cocolino said:
Oohashi's tests seem to indicate that there's some sort of modulation effect occurring somewhere as the supersonic content only seems to cause differences in the EEG and PET when accompanied by sound within the audio band. Supersonic sound on its own doesn't have the same effect.
se
cocolino said:
I don't know about anyone else, but based my own on experiments with a signal generator and headphones, my hearing is pretty much worthless above 12KHz. That's not to say the performance of gear above 12KHz is inaudible to me, just that any fundamental tones above 12KHz are hard for me to hear.
P.S. I'm 45 and have mowed many a yard and attended many a rock concert without ear plugs and so I have only myself to blame.
Phil
cocolino said:
Well, the first link is a precis of the second, but it is an interesting hypothesis, and I wouldn't rule it out. But without being able to read the original papers by Oohashi and Lenhardt it is hard to draw a proper conclusion. If Steve is correct in stating that these results only happen when the high frequencies are modulated by percievable ones, then the 15K finding is still valid in this discussion.
When I first saw that paper last year, I nearly started celebrating as it seemed to vindicate my personal feeling that audibility is not everything in our experience of sonic events.
In addition to the bone-conducted ultrasonic theory, I've been thinking that other parts of anatomy play a role. If subwoofers can excite certain thoracic transmission lines, then ultrasonic resonances must also be "felt" somehow. Unfortunately, I have no idea which of our body parts would sense and/or convey the experience to our brains, never mind figure out a test for it.
Still, I would like to attribute that special feeling of "the hair standing on the back of my neck" to the particularly sweet ultrasonic harmonics of extremely refined musical instruments. For me, that feeling is usually brought on by a particularly brilliant piece of music and/or a virtuoso performance.
🙂ensen.
In addition to the bone-conducted ultrasonic theory, I've been thinking that other parts of anatomy play a role. If subwoofers can excite certain thoracic transmission lines, then ultrasonic resonances must also be "felt" somehow. Unfortunately, I have no idea which of our body parts would sense and/or convey the experience to our brains, never mind figure out a test for it.
Still, I would like to attribute that special feeling of "the hair standing on the back of my neck" to the particularly sweet ultrasonic harmonics of extremely refined musical instruments. For me, that feeling is usually brought on by a particularly brilliant piece of music and/or a virtuoso performance.
🙂ensen.
pinkmouse said:
Here's the paper as published in the Journal of Neurophysiology:
Inaudible High-Frequency Sounds Affect Brain Activity: Hypersonic Effect
Here's the quote in support of what I'd said initially (in case you don't want to wade through the whole thing trying to find it):
Moreover, it is the combined presentation of HFCs and LFCs, not HFCs alone, that specifically induces the enhancement of alpha-EEG and activation in the deep-lying structures. We interpret these findings to mean that the hypersonic effect does not simply result from a neurophysiological response to isolated frequencies above an audible range, but from a more complex interaction to which HFCs and LFCs both contribute.
HFCs and LFCs standing for High Frequency Components and Low Frequency Components.
se
I don't know if the above examples are the one I'm thinking of or not, but i read of an an experiment conducted a few years ago where a 20k+ tone was added to sond in the accepted audible range and the subjects could detect it even though unable to detect the 20k+ tone alone. It was thought this proved that humans were sensative to sounds over 20k, however, a subsequent attempt to replicate the results filtered the combined signal to eliminate everthing avove 20K, the subjects could STILL tell the difference. I don't recall all the details, but the conclusion was that there were IM products below 20k generated by adding the +20k material.
It seems to me that any investigation in this area needs to routinely include a spectral analysis of the output for each case considered..
It seems to me that any investigation in this area needs to routinely include a spectral analysis of the output for each case considered..
I know this is not exactly to the point, but it makes an argument for not ignoring the ultrasonic contribution to what is being heard.
I’ve always figured that we hear the sum and difference (beat) frequencies created by harmonics in the range greater than 15kHz even if we can’t hear the pure frequency (at least not at its full level). This would not be a big issue if everything was recorded and played back in mono, but with multi-track recording and, at least, stereo reproduction, beat frequencies will be produced in the listening space.
IMO, if you limit your sights to less than 20kHz you’ll miss out on a lot of music and ambience. This is particularly true for vinyl. Just for fun and to illustrate the idea, checkout this site.
http://207.10.97.102/explrsci/dswmedia/tonebeat.htm
I’ve always figured that we hear the sum and difference (beat) frequencies created by harmonics in the range greater than 15kHz even if we can’t hear the pure frequency (at least not at its full level). This would not be a big issue if everything was recorded and played back in mono, but with multi-track recording and, at least, stereo reproduction, beat frequencies will be produced in the listening space.
IMO, if you limit your sights to less than 20kHz you’ll miss out on a lot of music and ambience. This is particularly true for vinyl. Just for fun and to illustrate the idea, checkout this site.
http://207.10.97.102/explrsci/dswmedia/tonebeat.htm
Roddy, the beat frequencies will be awfully low in amplitude- music follows roughly a 1/f power law, and the beat or heterodyning products are reduced in amplitude even further. I wonder about the application to real listening spaces with real background noise.
Vinyl is a different thing- most of the really high frequency stuff is noise, not signal.
Vinyl is a different thing- most of the really high frequency stuff is noise, not signal.
