Music Player Network Home Guitar Player Magazine Keyboard Magazine Bass Player Magazine EQ Magazine
Page 40 of 42 < 1 2 ... 38 39 40 41 42 >
Topic Options
#442559 - 01/13/04 11:27 AM Re: George, Watch this!!!....(96k)
Nika Offline
MP Hall of Fame Member

Registered: 03/22/01
Posts: 2938
Loc: Ft Wayne,IN,UNITED STATES
[QUOTE]Originally posted by David Sykes:
Moreover, (someone please re-direct me if I’m off base but…) my take on reading the interchanges between Paul Frindle, Hutch, Nika and a few others back in 8/2001 is that further refinements in techonology (convertors, filters, etc) are required before we can adequately study the impact/role of frequencies outside of 20hz-20khz on human perception.

David,

I'm not exactly sure what was said that implies this to be the case? One thing we have been saying is that any audible difference between base rate and extended sample rate responses does not come from the audibility of extra frequency content so far as we have been able to conjur, nor does it come from an inability for base rate responses to accurately capture audio in the Nyquist range.

Now one thing we discussed is that converter chips that are good enough to demonstrate this fact are difficult to come by, and therefore the converter itself (the filters within it) impedes the ability to accurate analysis on the subject when doing listening tests.

There is a difference, though, in saying that we need better converters in order to determine appropriate sample rates (which encompasses many issues) and saying that we need better converters in order to determine the threshold of human hearing (a subset of the former issue).

Nika
_________________________
For more...

Top
#442560 - 01/13/04 11:35 AM Re: George, Watch this!!!....(96k)
Nika Offline
MP Hall of Fame Member

Registered: 03/22/01
Posts: 2938
Loc: Ft Wayne,IN,UNITED STATES
[QUOTE]Originally posted by Bobro:
Perhaps I didn't make myself clear on the other thread- I listened to the group warming up before the concert at both 48k and 96k (Sek'd 2496, Sennheiser HD600) and made the decision at that point to record at 96k because of the difference.

So you were switching the sample rate on a given converter between 48k and 96k and you heard a difference? That tells us something about that particular converter but does not tell us anything about the inherent differences between sample rates.

Hopefully I did not say anything to give the impression that I'm running under that notion (cool expression), because that's the complete opposite of my feelings about "accuracy".

Ahh, OK. Then I'm with you. Using a poor playback system one can certainly get more distortion with 96KS/s sample rates. If you are interested in using high sample rates as an effects processor, specifically to impart non-linearities, then I see no dispute. I had thought we were talking about linear capture of audio, and my earlier points were related to this.

Nika.
_________________________
For more...

Top
#442561 - 01/13/04 11:40 AM Re: George, Watch this!!!....(96k)
Nika Offline
MP Hall of Fame Member

Registered: 03/22/01
Posts: 2938
Loc: Ft Wayne,IN,UNITED STATES
[QUOTE]Originally posted by NYC Drew:
Maybe the issue then is that we have become way too fixated on how the ear works in isolation.

Drew,

I'm with you on this idea. There is certainly the opportunity for the brain to receive information from other parts of the body. The only evidence I've found so far (discussion on bone conduction, sonar, and more) is that either the levels have to be so high as to be out of the realm of what we talk about with music performance, or the brain, while perhaps getting the information, can do nothing with it and does not recognize its presence (the results of the Oohashi study).

But I find this question to certainly be valid.

Nika.
_________________________
For more...

Top
#442562 - 01/13/04 11:46 AM Re: George, Watch this!!!....(96k)
skiboy Offline
Senior Member

Registered: 07/08/03
Posts: 37
Loc: The Pole
...so humans NEED DISTORTION to see, hear, feel comfortably...
Any stimuli that is mathematicaly perfect=ACCURATE is boring our perception ?
Nika states that, if some data doesn't reach the cerebral complex responsible for its interpretation, our perception will not register or process this data.

But how about the fact that the brain actually CREATES images, sounds, smells with no data input.
Recall a case patinet with a part of his eye's retina removed.
A while after the operation, he was seeing a grey "hole" in his world. But after some time, images began appearing in this hole; faces or architectural details mostly. These ghost images were an effort of his brain to "fill in" the missing parts of his vision.
Now consider cases of aphasia - due to internal brain damage, some nerve signal paths between cognitive regions get severed or rerouted. In some cases of aphasia the brain tries to "reconnect" these impulses by providing an auxiliary "imaginary" input to the cut-off regions.
While this is exciting for linguistical studies, it also bears relevance to how we percieve sound.
Or sight....the visual data reaching our brain (eye is also brain) is distorted, upside-down, probably black&white with some hue, the picture is smaller tha a post stamp and very dark. Thus the amount of processing to give us our succulent, 3D vision is AWESOME. This means that we can easily sense a movemnt within or just outside our vision and react in a split second WITHOUT even registering what is it, how or exactly where the movement comes from. So we react to the impulse, not to our perception of the impulse because the tiger might eat us while we are busy deciding what is it that we are perceiving.
:p
Can this be an analogy to how we hear ?
And could this argument be used by armchair scientists to support their "romantic" view of audio phenomena ?

Bill's work with autistic children would suggest that, while data might flow uninterrupted to the child's brain, it does not get interpreted correctly. Some parts of the sound are amplified and some are completely dampened out within 1/12 of an octave !

Also, I find the notion that high frequency audio components in our environment distort our cognition VERY interesting. I will try to study this further.

