R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 [2] 3 4 ... 10   Go Down

Author Topic: the "high frequency transients" fallacy  (Read 53938 times)

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: the "high frequency transients" fallacy
« Reply #15 on: March 01, 2005, 12:58:44 PM »

Joe Bryan wrote on Tue, 01 March 2005 05:54

Hi Dan-

I'm not quite ready to accept your assumption that the ear has the band-limited reponse your argument requires.

I'm not an expert on human perceptual physiology, so I'll leave this point for others to debate. However, I know how circuits behave, and if the physical issues I've mentioned have analogues in our hearing system, I wouldn't be very surprised.

In the mean time, I'll continue to believe my own ears when it comes to audio quality issues.

Cheers,

-Joe



Hi Dan-

I'm not quite ready to accept your assumption that the ear has the band-limited response your argument requires.

I'm not an expert on human perceptual physiology, so I'll leave this point for others to debate. However, I know how circuits behave, and if the physical issues I've mentioned have analogues in our hearing system, I wouldn't be very surprised.

Cheers,
-Joe

Your statement is that you understand circuits, therefore it gives you a sense of the mechanics is less that half baked. You just claimed some knowledge and left it at that, no explanations, and I think your statement actually helps “my case”.

I too know about circuits, and IF the physical issues you mentioned are indeed analogues to hearing system, that only makes my statement more correct! I do not know if that ear is behaving like a circuit, and making such assumption is going very far in my book. But microphones, speakers and dynamic mechanical systems have been modeled by electrical circuits, such as in the era of analog computers. I believe there was a time when automotive suspension was modeled by electrical means…  

Let me state some of the principles that make such analogy work:
A network of electrical circuit made of R,L and C is an expression of the relationship between voltage and current flow over time where:
Resistance relation of curent to voltage is proportional
Inductance has to do with rate of change in current
Capacitance relates to to voltage by accumulation (integration).

We can interconnect those elements in any configuration we wish, and it all boils down integral-differential equations for the circuit behavior, all with respect to time.

Now let us take a mechanical system. Making the ratio between distance of the motion over time (velocity) is analogous to electrical current
The mechanical spring action behaves like an electrical capacitor.  
The mass is relates to the derivative of velocity, acting like inductance.
The damping (such as viscous friction) is a derivative function (like electrical resistance).

And before you raise objections, all non linearity, gain and other factors can be modeled by electrical circuits. It is a difficult job, but electrical theory is under the same “bigger umbrella” of the “principles of nature”.

So in fact, the concepts of poles and zeros are not limited to electrical systems. It is more common to talk about poles, zeros, resonance, bandwidth and more in the area of electronics. But the fact is that microphones and speakers also have poles and zeros, and so do relays and motors… If a microphone has say a 3dB mechanical attenuation at 20KHz (the membrane material and air around it does not alow response to such fast variations), it also has a phase shift associated with that attenuation. A dynamic mechanical system is not unlike an electrical one!

In my view, anyone that does understand circuits and signals is in a position to see how steady state mechanics IS VERY MUCH CONNECTED to transient response. Why are you viewing them as 2 seperate worlds?

Let us forget about mass, spring coefficients and chock absorbers. Understanding of analog circuits requires knowledge of the strong connection between transient and steady state behavior! An impulse response is an expression of the bandwidth!!! Given the bandwidth, your staedy state response is limited.

And Vica versa!!! Given a bandwidth, the impulse response is cast in stone! Lower bandwidth sets a limit ob impulse speed (maximum rate of change)!!!

This is true for electrical systems, and for dynamic mechanical systems. Is it true for the ear? I do not know.

But since you wish to connect the ear to electrical systems, your “circuit understanding” should immediately put you in my camp!

Regarding your last comment: In the mean time, I'll continue to believe my own ears when it comes to audio quality issues., I really do not see how it ties in to this thread.

Regards
Dan Lavry  
www.lavryengineering.com


Logged

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: the "high frequency transients" fallacy
« Reply #16 on: March 01, 2005, 01:20:39 PM »

RobertRandolph wrote on Tue, 01 March 2005 13:24

I can say by all hearing tests Ive been subject to by licenses audiologist give me an average ability to hear.

BUT. I have done blind tests with monitors (specifically earthworks) that can pass frequencies above 20khz. Blind folded, and simply passing fullscale 25khz and 28khz signals at 96khz sample rate with apogee converters (not at my studio). Monitor sysetm was calibrated to the k-system  k-14.

