What is the advantage on high-sampling rate ?

Randy Yates wrote:

> Right. And we should start extending LCDs, CRTs, and other
> display devices to ultraviolet - you never know what the
> eye might do with those high frequencies. 

It would be OK to have proper spectrum in the
visible range, since this does not happen at all
with current displays (poor perfomance with
"flesh" color, bloody red missing, etc.)

Anyway, you know this is a different issue than
sound quality. For example visual systems are quite
far from human perception limits (in negative
sense), while sound equipment can achieve performance
of real instruments, and these are were they are,
in terms of human feelings.

So, you didn't answer the question. ;-)

bye,

-- 
   Piergiorgio Sartor

Arny Krueger wrote:

> The spectral content of *any* musical instrument is irrelevant to the basic
> question, which is audibility.

First, I think this is quite a superficial view,
you do not know possible non-linear (de)modulation
effects, which can (or cannot, I'm just using fantasy),
bring high (non perceiveble) frequency into the
audible domain.

Second, I'm not talking strickly about perception,
but about "Hi-Fi".
How we measure Hi-Fi?
Do we use hears?
Or do we use instruments?

> You can transmit anything you want to, but if the receiver doesn't respond
> to it, then your transmissions are futile.

But maybe, something else in the environment
will respond, and provide something to the
human listener...

> It does, but you also need to quantify that energy unless you don't believe
> in the concept of amplitude thresholds for hearing.

It's not only "hearing" the issue.

> In fact you can know what kind of non-linerities can happen, but again
> system nonlinearities are irrelevant to the basic question of the perceptual
> limitations of humans.

Nooo, because a non-linearity can bring the 100KHz
harmonic to 10KHz, so the sound from the violin and
the sound from the equipment will be different, and
we cannot talk about Hi-Fi.

> If you don't trust my files, analyze them to double-check my work, or even
> make some for yourself. This is not rocket science, it's just a walk in the
> park for people who even know just a little bit about audio production.

I trust your files, but I feel you're missing the point.

The problem is:

I do not know which effects can have the supersonic harmonics
of the violin when interacting with the enviroment, in relation
with the perceived sound.

Maybe there is nothing, I'm not stating anything, or maybe, in
certain conditions, something will popup, who knows?

So, the hi-fi is hi-fi because I perceive it very often as hi-fi
or because I measure that is hi-fi?

If you build a cross-over do you only listen to it or you try
to measure/model the performances?

If you build an amplifier how do you know it's working?
Do you just listen if it sounds good or you also measure?

Note that this is more a philosophical problem than an
engineering problem.

bye,

-- 
   Piergiorgio Sartor

Piergiorgio Sartor wrote:

> I mean, if the violin has some energy up to 100KHz, for
> example (I've no clue of reality, in this case), then it
> would make sense to claim "Hi-Fi" only a system that can
> reproduce that

No it wouldn't.  The actual acoustic waves in the air are
really irrelevant;  they're just an intermediate step in
a more complex process whose *actual* outcome is the
psychological effect in our brains.

True high fidelity can only be defined as a process (or
a system, or whatever) that reproduces *exactly* the
acoustic psychological effect in our brains.  From the
technological point of view, naturally, the only model
that we're capable to observe and understand, is the
acoustic waves in the air, and since that is highly
correlated to what happens in our brains, then we
choose to use that model.

Of course, your definition of Hi-Fi is a superset of
what I claim is the valid definition -- if a system is
able to reproduce the acoustic wave with molecule-by-
molecule identical properties  (this was a hyperbole),
then the acoustic psychological effect in our brains
must be the same.

Carlos
--

Arny Krueger wrote:

> In fact you can know what kind of non-linerities can happen, but again
> system nonlinearities are irrelevant to the basic question of the perceptual
> limitations of humans.

I think he was talking about non-linearities in the human ear.  You
don't have the same non-linearities in a microphone, and so when you
record (and eliminate) the 100kHz, it won't produce the same inter-
modulation at 10kHz that the human ear would.

It's a good point, in fact.

One argument I can throw against it is that the external ear canal
is a lossy resonator filter, and so as you keep going higher in
frequency, those sounds won't even make it to the timpanic membrane,
and thus they won't play any role in any possible non-linearity in
the ear.