I respect Nika Aldrich's work and value the time he spends to educate us. I would never insult people who spent years of their life studying a problem and trying to explain that to others.
I hope i just didn't.

Top
#442563 - 01/13/04 11:53 AM Re: George, Watch this!!!....(96k)
Nika Offline
MP Hall of Fame Member

Registered: 03/22/01
Posts: 2938
Loc: Ft Wayne,IN,UNITED STATES
Quote:
Originally posted by Bill Mueller:
Drew,

Bone conduction, albiet at a much lower volume, has been shown to conduct frequencies to the brain above 40khz. I wish we had ol' Rupert in on this conversation.

Bill
Bill,

Bone conduction at a much higher volume than is found in music reproduction - has been shown to conduct information to the brain. From my understanding, though, what the brain "hears" in this situation is not information above 20KHz, but rather the lower harmonics that help with speech intelligibility and comprehension. We can think of this as using the "missing fundamental" principle, or possibly intermod distortion that actually resonates the cochlea at below 20KHz.

The bone conduction in terms of decibels through the skull is fairly well understood from the reading I've done. The amplitude of audio that we hear "directly" (that being into the ear canal, through the middle ear, through the oval window, into the cochlea as fluid waves, then conversion) versus the audio that we hear "indirectly" (through the bones and organs of the body, which eventually cause fluid resonation in the cochlea) is documented.

Of course some situations create more indirect audibility than others. For example, stuff your ears with cotton and then tap on your forehead. Or plug your ears and cough. We hear those sounds more indirectly than directly. But in a situation where we are listening to typical audio, even at significant volumes, the level of transmission to the ear through bone is so low as to fall below the noisefloor of the ear, and does not play a part in our hearing - from the research I've done. One has to remove opportunity for direct sound in order to allow the indirect sound to be consequential enough to be audible. This is why bone conduction is used as in therapy for those with significant hearing loss, depending on the nature of the loss.

Of course, all of that is from the research I have done.

BTW, I've seen Mr. Neve discuss the topic of high frequencies many times and have not found his points to be anything beyond speculation.

Nika.
_________________________
For more...

Top
#442564 - 01/13/04 11:57 AM Re: George, Watch this!!!....(96k)
Nika Offline
MP Hall of Fame Member

Registered: 03/22/01
Posts: 2938
Loc: Ft Wayne,IN,UNITED STATES
Quote:
Originally posted by edmann:
Perhaps the most immediately useful part of this info is to find one's way out of manufacturer marketing mumbo-jumbo. Bob Ohlsson says that 96k converter chips offer significantly improved performance over their previous generation 48k counterparts w/re to jitter and accuracy (I am paraphrasing him). I have no reason to doubt this, and so given this scenario - it may be the case that one ~should~ use the newer technology but not run it to it's limits.
Ed,

There is no question that converter technology has improved. Converter chips that have been made in the last three years exceed the capabilities of most converters of years gone by. Using modern converters will almost inherently help the sound of recordings (all things being equal). This indeed has nothing to do with the sample rates they use - it has to do with the onboard processing, filtering, and other issues.

Nika.
_________________________
For more...

Top
#442565 - 01/13/04 12:06 PM Re: George, Watch this!!!....(96k)
Nika Offline
MP Hall of Fame Member

Registered: 03/22/01
Posts: 2938
Loc: Ft Wayne,IN,UNITED STATES
[QUOTE]Originally posted by skiboy:

And could this argument be used by armchair scientists to support their "romantic" view of audio phenomena ?


Skiboy,

I like your way of putting this.

The interesting question to me is why so many people are so pulled to the desire to have there actually be benefits to high frequency sampling. Why do we want to believe that the ear is limitless, that our capabilities about beyond what logic, reason, science, and research indicate? Why do we yearn to know that we haven't found the limits? What is it in the human psyche that keeps trying to find holes in the research that defines our capabilities?

To me, this desire is a part of a Judeo-Christian-western mentality that actually taps into a religious issue. To me it is a God issue - an evolutionary characteristic of humans to want to feel that we can't "know" things, for "knowing" things starts to make our lives more clearly less significant than we like to think. My thoughts on this subject are long and complex, and based on years of thought and observation but little "research" per se.

The compelling issue to me on this whole subject is not the issue of benefits or lack of benefits in high sample frequencies, but rather the desire by the majority of people to FIND benefits in higher sample frequencies. I find this utterly fascinating. I'd be interested in hearing what others have to say on the subject.

Why is it that we glom on to these "romantic" notions? Why do we seek out this "romance?"

Also, I find the notion that high frequency audio components in our environment distort our cognition VERY interesting. I will try to study this further.

You don't have to look far for one example: what happens when you push a whole lot of high frequency content (say, around 50KHz) into a speaker that is designed to go up to 20KHz?

Nika
_________________________
For more...

Top
#442566 - 01/13/04 12:55 PM Re: George, Watch this!!!....(96k)
cram Offline
Senior Member

Registered: 05/23/01
Posts: 493
Loc: sandy,UT,UNITED STATES
Or it could be something a bit more down to earth.

"My mix sounds like crap, it must be my gear."

Uh-huh. \:\)
_________________________
Tom Cram
dbx Senior Technical Support
(801) 568-7530
tcram@dbxpro.com

non fui, fui, non sum, non curo

Top
#442567 - 01/13/04 01:19 PM Re: George, Watch this!!!....(96k)
edmann Offline
Senior Member

Registered: 08/26/03
Posts: 314
Loc: MA
Quote:
Originally posted by cram:
Or it could be something a bit more down to earth.