Every single person was able to tell when the signal was on. And I can describe it very easily as simply feeling a change in the air. I cant claim to have "heard" it, but I know that of the 4 people there who participated in the test there was not a single incorrect answer... I dont recall the exact number of repititions but it was above 100.

I cant say for sure the result was due to resonances from the speaker mechanism in the audible band or simply the ability to feel the minute change in pressure by some human mechanism. Either way I have been convinced more than once that in many modern listening systems, frequencies above the audible range have an impact on the listener. Perhaps by imperfect listening systems, or by human ability. Regardless it exists.

And I invite anyone to try it.


I certainly belive that people "heard it". My point is simple - what ever you heard was within the audible range of human hearing. If you put a 50KHz tone or even 1MHz tone into the system, there are numerous ways it can find a way to modify the energy content at lower frequencies that we hear.

I do not wish to offend anyone, but I heard people reporting their ability to hear high frequencies where the source material was a tape (at about 20KHz), the original mic was probably not more then 20KHz and the speaker was FOR SURE limited to less then 20KHz.

So just because one hears a difference, does not make it due to high frequency sound. How does anyone with an ear decide that what they hear is really high frequency? Is it because the gear they listen to has 192KHz painted on it? Could it be say an alteration of the harmonic structure? Could it be a change in attack plus modulation (or less modulation) of the noise...
The for musical alterations within the human hearing range are endless!

regards
Dan Lavry
www.lavryengineering.com
 
Logged

RobertRandolph

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 54
Re: the "high frequency transients" fallacy
« Reply #17 on: March 01, 2005, 01:58:09 PM »

Dan I agree with you (I even stated what you said, briefly, in one of my other posts).

But if higher frequencies do have an effect on reproduction systems, do you not think more attention should be paid to them? I know this thread is about the idea of HEARING high freuency transients.... but I think we most certainly can hear the imperfect (and within the audible range) reproduction of any audio material outside the audible range.

Perhaps this is why I have alwywas preffered lower sampling rates (44.1/48). It just sounds.... right. Higher sampling rates tend to sound artificial and unusually good... in no way representing what I really think I recorded.

Either way... dont want to veer this too off topic.
Logged
Now I know a disease that these doctors can’t treat
You contract on the day you accept all you see
Is a mirror, and a mirror is all it can be
A reflection of something we’re missing

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: the "high frequency transients" fallacy
« Reply #18 on: March 01, 2005, 06:34:56 PM »

RobertRandolph wrote on Tue, 01 March 2005 18:58

Dan I agree with you (I even stated what you said, briefly, in one of my other posts).

But if higher frequencies do have an effect on reproduction systems, do you not think more attention should be paid to them? I know this thread is about the idea of HEARING high freuency transients.... but I think we most certainly can hear the imperfect (and within the audible range) reproduction of any audio material outside the audible range.

Perhaps this is why I have alwywas preffered lower sampling rates (44.1/48). It just sounds.... right. Higher sampling rates tend to sound artificial and unusually good... in no way representing what I really think I recorded.

Either way... dont want to veer this too off topic.



At this point I am concentrating on pointing out that bandwidth is bandwidth, no more no less. I am pointing out that one does not get to arbitrarily have 2 simultaneous sets of bandwidth – one is backed by scientific study of the human ear and the other is for BS purpose.

The people that want, for whatever reason, to convince you that you hear very high frequencies, know very well that any audiologist can prove them wrong by doing a standard listening test! You know, that test, where you have a sine wave played into a headphone, and the frequency is being raised until you can not hear it….

So the only way to “stay the course” and convince themselves and others that you can hear higher is to find a fault in the audiologist test. My best guess is that this how the concept of high frequency transients came about.

What would be the nature of a “high frequency transient”, that is a transient where ALL the energy is above say 20KHz? It has to be faster then a 20usec impulse (narrow), and it would need a lot of time settle (or wait) at a near zero value. A wider impulse is of course not allowed because it contains audible energy. Also, you will not be allowed to introduce another such “animal” for a long time, in order to prevent audible activity above a given level (be it -60dBFS or -120dBFs or whatever dynamic range you wish for). Adding a new impulse before the previous one "settled" will spill energy into the audio band.  

