Subharmonics of ultra high frequency information...

[Phelonious Ponk]

I've often heard this case made for the inclusion of frequencies above the audible range in reproduction -- that, for example, while a 30 kHz signal is above the audible range, its first sub-harmonic (15 kHz), second sub-harmonic (7.5 kHz), and third (3.75 kHz), are not.

But if the 30kHz signal is on the master tape, and you copy to Redbook, filtering above 20 kHz, wouldn't all of those subharmonics, below 20 kHz, be be copied to the CD? And wouldn't they all -- the audible subharmonics, not the super-audible fundamentals -- be reproduced by any system capable of 20 kHz? Would the harmonics generated by super-audible fundamentals be lost, even if you were filtering the original recording itself? Or wouldn't even a microphone cut off above 20 kHz, pick up these subharmonics coming from a cymbal as long as those harmonics were below 20 kHz?

I'm sure I must be missing something, but I needed to ask. It's not quite making sense to me.

Tim

 

[mep]

Tim-I'm going to take a stab at this. I'm assuming you are talking about transferring an analog master tape to redbook CD. If the mastertape contained the subharmonics you used as an example (15kHz, 7.5kHz, and 3.75kHz), they would certainly make it to the redbook trasnfer to CD and they would be heard by any decent system.

With regards to your last two questions, I think the answers are "no" and "yes." No, I do not think the subharmonics would be lost because you are only filtering above 20kHz. And yes, the mike should certainly pick up the subharmonics below the 20kHz filter point you assigned.

 

[Ethan Winer]

I'm sure I must be missing something, but I needed to ask. It's not quite making sense to me.

You are missing something, and it's very basic and simple: There's no such thing as sub-harmonics.

And yes, I'm totally serious. But I'll clarify further in advance in case someone brings these up:

1. There's a violin / cello bowing technique that creates frequencies lower than the open string. You draw the bow slower than normal, and also press harder than normal. This creates a scratchy sounding low frequency tone that would be classified as a special effect. But the idea that sub-harmonics occur naturally is incorrect.

2. In this Wikipedia article:

http://en.wikipedia.org/wiki/Subharmonic

The author talks about bells having sub-harmonics. But that's not really accurate. Bells (and bars) are some of the few sound sources in nature that create overtones not necessarily related to the fundamental pitch. So what can happen is overtones might mix non-linearly to create sum and difference frequencies. But this happens only when something distorts, and it's still nothing like "sub-harmonics" that would be related numerically to the fundamental if such a thing really existed in the first place.

--Ethan

 

[tony ky ma]

Hi Tim
This is my thinking, when you hear a sound from cymbal ,the real one not repro, you will have a memory that is a cymbal's sound, may be what you can hear is only the combination of harmonics because the main frequency is too high and can not audible but still will remember that is a sound of cymbal, in reproduction, in some reason the cymbal sound from the speakers is only the harmonics, main frequency doesn't exist ,your brain will tell you that is sound of a cymbal too because you don't have that main frequency in your memory but when compare to the real one, there will be a little different, because the main frequency not in the air, that is the difference between real sound and repro and that is what I believe in reproduction the ultra sound range low or high should be in the air too no matter it can be heard or not for closer to a real sound , also timing of harmonics' combine is a issue too, that is why a coupling cap can give a big different in sound quality
tony ma

 

[Phelonious Ponk]

Tim-I'm going to take a stab at this. I'm assuming you are talking about transferring an analog master tape to redbook CD. If the mastertape contained the subharmonics you used as an example (15kHz, 7.5kHz, and 3.75kHz), they would certainly make it to the redbook trasnfer to CD and they would be heard by any decent system.

With regards to your last two questions, I think the answers are "no" and "yes." No, I do not think the subharmonics would be lost because you are only filtering above 20kHz. And yes, the mike should certainly pick up the subharmonics below the 20kHz filter point you assigned.

So then what is the advantage in reproducing anything beyond the audible spectrum?

Tim

 

[mep]

Are you addressing your question to Ethan?

 

[Phelonious Ponk]

Are you addressing your question to Ethan?

Whoever has an answer. I just thought about the whole thing this morning, the notion that super-audible frequencies impact our listening experience because of the overtones they generate that are audible, and that's why it matters that we record/reproduce frequencies that are outside of the audible range. But it the overtones themselves are in the audible range, are captured on the recording, and reproduced by the sound system even if that recording/system doesn't capture/reproduce the fundamentals, what difference does it make?

