but all surrounding contact from the ear cup and actual headband; quite a bit of surface to the head and under ear.
And then there is in-ear transducers.
Cheers
Orb
Thanks for the explanation Groucho.
I appreciate it can be frustrating where it seems a technology is reduced from science to 'craft-based' messing, but then it is a very fine line as we have seen by the work from Gerzon/Craven/Stuart, and then also others who worked much further on dither/oversampling/etc historically before it was fully accepted.
A lot of the engineering is not really craft-based messing around, but core to digital, the overlap can be where trying to implement core theory into real-world application - especially if one focuses on the original core theory without it being expanded (many objectivists on forums it seems exclude the works done by Gerzon/Craven/Stuart amongst others).
I can accept talking about NOS DACs is kinda frustrating as it cannot be implemented perfectly, or even very well
But then it did have a possible place in audio up to around 2010 where more engineering-implementation work was done with the focus on digital filters and oversampling, before this it was rare for it to fine-tuned and bespoke development independent of the main DAC chip manufacturers; around 2000 only a very few applied specific and powerful hardware (relative to what was provided as a shared resourced by DAC chip manufacturers and general filter-oversampling capability) to this, one big example would be Chord Electronics with the DAC64.
Another such leap is looking at say the ESS Sabre DAC chips, and yet it was some exceptional engineers outside of ESSTech such as John Westlake that identified some issues with their higher end DAC chips in the early days and created their own workarounds and implemented them in the Audiolabs M-DAC.
Now some may say that is messing around, but it is engineering and science.
John Swenson has chatted about his own experience in the past with NOS DACs and that of oversampling designs with his own thoughts how this has changed over the years; like myself he feels NOS DACs did have a 'place' (limited and more from a concept) going back due to the resource constraints in terms of integrated functionality, hardware, and flexibility - he also feels NOS DACs is a flawed design but did some aspects well subjectively compared to say mainstream DAC chips going back and I agree with all that myself including its flaws (why I would never own one myself but can accept others did and comes back to what JA says from a limited context on SQ; "does it sound good because of its poor measurements or despite them").
Fair to say IMO like Swenson would there is really no need for NOS DACs when considering modern chip designs,hardware, and flexibility now available.
I think he felt the issue was the integrated cascade filter implementation and hardware resource available, personally I would think also goes beyond that albeit with more marginal differences.
A modern day comparable example with speakers; the Fujitsu-Ten manufacturer with their TD712 speakers and all out focus on time domain behaviour, this speaker could be deemed 'broken' in view of several on this thread because it has pretty strong limitations (frequency domain related being a big one), but does show an interesting performance/SQ from a time domain perspective compared to traditional loudspeakers.
Subjectively it shows possibly one area mainstream loudspeaker designs are going wrong, albeit creating other issues to resolve this one specific variable.
Science-engineering or 'craft-based' messing around; people will be split
Cheers
Orb
Again this strange assumption that everyone's hearing is the same (I assume within age and sex parameters)?
Vast majority of our members are over 40 years old. What are the possibilities that they even hear 20 Khz let alone higher? Have they tested themselves?
When I was at the height of my golden ear career where we I could hear a mosquito walking outside
, I bought a device that measured your ear's frequency response using a calibrated headphone and amplifier. I tested myself and almost smashed the thing at the wall when it said that my high frequency perception was already shot to hell. The device was I think $250. I bought it instantly as soon as I found out about it. Why won't audiophiles who spend far more than that on cables and such, buy such a device so that they know if they can even hear the audible band let alone above it?And why do we think the person who created our music could hear it? They are not 16 year olds either. Here is a great example. These are two-tone stimuli that are sold to make people relax, etc. I analyzed it and this popped up:
Putting aside the problems with noise and such, look all the way to the right. That sharp spike is from a Mac monitor that was used in 1980s to capture and edit this track from R2R! This is from the most popular source of such tracks so likely hundreds of thousands of people have sat there and subjected themselves to 16 Khz tone without hearing it. And of course the people who sell the stuff have no idea either because they don't hear it just the same.
Mind you, I think any transmission channel must be transparent for the full population. But among us chickens, let's not assume that we get a score A when it comes to high frequencies. Sadly most of us live in D to F category and have no leg to stand on to ask for frequencies well above anyone in the world as we ask for 96 Khz, 192 Khz, 384 Khz, etc.
Actually they are done on a standards specified artificial head with pinna (Keith Howard follows those test standards including use of an accurate artificial head-pinna with associated measurement equipment)
And none of the tests looked to investigate headband resonance that can transfer from one transducer to another using complex sound (which we can include pink noise), or how the sound/resonance transfers through the bones associated with a head.
Sean Olive (him and Keith Howard used near identical associated headphone standards testing equipment) has also been investigating headphones in recent years and his measurement-preference findings seem to align pretty close to Keith Howard, albeit Sean has not yet I think investigated the effect of headband resonance or the other thoughts just recently posted that touch on yours and Amir's response to the hearing mechanism related to ultrasonics.
Cheers
Orb
Last edited:
A standard "dog whistle" is 23 kHz; I don't know about the rest of you but when I was in high school and college there were always a few of us who could reliably hear it. I think I posted here recently that whether using on-line hearing tests or tones generated in Adobe Audition I can reliably (and blindly) hear to about 16.8 kHz if the volume is high enough. At age 62 I may be a bit of an outlier but I suspect not that much; I'll bet at least 5% of people my age hear as well or better.
As far as bone conduction vs ear hearing, both need to be considered; it's obvious to everyone where infrasonics are concerned, it shouldbe just as obvious for ultrasonics.
As far as bone conduction vs ear hearing, both need to be considered; it's obvious to everyone where infrasonics are concerned, it shouldbe just as obvious for ultrasonics.
But are you sure you aren't hearing a distortion artefact at half the frequency? I was attempting to repel rodents with a tweeter using a sine wave at 20 kHz. At a high enough level I could hear it, but it was, in fact, just a 10 kHz artefact.
In the guise of appearing knowledgable you are merely revealing your own naivete. For instance do you really think that only a small percentage of teenagers could hear a 11.5 kHz tone?? And didn't I say "if I turned the volume up high enough"? Because there come a point where the volume doesn't matter, you simply don't hear or perceive the sound at a given frequency.
Similar threads