"My mix sounds like crap, it must be my gear."

Uh-huh. \:\)
Conversely, some of the more naive amongst us (me included) may have spent weeks trying to squeeze quality performance out of gear that is just is not capable of producing quality results. That is one way to learn.

no reflection upon dbx intended, I am speaking of other gear.
_________________________
Ed Mann

Top
#442568 - 01/13/04 01:32 PM Re: George, Watch this!!!....(96k)
cram Offline
Senior Member

Registered: 05/23/01
Posts: 493
Loc: sandy,UT,UNITED STATES
Quote:
Conversely, some of the more naive amongst us (me included) may have spent weeks trying to squeeze quality performance out of gear that is just is not capable of producing quality results.
I'm with you, I've been there.

I just find the appeals to mysticism alarming. As Nika has stated our knowledge of the mechanics of the ear is pretty well down. So instead of discussing inadequacies of converter design, bone induction and possible 50k ripple is brought up.

Just trying to keep it real. \:D \:D \:D
_________________________
Tom Cram
dbx Senior Technical Support
(801) 568-7530
tcram@dbxpro.com

non fui, fui, non sum, non curo

Top
#442569 - 01/13/04 02:31 PM Re: George, Watch this!!!....(96k)
StoneinaPond Offline
MP Hall of Fame Member

Registered: 11/22/02
Posts: 2009
Loc: New York City
Quote:
So instead of discussing inadequacies of converter design, bone induction and possible 50k ripple is brought up.
So we go back to the question of what part of a good converter, that just happens to sample at 96, provides the increased joy or whatever it is that we hear?

I know some say they have listened to the same converter at 48 and 96 and hear no difference. Is that a true scientific test?

Because if you do hear a difference, how can you isolate which part played the bigger role in that equation?
_________________________
Yorik
Stone In A Pond


"Man will occasionally stumble over the truth, but most of the time he will pick himself up and continue on."

Top
#442570 - 01/13/04 06:17 PM Re: George, Watch this!!!....(96k)
blairl Offline
Senior Member

Registered: 04/03/00
Posts: 334
Can anyone give some examples of converters that sound the same at 48 and 96K? I have seen mentioned many times that a good converter design will sound the same at 48 and 96 K but haven't seen mention of any brands or models that actually do this. Paul Frindle has said that the Oxford converters are not discernable from the source. You can't beat that. I have seen it suggested that some db technologies or Lavry Engineering converters might sound the same at 48 and 96K but haven't seen anyone definitively say whether or not this is so. Does anyone want to chime in with an opinion?

Top
#442571 - 01/13/04 06:39 PM Re: George, Watch this!!!....(96k)
dsykes Offline
Member

Registered: 09/02/01
Posts: 9
Loc: New York,NY,UNITED STATES
Quote:
Originally posted by Nika:

The interesting question to me is why so many people are so pulled to the desire to have there actually be benefits to high frequency sampling. Why do we want to believe that the ear is limitless...

I find this utterly fascinating. I'd be interested in hearing what others have to say on the subject.

Why is it that we glom on to these "romantic" notions? Why do we seek out this "romance?"
...
Nika
Speaking from my armchair, I don't think it is about any romantic desire for human hearing to be in any sense unlimited. Human limitations are too abundant and too obvious for anyone to expect that. So I don't think I harbor any romantic notions in this regard.

On the other hand, the pursuit of sceintific understanding is all about romance: the human urge/quest to know "more better". That the latest "authority" (rememeber Galieo) or current state of understanding is always subject to being picked apart, challenged, questioned is what makes science work. We all know that.

So from my vantage point, it is not about accepting that hearing is limited. I don't think it would phase me to somehow "know" that in the year 3000 everyone records at 44.1K without controversy---i.e. everyone agrees there's absolutely no audible benefit to sampling any faster.

In the mean time, the fun, the romance, is in the learning process, the constant peeling back of the onion (as layman spectator or participant). Someday we will surely reach a point where the optimal sampling rate is beyond controversy...so "we" will find some other fuzzy area to banter about. In the meantime, there is fun in the peeling, while it lasts.

Finally, the fact that sustained, focused cross-discplinary research on brain function is not more than, what, 10 years old, suggests to me that this topic will be banter fodder for some time to come...or not?

David

Top
#442572 - 01/13/04 07:23 PM Re: George, Watch this!!!....(96k)
Nika Offline
MP Hall of Fame Member

Registered: 03/22/01
Posts: 2938
Loc: Ft Wayne,IN,UNITED STATES
Quote:
Originally posted by blairl:
Can anyone give some examples of converters that sound the same at 48 and 96K? I have seen mentioned many times that a good converter design will sound the same at 48 and 96 K but haven't seen mention of any brands or models that actually do this. Paul Frindle has said that the Oxford converters are not discernable from the source. You can't beat that. I have seen it suggested that some db technologies or Lavry Engineering converters might sound the same at 48 and 96K but haven't seen anyone definitively say whether or not this is so. Does anyone want to chime in with an opinion?
Blair,

An excellent question, but a tough one to answer. A part of the scientific process is making sure that the margin of error does not exceed the boundaries of the results of the test. In other words, if you are testing for a .1dB difference between to sources, but the margin of error in the testing protocol yields a margin of error of 6dB, then whatever results are released are inconclusive.