Does this mean we should ignore the effects?SY said:
The amplitude of the beat frequency should be sum of the amplitudes each of the individual frequencies. If the delta f is 400Hz, I'm sure this would be audible, even at very low levels.
With the complexity of music (as opposed to pure tones) I have to believe the effect is real and audible.SY said:
I have yet to hear a CD that can deliver the ambience comparable to some good vinyl recordings. That's not to say they don't exist, it's just I haven't heard it. There may be a number of reasons for this. I've always took one of the reasons for this to be the absence of the "ultrasonic" frequencies in the CDs.SY said:
Regarding beat frequencies: I can imagine that musical instuments' overtones can produce beat frequencies in the range below 20 kHz. But I tried really hard to figure out which overtone combination of John Curl's 20 kHz rectangular could produce beats in the audio range but found none ! So IMD is ruled out for his case IMHO. That leaves anything from ringing, up to the possibility that it is in fact caused by our hearing.
If you don't use a static signal like John's rectangular, HF restriction can deteriorate the perception of a sound source's location.
Regards
Charles
If you don't use a static signal like John's rectangular, HF restriction can deteriorate the perception of a sound source's location.
Regards
Charles
Rodd...may I interject a comment on your comment about ambiance...I have some old tapes that were recorded from my turntable to my NAK years ago and I still play them on occasion and they do posess a certain signature compared to a cd....but they used different recording hardware back then too....I equate alot of that today with our current media
DIRT®
DIRT®
phase_accurate said:
John said he used a 5kHz square wave.
As I remember, I set the function generator with a 5KHz square wave and deliberately limited the risetime to 3.5us, which is about a 100K response, with a quality film polystyrene cap to ground.
That would put all of the second order difference products of each harmonic pair at 10kHz, i.e. 15kHz - 5kHz, 25kHz - 15kHz, 35kHz - 25kHz, etc. And of course there will be harmonic distortion products of the square wave's harmonics. For example the second harmonic of the square wave's third harmonic is 30kHz, which with the square wave's seventh harmonic at 35kHz gets you a product at 5kHz, and so on.
So if John did indeed use a 5kHz sqaure wave, I can see where quite a lot of intermodulation energy could manifest down in the audio band.
se
O.K. O.K. O.K.
I didn'd have a close-enough look. The 10 kHz could definitely be perceivable, depending upon level. But your IMD-generated 5 kHz will not interfere, because it is much weaker than the 5kHz fundamental. I.e. it will alter the 5 kHz fundamental by a veeeeeery tiny fraction of a dB.
I still don't rule out that the changes COULD be caused by human perception.
IMO the assumption that you don't have to pay attention to a system's response above 20 kHz, simply based on tests with static sinusoidal signals, is a false conclusion.
Regards
Charles
I didn'd have a close-enough look. The 10 kHz could definitely be perceivable, depending upon level. But your IMD-generated 5 kHz will not interfere, because it is much weaker than the 5kHz fundamental. I.e. it will alter the 5 kHz fundamental by a veeeeeery tiny fraction of a dB.
I still don't rule out that the changes COULD be caused by human perception.
IMO the assumption that you don't have to pay attention to a system's response above 20 kHz, simply based on tests with static sinusoidal signals, is a false conclusion.
Regards
Charles
phase_accurate said:
Nor do I. I'm simply saying that all reasonable possiblities need to be considered until such time as those possibilities can safely be ruled out. Until that time, the results remain ambiguous.
That's not any assumption of mine. My only assumption is that there are still some unanswered questions which prevent other assumptions from being made.
se
My view
It is my personal view that you cannot hear the effect of sounds above 20k directly, but that it is certainly possible to hear the effect of an extended bandwidth signal on the equipment reproducing it.
My experience is a properly designed, carefully executed filter can be used to bandwidth limit signals without sonic side-effects, but this is hard to do, and hence the reasons for different results experienced by different people.
We should not overlook that it is entirely possible that feeding extended bandwidth signals through amplifiers not designed to handle those signals will have audible consequences. By designing a product to deal with these signals proeprly though we make our lives much harder in design terms.
This makes interesting reading: -
http://www.audiomedia.com/archive/features/uk-0400/uk-0400-listeningtest/uk-0400-listeningtest.htm
Of particular note is the engineer (Bob Katz) was pre-disposed to hearing the differences of extended bandwidth, based on his previous experiences. For him to change his view based on this test is of great interest, I suggest.
Andy.
It is my personal view that you cannot hear the effect of sounds above 20k directly, but that it is certainly possible to hear the effect of an extended bandwidth signal on the equipment reproducing it.
My experience is a properly designed, carefully executed filter can be used to bandwidth limit signals without sonic side-effects, but this is hard to do, and hence the reasons for different results experienced by different people.
We should not overlook that it is entirely possible that feeding extended bandwidth signals through amplifiers not designed to handle those signals will have audible consequences. By designing a product to deal with these signals proeprly though we make our lives much harder in design terms.
This makes interesting reading: -
http://www.audiomedia.com/archive/features/uk-0400/uk-0400-listeningtest/uk-0400-listeningtest.htm
Of particular note is the engineer (Bob Katz) was pre-disposed to hearing the differences of extended bandwidth, based on his previous experiences. For him to change his view based on this test is of great interest, I suggest.
Andy.
- Status
- Not open for further replies.