For example, say you are to create a file for your say 96KHz DA, with all zeros data except a single one sample at full scale, You play that and that impulse is too high to hear. But say you add just one more sample of full scale at 1 msec later, the sum of the 2 impulses does spill over to the audio band, stating at 0Hz all the way up to way over 20KHz. And of course it will not be a very loud activity – which is the nature of narrow impulses, which of course only strengthen my argument. In other words, the more high frequency transient energy you wish for, the more often you want the transients to occur which in turn spill more energy into the low frequencies. That whole concept of high frequency transients is flawed.

Is anyone arguing that they hear high frequency transients occurring very rarely? A systematic (periodic) set of narrow impulses at say 1KHz yields 1KHz energy and a lot of harmonics (2,3,4Khz…). Say we go for only 20 such transients per seconds. That will yield 20Hz fundamental and 40.60,80hz… Of course you will call it periodic activity, not high frequency transients. I agree. So lets “break up” the periodicity and go for say 20 impulses per second spaced at a random interval. That will result in random noise over the audio band!  

The bottom line: it is easy to come up with steady state high frequency stuff beyond hearing. But having transient (non steady state) high frequency signals with all the energy above the hearing region places some heavy restrictions on the signals, both in terms of duration (must be very narrow) and allowed frequency of repletion (must be rare), to the point of making the whole concept ridicules.  

Of course, many people, including EE’s that do not really get the theory, will just insist on getting the conversation into hearing and being open minded. It would be good if they were "open minded" to the fundamental physics and math that are used to convey scientific reality of real world occurrences.

It was a long and wordy one. I could have said it in a couple of equations, but I want the concepts understood by more than a hand full of people.

Regards
Dan Lavry
www.lavryengineering.com
Logged

RobertRandolph

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 54
Re: the "high frequency transients" fallacy
« Reply #19 on: March 01, 2005, 08:34:50 PM »

Sahib wrote on Tue, 01 March 2005 16:02

RobertRandolph wrote on Tue, 01 March 2005 17:00

I dont think anyone claimed we could hear sound above 20khz,

RobertRandolph wrote on Tue, 01 March 2005 13:24

I

Every single person was able to tell when the signal was on. And I can describe it very easily as simply feeling a change in the air. I cant claim to have "heard" it, but I know that of the 4 people there who participated in the test there was not a single incorrect answer... I dont recall the exact number of repititions but it was above 100.







A bit of contradiction there, I think.

Regards,
Cemal



Hearing, and perception are different things. Hearing is part of perception, but not all perception is hearing.

Perhaps you are not familiar with using quotes to insinuate that the word in quotes is not being used correctly, but is instead a suitable word not in meaning, but in intent.
Logged
Now I know a disease that these doctors can’t treat
You contract on the day you accept all you see
Is a mirror, and a mirror is all it can be
A reflection of something we’re missing

RobertRandolph

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 54
Re: the "high frequency transients" fallacy
« Reply #20 on: March 01, 2005, 08:41:55 PM »

Dan, I am not trying to insinuate that we can hear frequencies above 20khz. I strongly beleive that is not true.


But I think one, or both of these statements is correct and has great impact on what we, as recording engineers (we ARE the ones using the products) do.

Correct: Frequencies outside the audible range can manifest themselves in mechanical reproduction of themselves within the audibel range... i.e. we cant hear 30khz but a 30khz signal that is produced through a modern reproduction system is audible due to the various distortions in the system that create sound in our audible range.


possible: simply, we can somehow perceive (but not necassarily hear), pressure changes in the air occuring faster than 20,000 cycles a second.

I think this is very close to turning into a purely semantic discussion. If one is to look at the meaning behind the words of those involved.. it seems we are all saying the same things from a different perspective.
Logged
Now I know a disease that these doctors can’t treat
You contract on the day you accept all you see
Is a mirror, and a mirror is all it can be
A reflection of something we’re missing

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: the "high frequency transients" fallacy
« Reply #21 on: March 02, 2005, 11:53:24 AM »

I did not state high we hear. I did not say that we can not hear or feel beyond 20KHz. It is not common but there are some people that can hear over 22KHz in an audiologist test. There were some less well substantiated claims about "feeling" signals at 26Khz, and I am not at all commenting on the specific limit, be it 20Khz, 30Khz or more.

What I am saying is: If your mic can pick up to say 30KHz, if your speaker can play 30KHz, if the rest of your gear can accommodate 30KHz, then it would be doing it for steady state signals as well as transient energy. There is no such thing as 2 separate bandwidth, one for steady tone the other for transients.