What we are talking about, here, is detecting ANY difference at all between two sources. This is obviously impossible to test for if there is ANY margin of error in the testing protocol. Think about it for a moment - if box A puts out a signal that, due to its power supply or jitter or anything else is .1dB different than box B then the margin or error in the testing protocol exceeds the bounds of the differences that are being tested for.

Just the same, if the testing is done in a room wherein the noisefloor is 20dB SPL and the difference between the two boxes isn't manifested except for at a level of 3dB SPL then in that one situation the two boxes may sound the same, but in some situations they may not.

Therefore, if we want to test for the audible differences between two solutions that are to yield absolutely identical results then we have to couple an empirical test with an analysis of the validity of the test. In other words, at some point we have to start testing the result "sideways," by proving that there are no audible differences by determining what all it would take to yield a difference and then analyzing whether the two boxes yield that much of a difference. In other words, you test the ear and figure out what it takes to register a difference in the ear. Then you figure out whether or not the converters yield differences that exceed the registerable difference recognition skills in the ear. Then you determine, indirectly, whether or not the converter yields a difference.

Then you can start to test and see if this holds true in listening situations by using an ABX test and check for the statistical probability of the subject to be able to guess A or B.

To answer your question, the only converter that I know of to have endured this extensive process, and therefore the only one I know is safe for conversation on this topic is the Oxford converter, which is not publicly available in a 96k form (out of a determination that it was unnecessary because it yielded no benefit). I have tested various other converters and heard of testing of other various converters where there was no audible difference, but not in a capacity as to establish that there was indeed NO difference - only that the difference would have been outside of certain boundaries of the listening test. I have been told, for example, that the Rosetta800 from Apogee fits the criteria, but have not tested it myself and don't know how rigorously the testing was done.

In every situation I know of in which the Oxford converters yielded differences it actually turned out that the difference exposed a problem with the test and not a difference with the converters. Paul Frindle wrote in a paper that he had a listener in their ABX booth to test out a particular process, for example, and the listener identified the difference. It was then determined that the testers had accidentally loaded the wrong coefficients in. If the converter is truly transparent and yields no difference between 44.1 (48) and 96 and a difference is heard then it would identify a problem with the testing protocol.

So, back to the question, I have not done significant enough testing on converters to hang my hat on a particular one and say "this one meets the criteria." I can tell you, however, that the specification on the converter that makes the difference is the audibility of the filter, and Bob Katz had a filter design that was completely transparent to 20 of 20 mastering engineers that he had test it. Mastering engineers have good testing facilities, as a general rule, as well as the knowledge of how to conduct a test and the desire to trump the test.

Sorry for the lengthy reply.

Nika.
_________________________
For more...

Top
#442573 - 01/14/04 12:02 AM Re: George, Watch this!!!....(96k)
Bill Mueller Offline
Senior Member

Registered: 06/22/03
Posts: 147
Loc: Maryland
Nika,

" From my understanding, though, what the brain "hears" in this situation is not information above 20KHz, but rather the lower harmonics that help with speech intelligibility and comprehension. We can think of this as using the "missing fundamental" principle, or possibly intermod distortion that actually resonates the cochlea at below 20KHz."

I agree exactly, but can’t this very process be the mechanism by which the brain “synthesizes” a 24khz square wave? I am talking perception, not raw mechanical/electro chemical conversion in the ear. In the old days we had a Scully tape machine with a bias frequency of 150 khz. The bias occilator beat with certain synth tracks and generated whole new pallets of unwanted sound.

The issue of what we "hear" and what we "experience" is important because of a brain function called habituation. With habituation, our brain normalizes or suppresses stimulus that it subconsciously determines to be unimportant. I happens every moment of the day and night with all of our senses. Only when the brain determines a stimulus to be important does it bring that information to the conscious mind. The brain can also give extra perceptive “weight” to certain stimulus. The “cocktail party effect” is a good example.

There are people who have hearing that surpasses the average person in both sensitivity and frequency response. Some of these people can also mentally record an acoustic event to the finest degree and remember it perfectly forever. These people are of course autistic savants. I would suggest that these children hold secrets to our potential auditory capacity because their brains cannot normalize stimulus and therefore must experience every moment of every event at maximum attention. We work very hard to give them control to allow dampening of the auditory experience. Without that control they retreat into a world of self stimulation to block out constant bombardment of uncontrollable stimulus. They warrant further study by folks like yourselves.

I don’t offer these issues as a red herring, only to say that studying normal human hearing must by definition include a great amount of neurological manipulation/conditioning/influence and may not lead to understanding the whole auditory system.

Thank you all for your contributions. I am learning a lot for an old guy.

Best regards,

Bill

Top
#442574 - 01/14/04 06:32 AM Re: George, Watch this!!!....(96k)
Bobro Offline
Platinum Member

Registered: 01/06/01
Posts: 1467
Quote:
Originally posted by Nika:


Why do we want to believe that the ear is limitless, that our capabilities about beyond what logic, reason, science, and research indicate? Why do we yearn to know that we haven't found the limits? What is it in the human psyche that keeps trying to find holes in the research that defines our capabilities?

To me, this desire is a part of a Judeo-Christian-western mentality that actually taps into a religious issue. To me it is a God issue - an evolutionary characteristic of humans to want to feel that we can't "know" things, for "knowing" things starts to make our lives more clearly less significant than we like to think. My thoughts on this subject are long and complex, and based on years of thought and observation but little "research" per se.