Another argument against is that it has to be more than clear that
any possible imperfection in the sound reproduction due to the lack
of 100kHz (and the related non-linearities in the ear) has to be
*many* orders of magnitude below the imperfections that exists with
the best, state-of-the-art microphones and loudspeakers in the
audible band.

So, let's worry first about the super-obvious differences that 99.99%
of humans can notice, and then we'll worry about differences that
*maybe* 0.000001% of humans *could* notice under extremely controlled
listening conditions.

Carlos
--

My preference would be for 64K/24-bit for storage since we really can't do much
better than 20-21 bits now and I think I've heard that thermal noise and such
will limit the ability to ever reach true 24-bit quality in real-world
ADCs/DACs.  (On the other hand, true 16-bit quality was close to impossible a
few decades ago...)  Also, dynamic range of human hearing from threshold of
audibility to threshold of pain is somewhere close to 24-bits.  I wouldn't want
to be burdened with 33% more data across the board just in case I wanted to
reproduce both a flea and the big bang in the same performance!

But in systems where bits are plentiful and 32-bit words are more convenient,
then by all means.

"Randy Yates" <randy.yates@sonyericsson.com> wrote in message
news:xxpisfr7ch6.fsf@usrts005.corpusers.net...
> danlavry@mindspring.com (dan lavry) writes:
>
> > I agree with you, and 88.2KHz and 96KHz are certainly fast enough, an
> > overkill in fact.
> >
> > The argument that filtering near 20KHz while at 44.1KHz is true, but
> > going to 88-96KHz solve it. In fact, while douing that (going faster),
> > the ability to work nera zero (say 20Hz or other low frequencies)
> > takes gets tougher. So 96KHz is ok with me. 192KHz is not.
> >
> > BR
> > Dan Lavry
>
> I agree. That's why I proposed a 64/32 scheme awhile back. 64 kHz -
> enough room for those folks who have the hearing of a bat and to let
> the antialias filters breathe - and 32 bits - which covers the range
> from the energy of a flea fart in Kenya to the total energy of the
> universe.
>
> Of course it's a "pipe scheme" - not likely to be adopted. But it
> seems reasonable to give some thought to these matters.
> -- 
> Randy Yates
> Sony Ericsson Mobile Communications
> Research Triangle Park, NC, USA
> randy.yates@sonyericsson.com, 919-472-1124

Piergiorgio Sartor wrote:
> Arny Krueger wrote:
>
>> The spectral content of *any* musical instrument is irrelevant to
>> the basic question, which is audibility.
>
> First, I think this is quite a superficial view,
> you do not know possible non-linear (de)modulation
> effects, which can (or cannot, I'm just using fantasy),
> bring high (non perceiveble) frequency into the
> audible domain.

OK, so you apparently think that you can read my mind, and state right here
in front of the whole forum that I (me, myself personally)  can't possibly
know whether or not any particual high frequency signals can be brought into
the audible domain by means of some nonlinear distortion or another.

Of course, everybody with a brain is going to say that you are presuming a
heck of a lot that you can't possibly know. How can you know what I know and
don't know on this topic? How can you know all the education and experience
that I have that is relevant to this question? Quite obviously, you can't!

So, your claim can be safely dismissed on the grounds that is is speculative
and presumptious.

> Second, I'm not talking strickly about perception,  but about "Hi-Fi".

> How we measure Hi-Fi?
> Do we use hears?

In the end, the final basis for judgement of what constitutes high fidelity,
is what we can reliably hear.

> Or do we use instruments?

Test equipment is just a short cut. Listening tests are a lot of work and
time-consuming. If we can efficeintly reduce some aspect of high fidelity to
a simple test done with test equipment, we've saved time and money. Also,
listening test have some inherent limits that we may wish to circumvent.

>> You can transmit anything you want to, but if the receiver doesn't
>> respond to it, then your transmissions are futile.

> But maybe, something else in the environment
> will respond, and provide something to the
> human listener...

So you're saying that if a bat responds, but no human known ever has
responded to a given stimulus, and lots of humans have had every chance in
the world to respond, that this is supposed to be significant to the pursuit
of high fidelity?