I am also saying that the concept of hearing high frequency transients is ridicules, because the combined impcat of such transients would spill over to the audio band. The only way to avoid such occurrence is to: a: have the transients very narrow (short lived in time) b. There should be "a lot!" of time separating the transients.
If you wish for a clean audio bandwidth say down to 20Hz, you are not allowed to have the transients nearer than about 50mesc (a maximum of less then 20 such transient a second).

Why are we not having a listening test to see if we can isolate our ability to listen to high frequency transients? Because you first have to generate a signal containing ONLY high frequency transients.  There lies the problem! The signals will contain audio range energy.

Again, it is easy to generate high frequency steady state signals. If you want to have a signal that carries ONLY high frequency, and you do not want it to be a steady tone, you must settle for few and far between very narrow impulses. Is that what we are talking about? A musical content made of such “elements”?

Say we want to test audibility of high frequency transients. We need say a narrow impulse generator with controlled rate, preferably with a randomizer feature. First we adjust a single impulse (say at impulse one per 5 seconds) to be so narrow that we can not hear it. Next we start raising the repletion rate (it can be randomize to avoide steady state behaviour). Guess what, at some point we will hear those impulses. The energy we hear will be in the audio band!

I know we are talking about analog, but think of conversion for a second. We are able to “register” music with a bunch of narrow impulses. Each impulse is very narrow. But setting them side by side, you can have one big audio signal!  

Therefore, the subject of how high we hear, or feel (be it 20Khz, 30KHz or what not) is a different subject.

But IF we hear say 30KHz, or feel it, we can not say it is not due to “high frequency transients” (above where we hear steady state), because such transient activity is coupled with a lot of activity in the lower frequencies (unless we go out of our way to prevent it).

This thread is about the “notion” of hearing or processing “high frequency transients” above “the rest of the music”.

Regards
Dan Lavry
   

Logged

Joe Bryan

  • Newbie
  • *
  • Offline Offline
  • Posts: 19
Re: the "high frequency transients" fallacy
« Reply #22 on: March 03, 2005, 01:14:42 AM »

Dan-

You're making a lot of assumptions about the human auditory system without backing them up with reviewable reseach.

Your argument would hold *if* the ear were a simple linear frequency analyzer (like an FFT analyzer), but it most certainly is not.

Irino and Patterson "A compressive gammachirp auditory filter for both physiological and psychophysical data" (JASA 2001 May), and Duke and Julicher “Active travelling wave in the cochlea” (Phys. Rev. Letters 90, 2003) model the resonant hair bundles in the cochlea as nonlinear sensors whose bandwidth and sensitivity vary over time with the signal.

Also, Krumbholz et al "Microsecond temporal resolution..." (JASA 2003 May) show that temporal resolution is as little as 10-20 microseconds.

As I said earlier, I'm by no means an expert in human auditory physiology, and I respectfully submit that you are not either.

I am not arguing that "high frequency transients" is a valid concept or not. I'm simply stating (again) that there is a *lot* more to it than you claim, and that I'm unwilling to dismiss this concept based on your argument.

If you want to make your argument, you will have to try a lot harder than with the ultra-simplistic linear analysis presented so far.

-Joe

Joe Bryan
VP Engineering
Universal Audio

"EQ, IQ, whatever it takes." -Jack Hotop
Logged

maxdimario

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3811
Re: the "high frequency transients" fallacy
« Reply #23 on: March 03, 2005, 05:40:50 AM »

The way a system reproduces transients is essential to realism and imaging.
I don't believe ultrasonic bandwidth is of prime concern, because a fast accurate amp that reproduces the wavefront of transients accurately (gives a clear indication of when the transient begins -- in time -- and how it develops...or the 'shape' of the transient) will work even if the signal is limited at 15 KHz.
I have done experiments, and if the amp is clean and simple with little or no feedback, or implements feedback that does not pass through a capacitor, it will retain 99 per cent of the 'transient' or imaging information, even with a 12 db/octave 15KHz rc filter.

The importance of transient reproduction is paramount because the brain localizes objects through the interpretation of transients.

but maybe the reproduction of 'impulses' is also a correct way of defining what audiophiles try and preserve in recording/reproduction (strangely few 'modern' recording engineers are conscious of this important issue).

dolphins can 'see' the shape of objects in water by analyzing reflected impulse waveforms.