Tim

 

[amirm]

Our library has a ton of articles on this front guys . This research while disputed, does talk to possible cause and effect (i.e. it is the ear whose response changes due to hypersonics): http://www.linearaudio.nl/Documents/high freq inpact on brain.pdf

"Hypothetical explanation of neuronal mechanisms of the hypersonic effect

From an authentic view of human auditory physiology, it is not straightforward to explain the neuronal basis of the hypersonic effect characterized by the fact that HFCs showed signi?cant physiological and psychological effects on listeners only when presented with audible sounds. Although how inaudible HFCs produce a physiological effect on brain activity is still unknown, we need to consider at least two possible explanations.

The ?rst is that HFCs might change the response characteristics of the tympanic membrane in the ears and produce more realistic acoustic perception, which might increase pleasantness. However, this hypothesis is unlikely to explain the fact that the subjects who showed signi?cant hypersonic effect were not necessarily aware of the difference of sounds in a conscious manner. An alternative explanation is that HFCs might be conveyed through pathways distinct from the usual airconducting auditory pathway and therefore might affect the CNS, including the deep-lying brain structure. It was reported that the vibratory stimulus of ultrasound modulated by the human voice activated the primary auditory cortex (Hosoi et al. 1998) and was successfully recognized by people with normal hearing as well as those whose hearing is totally impaired (Lenhardt et al. 1991)."

 

[JackD201]

In another thread one of our members clearly explained the limitations of tests conducted by audiologists. I believe this is so because the primary purpose of audiologists is not to see the limits but rather to aid us first and foremost for speech recognition/communication. What we have to remember is that hearing is not confined to just our primary hearing organ which is the ear. "Hearing" actually encompasses how the human body reacts to vibrations and that means stimuli from the hair follicles and skin all the way into nerves around our skeletal structure. This bandwidth is not 20Hz to 20kHz. There are many studies being conducted worldwide investigating this but as far as I know no hypersonic studies have yet been fully accepted and are still undergoing peer review. The observations that spurred the investigations/studies however are quite widespread. If we go into infra-sonics we see how people become nauseous or even short of breath when exposed to high SPLs that the are below the average human threshold of hearing. In an extra example of ultrasonics, women undergoing cosmetic ultrasound treatment for fat reduction "hear" frequency sweeps in the 8kHz to 4kHz range.

Now for Ethan to say that bowing technique is a "special effect", especially since he plays the cello is to my mind ludicrous unless his idea of making music is limited to strict adherence to the sheet music leaving no room for interpretation and expression. Still, to call it a "special effect" is still absolutely weird. A special effect is an EXTERNAL manipulation of what the instrumentation put out. A variation of bowing technique is a manipulation of the instrument itself.

As for quoting Wiki....... I think it's okay for casual investigation but everybody knows that Wiki articles are in no way acceptable in any court around the world as proper evidence. A 10th grader could have written that article and nobody would ever know. Having said that, "sub-harmonics" is probably as misnomer. Harmonics are harmonics and sub or not, they DO exist.

There is a major caveat. If one were a headphone listener with sensory input limited to the ears, Hyper and especially Infra- Sonics is a other than harmonics within the band are of little or no consequence. In a recording studio this is the reason monitors are used but headphones are also used for specific purposes usually confined to individual tracks but not for mixing to either mono or stereo.

 

[Phelonious Ponk]

Now for Ethan to say that bowing technique is a "special effect", especially since he plays the cello is to my mind ludicrous unless his idea of making music is limited to strict adherence to the sheet music leaving no room for interpretation and expression. Still, to call it a "special effect" is still absolutely weird. A special effect is an EXTERNAL manipulation of what the instrumentation put out. A variation of bowing technique is a manipulation of the instrument itself.

I took him to mean it is not a normal bowing technique or a normally accepted cello sound. If the cello were a jazz instrument, one might call it "out" instead of a special effect. It's semantics in any case.

There is a major caveat. If one were a headphone listener with sensory input limited to the ears, Hyper and especially Infra- Sonics is a other than harmonics within the band are of little or no consequence.

The consequence is little whether you use headphones or not, but perhaps we're into semantics again.

In a recording studio this is the reason monitors are used but headphones are also used for specific purposes usually confined to individual tracks but not for mixing to either mono or stereo

Yes, mixing with headphones creates some difficulties. In the studio, headphones are primarily used for listening in for the smallest audible details, particularly unwanted noises and distortions. Even nearfield monitors in a well-treated room can't seem to compete for this purpose.

Tim

 

[JackD201]

Hi Tim,

Take flamenco guitar as an example. "Normal" guitar play doesn't include banging on the body which produces not just tones but harmonics lower than an open string. It's still a guitar though and as far as Flamenco guitar playing goes it is quite normal. If the lady found a way to get frequency output lower from her violin has than anybody else has using technique, it's still a violin. A non-traditional application of an instrument yes but not a special effect. Should the technique become widespread, it will just become as normal as anything else a violinist does. I can't see how it can be a matter of semantics.