The compelling issue to me on this whole subject is not the issue of benefits or lack of benefits in high sample frequencies, but rather the desire by the majority of people to FIND benefits in higher sample frequencies. I find this utterly fascinating. I'd be interested in hearing what others have to say on the subject.

Why is it that we glom on to these "romantic" notions? Why do we seek out this "romance?"

Nika
An interesting view, though ethnocentric- since when has the "West" had a monopoly on mysticism?

And, from Man being fashioned in the image of God, to Adam's naming of all the animals, to guys getting locked up for suggesting that the Earth is not the center of the universe, it seems to me that "know-it-all arrogance" would be a more defensible complaint about "Judeo-Christian-Western" thinking than what you've proposed.

Anyway, as far as the topic at hand, it seems to me that if you must look for "Judeo-Christian-Western" thinking in these discussions, the fact that we're talking about tools makes it obvious that, in the offline world of bills, clients, workmanship, and all that, the taproot here is:

"If a tool is blunt and a workman does not sharpen its edge,
he must exert a great deal of effort;
so wisdom has the advantage of giving success.

-Ecclesiastes 10:10

Surely there are some discussing sampling rates with an ax to grind, so to speak, \:\) but IMO most are motivated by simple, practical concerns about having sharp tools.

Quote:
Originally posted by Nika:

The interesting question to me is why so many people are so pulled to the desire to have there actually be benefits to high frequency sampling.
Nika
Trippy- you'd certainly have something to say about "so many" people, having for example read thousands of posts here and at your forum, but it seems to me that there must be many who are, as I am, hoping that there are not benefits to recording at higher than say 48k and are downright pissed off, on a practical monetary level, when a specific experience seems to indicate a higher sampling rate as a noticably musically beneficial thing. Those are some big-ass files- but when you hear something sweet...

Quote:
Originally posted by Tom Cram:

Or it could be something a bit more down to earth.

"My mix sounds like crap, it must be my gear."

Uh-huh. \:\)
No, "That sounds sweet, let's work on getting more of it" is the thought behind gear discussions, at least for everybody I know.

-Bobro

Top
#442575 - 01/14/04 11:08 AM Re: George, Watch this!!!....(96k)
Nika Offline
MP Hall of Fame Member

Registered: 03/22/01
Posts: 2938
Loc: Ft Wayne,IN,UNITED STATES
[QUOTE]Originally posted by Bill Mueller:
Nika,

" From my understanding, though, what the brain "hears" in this situation is not information above 20KHz, but rather the lower harmonics that help with speech intelligibility and comprehension. We can think of this as using the "missing fundamental" principle, or possibly intermod distortion that actually resonates the cochlea at below 20KHz."

I agree exactly, but can’t this very process be the mechanism by which the brain “synthesizes” a 24khz square wave?


Can you elaborate on this?

Since the ear/brain connection is a digital system that has fixed boundaries, the brain interprets anything coming into it as though it fits within the boundary of that communication.

Let me give an example: You are working on a 48k session in a DAW and you add a limiter to your tracks. A limiter, as a form of compressor, adds all kinds of high frequency componentry to your waveform. This creates frequency content in your digital signal that is higher than the Nyquist frequency, because nothing was put in place to block that high frequency, "illegal" content from existing. So now you have a digital waveform with all kinds of "illegal" content in it. What happens with this waveform when you convert it at the D/A?

The D/A "interprets" this waveform in the only "legal" way that it can, taking all of that high frequency information and reproducing it as aliasing in the audible range. Since the digital system has fixed boundaries, information that enters it that is outside of those boundaries ends up being interpretted as "within" the boundaries as some form of distortion.

Since the connection between the brain and the ear works in a similar capacity a similar result would occur. Each nerve in the ear attached to an inner hair cell produces digital data that is based on a very specific frequency range. None of the inner hair cells exceed the ability to spit back information above a given frequency (around 1KHz). The combination of multiple of these nerves firing asynchronously gives our ear the sense of all audio that exceeds 1KHz. But the brain knows the response curves of each nerve, so as it gets these 3500 digital data streams it knows what each represents and can assemble a complete picture based on the cumulative signal.

Think of it this way - you have 10 A/D converters, each of which has strict filters on it that force that converter to only cover a 2KHz range. The combination of each of those converters can cover 20Hz to 20KHz, so long as you know which data stream is coming from which converter and which range that converter is representing. While simplified, the ear works much like this.

But since the brain knows the response of each nerve, what range it covers, and what its signal represents, the brain's ability to "synthesize" mental audible pictures during listening of things it can't hear is inherently limited. Just like your 48KS/s system can't reproduce anything above Nyquist, the brain can't "reproduce (an audible picture)" of anything that exceeds the boundaries of the communication format from the ear.

I am talking perception, not raw mechanical/electro chemical conversion in the ear. In the old days we had a Scully tape machine with a bias frequency of 150 khz. The bias occilator beat with certain synth tracks and generated whole new pallets of unwanted sound.

This is called "intermodulation distortion" and is very common, and one of the reasons that high sample frequency recording and playback can be a problem. When you talk about putting all kinds of HF content into a speaker you can create IM distortion that winds up in the audible range. Big problem! Until speakers are capable of reproducing linear results in the range we push into them we make ourselves susceptible to all kinds of audible distortion.