I believe they have a very large part of their brain that is dedicated to analyzing impulses, transients, or generally speaking sonic localization information.

It's no wonder this is a controversial and misunderstood issue, because for the last 20 or thirty years studio equipment has had terrible 'transient' or 'impulse' reproduction.

this is due to the ssl type op-amp based desks which lose almost all of the fine detail of the aforementioned type. The general marketing-oriented race for ultra low noise and distortion specs (which are a lot easier to obtain than good audio) has also made matters worse.

consider that passing through electrolytics will distort the above information considerably, and since modern equipment has tons (and if they age they become worse) the output has little to do with the input.

The ears know best.

Logged

David Satz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 661
Re: the "high frequency transients" fallacy
« Reply #24 on: March 03, 2005, 08:38:13 AM »

Joe, the "20 - 30 microsecond resolution" figure that you quoted--wasn't that the finding for minimum perceptible interaural time difference (i.e. time slippage between channels)? I seem to recall that it was.

It's an extraordinary finding in its own right, but the ability of our nervous systems to respond to tiny differences in signal onset time between our two ears says very little about the audio bandwidth to which those ears can respond. Two versions of an audio signal that are identical but 20 microseconds apart in time can be put through a pair of identical 3 kHz low-pass filters--digital or analog--and they will still be identical (and 20 microseconds apart) when they come out the other side. That doesn't depend on super-high sampling rates, either.

I frankly think that you and some other people here are steering the discussion into the much broader area of psychoacoustics where you can say (with perfect honesty and certainty) that research has turned up interesting, divergent results, and that none of us really knows enough.

But if you take the simple statement which Dan made, and don't transpose it into other areas of discussion, I think you have to admit that he's put his finger on a myth which many people cling to. This apparently makes certain people rather uncomfortable, as perhaps it should.

--best regards
Logged

zmix

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2828
Re: the "high frequency transients" fallacy
« Reply #25 on: March 03, 2005, 09:07:54 AM »

RobertRandolph wrote on Tue, 01 March 2005 13:58

But if higher frequencies do have an effect... do you not think more attention should be paid to them?



I would like to provide a simple illustrative analogy here.

The reported lower range of human hearing ends at 20cps, the eardrum simply does not respond below that.

The tremolo circuit on a fender guitar amp can operate in the 3-8cps range.
We can certainly hear the effect that this modulation has on the audible signal, but this is NOT the same as saying that we can 'hear' 3cps.

Now, at the other end, a microwave oven operates at a VERY high frequency, so high that it causes water molecules to resonate, hence produce heat. This could be misstated by saying that this PROVES that our bodies respond to extreme HF energy...

food for thought...

-CZ

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: the "high frequency transients" fallacy
« Reply #26 on: March 03, 2005, 11:38:24 AM »

[quote title=Joe Bryan wrote on Thu, 03 March 2005 06:14]Dan-

Your argument would hold *if* the ear were a simple linear frequency analyzer (like an FFT analyzer), but it most certainly is not.

Also, Krumbholz et al "Microsecond temporal resolution..." (JASA 2003 May) show that temporal resolution is as little as 10-20 microseconds.

-Joe

First, my statement about impulse response being a time domain representation does not assumes linearity, I did not assume sinc function impulse (brick wall transfer bandwidth) either.

Second, if I recall, that usec resolution has to do with delay, not with bandwidth. Yes, I accept that the ear is very sensitive to sudden change in delay, even in the few usec range

Third, you stayed away from the point that I made about signals containing high frequencies ONLY, with no energy in the audio band. That has nothing to with linearity.

Forth, I believe when I stated the equivalence of LRC to to mass, spring and viscosity, I did also state while complex, the modeling of non linear mechanical dynamics can be done with non electrical linear circuits. If I did not say it clearly before, I am saying it now.

I never think of the ear as an FFT! Why suggest it? It shows me that you really do not at all understand what I said!

I read some but not all of the papers you pointed out. I would rather not have a technical discussion where the response is "So and So wrote a paper, go read it, you are wrong".

I came up with a few arguments, such as the impulse IS bandwidth. Another is that high frequency energy without the lower frequency requires meeting some very stiff requirements. I showed what the fastest signal within a given bandwidth is - the slew rate of the highest frequency sine wave within the bandwidth (2*p*f*V) and also said a couple of other points.