The consequences in the signal, I fully agree, is little headphones or not. In free field however, during and immediately after final transduction, new harmonics including those that can be called colorations, from such things as, driver flexure or ringing, cabinet resonances and even room resonances can make significant contributions. These are a new set of harmonics created by the back end of a chain nothing even the flattest measuring speaker is immune to. Systems with wide enough FR, high enough SPL capability and rooms large enough for the long wavelengths of LF to develop, fed high resolution files with information in ultra or infra sonic information will also likely generate harmonics within the audio band. This can be a problem in many ways both electrically and acoustically.

MEP pretty much covered it in that harmonics within the audio band are picked up by microphones and do become part of the signal and given any half decent reproduction system should be audible during playback.

 

[Kal Rubinson]

The issue of subharmonics of nominally supersonic sounds is, often, conflated with the interaction of the supersonic tones with ones in the normal range and the resulting intermodulation/beating is in normal hearing range. That is one reason why the addition of hypersonic tweeters is often so obvious despite such reduced sensitivity at those frequencies, especially if the user has not employed proper filtering to separate the bandwidths for each driver. IMHO, in a similar way, there are no subharmonics, per se, but the potential for creating sympathetic resonances, below the fundamental, in other structures and materials. Not that you would want that in a reproducing system.

 

[Ethan Winer]

the interaction of the supersonic tones with ones in the normal range and the resulting intermodulation/beating is in normal hearing range.

Yes, and two frequencies, both ultrasonic, can of course beat to produce a difference tone in the audible range. But the key ingredient is non-linearity. Such beating does not occur unless something is distorting. That "something" could be the amp, the speakers, the ears, or all three.

As far as I know the ear is a low-pass filter that simply can't vibrate faster than it's size and mass allow. So I don't see any benefit to having a response that extends past what humans can hear. It's just wasted bandwidth.

Even if it could be shown that in highly specific cases some people can just barely perceive effects from ultrasonic frequencies, who cares? Clearly this is not needed for a highly satisfying listening experience. I'll never understand why some people obsess endlessly over whether or not we can "feel" ultrasonic frequencies, while ignoring stuff that is infinitely more important. For example, having enough amp power to avoid clipping peaks, or missing that most listening rooms skew the response by a staggering 30 dB or more. Priorities people!

--Ethan

 

[Phelonious Ponk]

I'll never understand why some people obsess endlessly over whether or not we can "feel" ultrasonic frequencies, while ignoring stuff that is infinitely more important.

Because it's more fun to buy super tweeters than it is to regret that your wife won't allow bass traps in the living room?



Tim

 

[Bruce B]

Even if it could be shown that in highly specific cases some people can just barely perceive effects from ultrasonic frequencies, who cares?--Ethan

We care Ethan... and so do the thousands of other people downloading and buying hi-rez material.

 

[Ethan Winer]

Bruce, I understand aiming for the highest quality possible, no holds barred, no cost too high. I'm happy with 99 percent, but that's just me. But nobody has ever shown that we can hear, or perceive, or be affected by frequencies higher than the known limits of about 20 KHz. So why is this still such an issue?

--Ethan

 

[JackD201]

Ummm because the increase in samples that gives you the higher frequencies also gives you more samples in the 20kHz to 20Hz range hence giving more data points to make up the signal ergo better resolution across the entire spectrum.

 

[FrantzM]

Ummm because the increase in samples that gives you the higher frequencies also gives you more samples in the 20kHz to 20Hz range hence giving more data points to make up the signal ergo better resolution across the entire spectrum.

Agree with Jack here ...

Now I am not sure I understand the concept of "subharmonics" , sounds to me (pun intended) as a bastardization of Fourier's ...

My relatively new experience with High Def seems to suggest there are some real advantages with High Def digital.. The future lays in that direction.. The question remains however to how high do we need to go?

 

[JackD201]

The beauty of it is that the sky is the limit now that the files have been freed from a physical medium. Download speed, processing power and memory costs look to be the limiting factors.

 

[amirm]

Ummm because the increase in samples that gives you the higher frequencies also gives you more samples in the 20kHz to 20Hz range hence giving more data points to make up the signal ergo better resolution across the entire spectrum.

That is not true as a matter of math Jack . As long as we have twice as many samples as we have bandwidth, we can fully represent the signal. More samples are redundant and don't contribute to proper reconstruction.

What we can do with more samples and hence bandwidth is to have more variations in reconstruction filter design. Fore example we could have a slow ramping filter close to hearing range and then a sharper one way above. That way the ringing for the latter won't impact the audio range. So in that sense, we can help the lower frequencies but it is not due to having more samples per-se .