And for those that question this concept on the obvious grounds that analog tape did the same thing, this is correct, but digital audio has FAR more HF content than analog tape did due to the noiseshaping used in A/D and D/A converters.

To my knowledge, though, this concept has nothing to do with what occurs in the brain.

The issue of what we "hear" and what we "experience" is important because of a brain function called habituation. With habituation, our brain normalizes or suppresses stimulus that it subconsciously determines to be unimportant. I happens every moment of the day and night with all of our senses. Only when the brain determines a stimulus to be important does it bring that information to the conscious mind. The brain can also give extra perceptive “weight” to certain stimulus. The “cocktail party effect” is a good example.

Yup! The brain is pretty durned nifty!

I don’t offer these issues as a red herring, only to say that studying normal human hearing must by definition include a great amount of neurological manipulation/conditioning/influence and may not lead to understanding the whole auditory system.

I would advise looking into how, exactly, the studies on the ear have been done. It's pretty fascinating. I have a book by Yost you would probably enjoy. There are some books that are much thicker as well, some of which focus less on the physiology of hearing and more on the psychology(?) of hearing. You can also read the works of Fletcher, etc. I interviewed, when writing my book, Jont Allen of AT&T research who is lofty in the field of audiology. I also interviewed extensively Jim Johnston, formerly of AT&T research and currently of Microsoft.

Nika.
_________________________
For more...

Top
#442576 - 01/14/04 11:22 AM Re: George, Watch this!!!....(96k)
Nika Offline
MP Hall of Fame Member

Registered: 03/22/01
Posts: 2938
Loc: Ft Wayne,IN,UNITED STATES
[QUOTE]Originally posted by Bobro:
An interesting view, though ethnocentric- since when has the "West" had a monopoly on mysticism?

This is prone to start a debate, but western mentality, throughout time, has had more of a "nothing can stop us" approach than other cultures or religious groups. Manifest destiny, going to the moon, conquering new continents and civilizations have all been very European/western traditions. Couple this with the western approach to science as opposed to the eastern approach to science and the goal is always to dig in and find "something more," to continue to show that "there is more to uncover" etc. I see it as a western mentality thing. Others may not.

Surely there are some discussing sampling rates with an ax to grind, so to speak, \:\) but IMO most are motivated by simple, practical concerns about having sharp tools.

Hmm. Having been involved in the 1300 posts on this thread and equivalent numbers on other threads, I see this attitude as less pervasive in these discussions. There are certainly those that want to know, but there are an awful lot that simply want to disprove the validity of fixed boundaries.

Trippy- you'd certainly have something to say about "so many" people, having for example read thousands of posts here and at your forum, but it seems to me that there must be many who are, as I am, hoping that there are not benefits to recording at higher than say 48k and are downright pissed off, on a practical monetary level, when a specific experience seems to indicate a higher sampling rate as a noticably musically beneficial thing. Those are some big-ass files- but when you hear something sweet...

Ahh, if you're looking for an explanation of why your tracks sound different at 48k than 96k that is certainly a conversation that can yield all kinds of fruitful results. But it is a different thing to suppose that an inherent benefit exists in changing sample rates simply because a specific situation showed a difference.

Nika.
_________________________
For more...

Top
#442577 - 01/14/04 11:39 AM Re: George, Watch this!!!....(96k)
StoneinaPond Offline
MP Hall of Fame Member

Registered: 11/22/02
Posts: 2009
Loc: New York City
Nika,

With all due respect, and I mean that most sincerely, you keep focusing exclusively on the ear. I think we can accept what you're postulating as fact.

If I am blind, I can still feel sunlight on my skin. Conversely, imagine what it might be like to be in hot sunshine and have absolutely no sensation from the skin.

A bit like listing only with the ears?
_________________________
Yorik
Stone In A Pond


"Man will occasionally stumble over the truth, but most of the time he will pick himself up and continue on."

Top
#442578 - 01/14/04 12:22 PM Re: George, Watch this!!!....(96k)
Nika Offline
MP Hall of Fame Member

Registered: 03/22/01
Posts: 2938
Loc: Ft Wayne,IN,UNITED STATES
Yorik,

First, I'm not sure we're all on the same page with respect to the ear. Bone conduction, for example, simply stimulates the ear, and we "hear" through bone conduction the same way we hear through air conduction, yet it was postulated that the "hearing" of bone conducted material might exceed the capabilities of the ear/brain system.

Second, though, I am more than open to the notion that we can "perceive" sonic frequencies in other ways, but have found no evidence other than pure, mystical speculation to support this, but have seen evidence to the contrary.

We have ascertained, for example, that light waves that are lower in frequency than visible light can be sensed by the body (as heat). We have also ascertained that sound waves that are lower in frequency than audible frequencies can be sensed by the body (as vibrations). There is nothing to indicate, however, that we can sense higher frequency sound waves at the amplitude with which they exist in audio playback. Meanwhile, I have found evidence to indicate that this does not happen.

So are people approaching from this direction simply trying to find any substantiation to record at higher amplitudes, or do they have some sort of evidence that this type of perception exists?

Nika.
_________________________
For more...

Top
#442579 - 01/14/04 12:45 PM Re: George, Watch this!!!....(96k)
StoneinaPond Offline
MP Hall of Fame Member

Registered: 11/22/02
Posts: 2009
Loc: New York City
Nika,

Excuse me for quoting myself (slightly ammended.)

Quote:
So we go back to the question of what part of a good converter, that just happens to sample at both 48 and 96, provides the increased joy or whatever it is that we hear?