As far as I am concerned, ANY ONE OF THE POINTS I RAISED is SUFFICIENT to shoot down that hearing "high frequency transient, above the normal audio". You have to be able to dispute ALL my points, not just one or two. So far you have not disputed any, in my view.  

Start at the beginning: Show that impulse and bandwidth are not the same, and you can transfer a narrow impulse through a band limited system. Doing so will REMOVE all the high frequencies above the bandwidth limitation, and what remains is the portion of the energy in the impulse that is within the bandwidth.

There is nothing simplistic about that. There is something fundamental about it!

If you read my paper "Sampling Theory" you will see that a sinc impulse is the sum of all the cos waves (infinite numbers of them) over the bandwidth) at equal amplitude. If you change the shape of the impulse from sync to say anything you will end up with a transfer function different from a brick wall, but the impulse is still an expression of bandwidth and Visa versa.

Assuming you can pass that impulse bandwidth argument, you have to take on the next argument about the signal characteristics of "high frequency transients without low frequency content".
After you "combat" successfully my assertion that such signals must occur below some maximum rate (very low rate), or explain why some 20-50 or so very narrow impulses are a part of the music made by the musical instruments, lets see an argument that shows what is wrong with my first post about maximum slew rate within bandwidth.

Then, last but not least, you talk about the complexity of hearing. This is a technical forum, where we do not just brush thing off as "the ear is complex". Lets talk about how to test the assertion that people can hear high frequency transients above normal audio.

Like Rich Cabot said: "IF you can demonstrate it reliably..."
I do not see it being demonstrated reliably. You want to bring up papers and articles? Where is a single credible listening test that shows that people can hear high frequency bandwidth above "normal audio"? What was the source material for such transients that was free of lower frequencies? I am not holding my breath...  

Regards
Dan Lavry
www.lavryengineering.com



 
Logged

Joe Bryan

  • Newbie
  • *
  • Offline Offline
  • Posts: 19
Re: the "high frequency transients" fallacy
« Reply #27 on: March 04, 2005, 08:51:21 AM »

Dan-

Just a quick note...

First, I interpret the phrase "high frequency transients" to mean "non steady-state signals with high-frequency energy". Your initial post in this thread uses band-limiting to demonstrate how rise-time is a simple function of the high-frequency energy, and goes on to show how band-limiting limits this energy and thus the rise-time. Simple enough.

The basis for your initial argument was that since you can't hear energy above a certain point, and because neither instruments nor transducers can reproduce these (nonexistant) signals, there's no need to account for them because they don't exist nor need to exist for perceptual purposes. This is where I disagreed, and sited examples why this is true (human perception is sensitive to HF energy, equipment and instruments can produce these signals, and systems need to handle them for various real-world reasons).

It appears you're now using the phrase "high frequency transients" to mean something else.

-Joe
Logged

zmix

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2828
Re: the "high frequency transients" fallacy
« Reply #28 on: March 04, 2005, 09:36:55 AM »

Joe,
Can you point to some research that would indicate that these transients contain energy beyond the audio passband?

-CZ

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: the "high frequency transients" fallacy
« Reply #29 on: March 04, 2005, 04:00:40 PM »

Joe Bryan wrote on Fri, 04 March 2005 13:51

Dan-

Just a quick note...
...It appears you're now using the phrase "high frequency transients" to mean something else.
-Joe




Joe said:
First, I interpret the phrase "high frequency transients" to mean "non steady-state signals with high-frequency energy".
“It appears you’re now using the phrase “high frequency transients” to mean something else.”


Dan Says:
Yes, "high frequency transients" means "non steady-state signals with high-frequency energy"

However, one must remember to MAKE THIS DISTINCTION: “high frequency transients above human hearing” means “non- steady state signals CONTAINING ENERGY AT FREQUENCIES ABOVE HUMAN HEARING.”  

Once again, bandwidth defining what you hear (whatever the specific number is), sets a limit for musical content (signal) transient or not. The highest frequency transient is not higher than the highest frequency steady state.

Transients (made by musical instruments) containing high frequencies beyond human hearing ALSO contain energy at frequencies within human hearing. It is impossible to claim that sonic alterations are due to high frequency when introducing such energy also alters the lower frequencies.

Dan Lavry
www.lavryengineering.com

Logged
Pages: 1 [2] 3 4 ... 10   Go Up
 

Site Hosted By Ashdown Technologies, Inc.

Page created in 0.061 seconds with 21 queries.