>> It does, but you also need to quantify that energy unless you don't
>> believe in the concept of amplitude thresholds for hearing.

> It's not only "hearing" the issue.

What else is important to the pursuit but that which is in some sense
audible? BTW, let's not try to separate infrasonic stimulus from that which
is audible, because we have this convention of calling low-frequeny sound
"infrasonic", and there's plenty of evidence that the body responds to
infrasonic stimulus, even if the process isn't exactly hearing. Infrasonic
stiumulus is irrelevent to a discussion of high sampling rates since even
relatively low sampling rates can accurately convey infrasonic signals.

>> In fact you can know what kind of non-linerities can happen, but
>> again system nonlinearities are irrelevant to the basic question of
>> the perceptual limitations of humans.

> Nooo, because a non-linearity can bring the 100KHz
> harmonic to 10KHz, so the sound from the violin and
> the sound from the equipment will be different, and
> we cannot talk about Hi-Fi.

It could happen, but in the real world it's never been found to happen.

>> If you don't trust my files, analyze them to double-check my work,
>> or even make some for yourself. This is not rocket science, it's
>> just a walk in the park for people who even know just a little bit
>> about audio production.

> I trust your files, but I feel you're missing the point.

I don't think you really have a point.

> The problem is:

> I do not know which effects can have the supersonic harmonics
> of the violin when interacting with the enviroment, in relation
> with the perceived sound.

That would be your problem. Buy the mic, buy the 200 Khz sample rate audio
interface,  buy the speakers and do the listening test!

> Maybe there is nothing, I'm not stating anything, or maybe, in
> certain conditions, something will popup, who knows?

OK, so you're just wildly speculating, is that it?

> So, the hi-fi is hi-fi because I perceive it very often as hi-fi
> or because I measure that is hi-fi?

Ear versus gear is irrelevant to this discussion.

> If you build a cross-over do you only listen to it or you try
> to measure/model the performances?

In the end it's reliably percieved sound quality that is the final arbiter
of audio quality.

> If you build an amplifier how do you know it's working?

Listen to it, properly and carefully.

> Do you just listen if it sounds good or you also measure?

Both, when relevant.

> Note that this is more a philosophical problem than an
> engineering problem.

Not really. It's pretty darn simple. Just form a testible hypothesis and
test it!

mark_ovchain@yahoo.com (Mark Ovchain) wrote in message news:<d61ab25b.0404221910.11b30855@posting.google.com>...
> "Arthur" <arthurc99@yahoo.com> wrote in message news:<c65haq$1dgf$1@news.hgc.com.hk>...
> > Hello all,
> Well, I suspect that it helps the manufacturers sell more equipment.
> 
> Think of it as a form of "trickle-down", why don't you?

Think of it as "trickle-down"? Like - they are standing on the roof
with the zipper open, trickling down on their customers? :-)

Dan Lavry

Randy Yates wrote:

> 
> Oh, that. Yep, they convinced me. The other 4 billion people on the
> planet must be wrong. They must have discovered something that no
> other researcher has been able to find. Yessirr, that's a taker. 

Not sure what you mean here.  Do you consider it impossible 
that the brain registers some effects of supra-audible 
frequencies without telling perception that it is "sound" by 
itself?  What comes out of the sensor could be perceived in 
all kinds of ways.


Bob
-- 

"Things should be described as simply as possible, but no 
simpler."

                                              A. Einstein

Arny Krueger wrote:

> If so, then do YOUR OWN  DBTs with the provided software and data file, and
> a computer with a 24/96 audio interface and the speakers or headphones of
> your choice.  24/96 and even 24/192 audio interfaces are now plentiful and
> relatively inexpensive. The M-Audio "Revolution 7.1" is a good card to use
> as is their "Audiophile 24/96" or Echo's Mia. The SoundBlaster Audigy stuff
> tends to be so flawed (high frequency IM) as to be questionable for this
> application.
> 
> If you don't trust my files, analyze them to double-check my work, or even
> make some for yourself. This is not rocket science, it's just a walk in the
> park for people who even know just a little bit about audio production.
> 
> It's safe to predict that you'll sing at least a slightly different tune
> after you get your hands dirty with the real world!