I know some say they have listened to the same converter at 48 and 96 and hear no difference. Is that a true scientific test?

Because if you do hear a difference, how can you isolate which part played the bigger role in that equation?
How do you seperate the clocking from the filtering? Perhaps the question is illogical or based on faulty assumptions. But if you understand it and it has merit, can you respond?
_________________________
Yorik
Stone In A Pond


"Man will occasionally stumble over the truth, but most of the time he will pick himself up and continue on."

Top
#442580 - 01/14/04 06:16 PM Re: George, Watch this!!!....(96k)
LawrenceF Offline
Senior Member

Registered: 10/11/01
Posts: 481
Loc: ,,UNITED STATES
Wow. What a thread.

Let me give my own opinions. No science involved just my common sense opinions.

1. If extremely high sample rates were required to duplicate sound that could be heard by humans, the guys at Bell Labs would have known that.

2. Business (in this case music) cannot survive without "improving". If they didn't convince people that 96k "sounds" better, how would they sell you the new 96k (read overpriced) gear.

3. I personally don't give a shit.

4. End users (translation: CD buyers) don't give a shit.

5. I've read some of Nika's posts over and over to try and understand what he's talking about and I've come to two conclusions...

A.) He's very smart and his reasoning makes sense.

B.) He has a huge ego that demands he let us know how smart he is. Although this does not cancel out A.

6. All this crap about "sensory perception" of sounds that the ear can't detect, going through the bones etc. may indeed be true. BUT, back in the 70's when the Temptations were playing on my record player it did not come close to the fidelity or reality of 44.1/16 digital but I loved it. It made me feel something. I'm quite sure there were sounds that escaped being captured by the recording medium of the day. You know what? I DON'T CARE. The music is what mattered then. The music is what matters now.

Beleive it or not...

48/24 is more than capable of accurately reproducing ANY sound that humans are capable of hearing. I don't know why so many people argue that fact in the face of ANY proven and repeatable (REPEAT-ABLE) experiment to the contrary.

8. 96k converters may "sound" better due to filtering and other things but I've seen nothing that convinces me that it is the higher sample rate that's responsible for that.

9. If I worked for Digidesign or another company that's heavily vested in 96k technology, I'd find a way to support the opposite argument. \:\)

Lawrence

Top
#442581 - 01/14/04 07:04 PM Re: George, Watch this!!!....(96k)
gm Offline
MP Hall of Fame Member

Registered: 02/10/00
Posts: 2184
Loc: Williamson County, TN, USA
Lawrence,

Brilliant post, man.

Thanks,
George
_________________________
George Massenburg

http://www.massenburg.com

Top
#442582 - 01/14/04 07:25 PM Re: George, Watch this!!!....(96k)
kk@jamsync.com Offline
Gold Member

Registered: 10/16/00
Posts: 649
Loc: ,,UNITED STATES
_________________________
KK Proffitt
Chief Audio Engineer, JamSync, Nashville
JamSync

Top
#442583 - 01/14/04 08:04 PM Re: George, Watch this!!!....(96k)
StoneinaPond Offline
MP Hall of Fame Member

Registered: 11/22/02
Posts: 2009
Loc: New York City
Quote:
All this crap about "sensory perception" of sounds that the ear can't detect, going through the bones etc. may indeed be true. BUT, back in the 70's when the Temptations were playing on my record player it did not come close to the fidelity or reality of 44.1/16 digital but I loved it. It made me feel something. I'm quite sure there were sounds that escaped being captured by the recording medium of the day. You know what? I DON'T CARE. The music is what mattered then. The music is what matters now.
Aw, come on, both you and George know that people in the 70s were pushing the envelope to try and improve the quality of audio reproduction.

Of course the music is paramount, for goodness sake.

But some of us are still uncomfortable with what digital audio has to offer at the level at which we participate. Do we just stop now?
_________________________
Yorik
Stone In A Pond


"Man will occasionally stumble over the truth, but most of the time he will pick himself up and continue on."

Top
#442584 - 01/14/04 11:06 PM Re: George, Watch this!!!....(96k)
LawrenceF Offline
Senior Member

Registered: 10/11/01
Posts: 481
Loc: ,,UNITED STATES
Quote:
Originally posted by kk@jamsync.com:
http://world.std.com/~griesngr/intermod.ppt
Apparently you skipped over number 3 in my post. \:\)

Lawrence

Top
#442585 - 01/14/04 11:23 PM Re: George, Watch this!!!....(96k)
Griffinator Offline
TPS cook & bottle washer
20k Club

Registered: 03/28/02
Posts: 20318
Loc: Lynchburg, VA, USA
Yikes. I forgot about this one - I remember being in Alphajerk's position (the one he held at the beginning of this behemoth thread) about 2 year ago, only I was significantly more ignorant of Nyquist's laws and the science of D/A conversion. I too made the comparisons to video D/A, and was sorely trounced by the logic handed to me by Nika et al.

I still hold to the idea that 96K is pushed as a way for the manufacturers to produce extremely poor quality converters that still sound good - because their awful anti-aliasing filters' noise and IM distortion exist well outside our hearing range.

Top
#442586 - 01/14/04 11:35 PM Re: George, Watch this!!!....(96k)
Bill Mueller Offline
Senior Member

Registered: 06/22/03
Posts: 147
Loc: Maryland
"Nika,

I agree exactly, but can’t this very process be the mechanism by which the brain “synthesizes” a 24khz square wave?