Arny, in my case this would be a total waste of time so 
would you mind providing the results that you have acccumulated?


Bob
-- 

"Things should be described as simply as possible, but no 
simpler."

                                              A. Einstein

"Arthur" in news:c65haq$1dgf$1@news.hgc.com.hk...
> Recently, there is a tendency to use high sampling rate 96 KHz, 192 KHz in
> the audio application.  To my knowledge, about ~ 20 KHz, everything will
be
> cut off from the loudspeaker, amplifier, etc..  So, why 96 KHz, 192 KHz
> instead of 48 KHz..

I've just read 48 responses and many of them did not address the questions,
a few such as Carlos Moreno's and Jon Harris's did, some (maybe even
seriously!) presumed that the point of higher sampling rate was super-20kHz
audio bandwidth and argued details therein; and a few replies even resembled
consumer-audio product claims.

Here is some other perspective.  Bear with me, you might enjoy it.

Concepts like "filter," "quantizer," and "decimator" are tossed around with
great ease by people (even professionals) who do not have to build them or
fight their practical limitations.  If only the hardware could do this too!

Around 1989 I began incorporating, into seminars on oversampling A-D / D-A
interfaces, a series of three short sketches showing diagrams and language
on "analog-digital interfaces" as represented in current journals in the
three separate technical cultures of DSP, communications, and solid-state
circuits.  Not only was little overlap visible across these cultures, but
technical audiences would react to these sketches, across the boundaries
("so THAT's what's in a D/A converter!").

Then in 1990 when Bob Adams in AES invited me to give a large tutorial talk
on modern analog-digital interfaces at an AES conference (September 1990 in
Los Angeles), I commissioned a scientific photographer to prepare a sequence
of two slides.  First was a stylized version of the map of L. Frank Baum's
fictional world of Oz, with four quadrants, each a solid color, representing
Oz's four color-coded provinces: the lands of the Winkies, Quadlings,
Gillikans, and Munchkins.  The second slide was four contemporary
archival-journal covers addressing analog-digital interfaces (by
coincidence, the four cover colors paralleled the Oz map).  Of interest to
the AES audience was the insight that the blue AES Journal made them the
Munchkins.  Everyone loves the Munchkins: they figured in the 1939 movie,
_The Wizard of Oz._  I have been hearing about those slides ever since.  (I
should put them on a Web site.)  That cross-disciplinary tutorial led to my
survey paper and bibliography in the AES Journal (January/February 1991
issue), available from the AES at:

http://www.aes.org/journal/search.html

(Enter keyword phrase "Principles of Oversampling")

Much of the relationship between analog bandwidths and sampling rates, as
others pointed out, concerns practical analog band-limiting.  Oversampling
can simplify the (still inevitable) continuous-time filtering requirements
for this but it is a two-edged sword; digital filtering (sometimes
considerable) replaces the analog.  Many all-analog continuous-time
antialias or reconstruction filters are of order four to eight (they have
four two eight poles).  An example of a modern analog "brick-wall" lowpass
filter is the LTC1564 chip, implementing eight complex poles and four
imaginary-axis zeroes for 100 dB roll-off in a factor of 2.5 in frequency
(with concomitant noise and distortion performance, just as critical as
antialiasing).  That is a reasonably aggressive example of analog roll-off.
This analog block, however, can off-load some millions of
operations-per-second from a busy DSP engine, freeing the ops for more
cerebral tasks.

In a later tutorial review in 1992, I cited contemporary authors treating
oversampling A/D conversion (for example) as solely a filtering problem,
while others showed no awareness of any filtering aspect; applications and
implementation imposed constraints separate from the mathematical theory of
oversampling.  Also that real systems (unlike real authors) cannot be
selective in their awareness of these factors and still get a nice result.
That paper by the way is

Max W. Hauser.  "Oversampling data converters: A concise summary."
_Proceedings of the 1992 International Analog Applications Conference,_
Santa Clara, California, September 1992.  (Videotape through Media
Productions, New York.)

(I believe these points to be on topic.)    -- Max Hauser