Can you elaborate on this?"

See Below:

"Since the ear/brain connection is a digital system that has fixed boundaries, the brain interprets anything coming into it as though it fits within the boundary of that communication."

It is a natural conceit that because we can build a mechanical system that approximates a function of the body (levers, pumps, bellows, this computer) that we believe the body then works like our invention. Sometimes this can be true but in the case of the brain and sensory perception, we are still far away. When perceiving information, the brain is constantly comparing the incoming info against a database of past events and synthesizing an experience from it. Therefor the experience of listening to a recording of a concert grand piano is better and more realistic (or far worse) for those of us who have actually stood next to a concert grand piano than someone who has only ever heard recordings. This is another form of habituation.

The most obvious form of this auditory preconception is language. We listen to a string of discordant phonems and synthesize a thought from it! This could not happen if the ear/brain worked in the kind of vaccum that is described when you only address the electrochemical/mechanical conversion of disturbances in an elastic medium.

"This is called "intermodulation distortion" and is very common, and one of the reasons that high sample frequency recording and playback can be a problem. "

Yes of course.

I would advise looking into how, exactly, the studies on the ear have been done. It's pretty fascinating. I have a book by Yost you would probably enjoy. There are some books that are much thicker as well, some of which focus less on the physiology of hearing and more on the psychology(?) of hearing. You can also read the works of Fletcher, etc. I interviewed, when writing my book, Jont Allen of AT&T research who is lofty in the field of audiology. I also interviewed extensively Jim Johnston, formerly of AT&T research and currently of Microsoft.

Thanks, I will try to fit one inbetween reading my DM2000 manual.

Best regards,

Bill

Top
#442587 - 01/14/04 11:42 PM Re: George, Watch this!!!....(96k)
kk@jamsync.com Offline
Gold Member

Registered: 10/16/00
Posts: 649
Loc: ,,UNITED STATES
Quote:
Originally posted by LawrenceF:
Quote:
Originally posted by kk@jamsync.com:
http://world.std.com/~griesngr/intermod.ppt
Apparently you skipped over number 3 in my post. \:\)

Lawrence
Ahhh...now I see. OK, everyone EXCEPT Lawrence should read it.

At first glance, I thought you had posted the link...
_________________________
KK Proffitt
Chief Audio Engineer, JamSync, Nashville
JamSync

Top
#442588 - 01/15/04 12:22 AM Re: George, Watch this!!!....(96k)
LawrenceF Offline
Senior Member

Registered: 10/11/01
Posts: 481
Loc: ,,UNITED STATES
Quote:
Originally posted by StoneinaPond:
But some of us are still uncomfortable with what digital audio has to offer at the level at which we participate. Do we just stop now?
Of course not. That would be very "un-human" like. I just have to agree with Nika when he say's he's baffled by something being bought en-masse by so many that is apparently imperceptible to most. That's all. Keep the research going forward.

I've still not seen a repeatable blind listening test where 96k consistantly comes out on top. Give me the link, I'll be glad to read it. MFG's show these colorful displays at trade shows that (according to them) explain why it's better. Don't you ever wonder why they never present the "repeated blind test by audio pro's who picked 96k 76% of the time"? Surely Digi or some company has commissioned such a test or two. Where is it? Dunno. Surely if it existed and 96k came out on top it would be posted everywhere for everyone to read.

It just seems to me that this (96k infatuation) is more faith than fact. I CAN see a difference between HDTV and "regular" TV. I can't afford the HDTV's at the current prices but I understand why people buy them, there is an obvious quality difference.

I've listened and listened and 96k looks more and more like "The Emperor's Clothes" to me. I listen to a high quality jazz CD recorded at 48k and think "Would that have been better with 96k recording?" I don't really see how when every nuance of every musical part comes through and makes the hair on my neck stand up. There is NOTHING missing. It sounds like I'm right there at the show. Perhaps 96k would make that 43rd hair stand up where now it's only 42? Yeah, that's worth $2000.

I COULD hear a difference between...

LP's (of the day) vs. 8 Tracks (no needle noise among other things)

8-tracks and cassettes (better sound, no interruption of the music)

Regular cassette vs. Cassette w/dolby or metal bias(clearer high end / less noise etc.)

Cassette vs. CD (obvious)

Notice how there was really no debate about those things because everyone COULD HEAR IT.

I cannot hear a difference between 96k and 48k in a blind test. If I can't perceive it or hear it the real question becomes does it really exist? Or maybe my ears are not up to it.

Much like religion. Either you "feel" it and believe in it or you don't, there is no way to conclusively prove it in the conventional sense. In the case of 96k mark me as an atheist.

So to answer your question....

The human quest for something better in all things mechanical, scientific and otherwise will never end and should never end. Much like our space program, at times we hit a point that changes everything. Often there are many, many years between those times when something tangible is achieved.

I think this is one of those times for digital audio as it relates to sampling. It'll be many years before anyone discovers anything that is IMMEDIATELY perceptible as "yeah, that's much better" than 48/24 to the vast majority of listeners, pro and consumer alike. If ever.

I'm not saying it won't ever happen. I (try to) never say never. Perhaps an entirely new method of sampling audio that hasn't even been discovered yet? Humans are really infants in the universe after all. Dunno.

I'm just saying that 96k aint' it.

Lawrence

Top
Page 40 of 42 < 1 2 ... 38 39 40 41 42 >