Arny Krueger wrote:

Seems like a remarkable claim for which evidence should be
made available.

I'll see what I can dig up.

The provided recording's quietest passage is only about 55
dB below FS.

That's right.

With a quiet concert hall and a dynamic composition, 90+
dB of dynamic range is not rare.

In terms of commercial recordings such a thing is
exceedingly rare.

You're right, but musicians experience it often. Just sit in front of
the trumpets in a studio. Or play bass standing next to the drummer's
crash cymbal for a few nights.

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

Chel van Gennip wrote:

This excerpt doesn't have the quietest passage so you can't hear the
full dynamic range. I estimate that from quiet concert hall to triple
forte crescendo exceeded 100 dB of dynamic range.

The dynamic range of the excerpt is 42 dB, so there must be another 60 dB
in the rest.

There's also internal detail that isn't captured by looking at the
dynamic range of the envelope of the recorded waveform. How far down
into that envelope do we (can we) hear?

With a quiet concert hall and a dynamic composition, 90+ dB of dynamic
range is not rare.

I think it is. A quiet concert hall will have a sound level above 30-40dB
most microphones stop working at sound levels of 130-140dB.

I don't know what kind of microphones you're using, but recording in
local rock clubs I regularly see levels of 130 dB and above. Mics don't
stop working.

At close range it is opssible to get higher sound levels, but it isn't a
piano sound anymore.

That's arguable. The pianist hears a very different sound than the
audience up in the third tier balcony. Which one is the sound of the
piano?

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

Arny Krueger wrote:

Having *played the game* the sticking point is the "quiet
concert hall". It seems like putting living-breathing
musicians into the room kinda messes that wonderful noise
floor up just enough...

And in studios?

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

Len Moskowitz wrote:

There's also internal detail that isn't captured by looking at the
dynamic range of the envelope of the recorded waveform. How far down
into that envelope do we (can we) hear?

All the way down, EVEN when the bottom is below the noise floor. So this
is basically a linearity issue and not relevant to a noise discussion.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

John Blankenship wrote:

"MicroTrack 24/96 - Professional 2-Channel Mobile Digital Recorder"

I think we're fully justified in challenging that assertion.

....and a Soundblaster Pro... has "pro" right on the box!

The word "pro" is meaningless when it's written on the back of the box.

"Len Moskowitz" wrote in message
...
In fact *none* of them offer 24 bit noise specs. You mean they actually
exceed 16 bit, by a bit or two.
I've always found a *genuine* 16 bit performance quite adequate for live
recordings anyhow. Most people just like to *pretend* they are exceeding
that.

Our Mic2496 offers noise levels typically down around -140 dBFS in a 1 Hz
bandwidth. That's roughly 23 bits of dynamic range with very low noise.

Isn't that more than "a bit or two?"

No. It's a fallacy to compare a 1Hz BW with a full band spec.
It's probably about 18 bits I'll bet.

MrT.

"Len Moskowitz" wrote in message

Arny Krueger wrote:

For example, to make a recording with 140 dB dynamic
range it is required to record first the 140 dB sound
without distortion, and then 0 dB without excessive
corruption by noise.

Or from +20 to +160. Or from +40 to +180.

Agreed.

At close range, a fighter jet on an aircraft carrier deck
measures 150+ dB SPL.

Agreed, but if you make a recording just about *anyplace* on
an operational aircraft carrier, the background noise level
is way up there. They are noisy from 100's of feet away.

Just kinda picking numbers from space, but my recollection
of the last time I was in a modern passenger ship under way,
puts the background noise level maybe as high as 50 dB.

"Len Moskowitz" wrote in message

Arny Krueger wrote:

Seems like a remarkable claim for which evidence should
be made available.

I'll see what I can dig up.

The provided recording's quietest passage is only about
55 dB below FS.

That's right.

With a quiet concert hall and a dynamic composition, 90+
dB of dynamic range is not rare.

In terms of commercial recordings such a thing is
exceedingly rare.

You're right, but musicians experience it often. Just
sit in front of the trumpets in a studio. Or play bass
standing next to the drummer's crash cymbal for a few
nights.

Again, I get the loud, but what about the soft?

As soon as you put 20-100 musicians on a stage, it ain't a
quiet place.

"Len Moskowitz" wrote in message

Arny Krueger wrote:

Having *played the game* the sticking point is the "quiet
concert hall". It seems like putting living-breathing
musicians into the room kinda messes that wonderful noise
floor up just enough...

And in studios?

Same situation with the musicians.

"Len Moskowitz" wrote in message

Chel van Gennip wrote:

This excerpt doesn't have the quietest passage so you
can't hear the full dynamic range. I estimate that
from quiet concert hall to triple forte crescendo
exceeded 100 dB of dynamic range.

The dynamic range of the excerpt is 42 dB, so there must
be another 60 dB in the rest.

There's also internal detail that isn't captured by
looking at the dynamic range of the envelope of the
recorded waveform. How far down into that envelope do we
(can we) hear?

The usual number for the instantaneous dynamic range of the
human is about 70 dB.

With a quiet concert hall and a dynamic composition,
90+ dB of dynamic range is not rare.

I think it is. A quiet concert hall will have a sound
level above 30-40dB most microphones stop working at
sound levels of 130-140dB.

I don't know what kind of microphones you're using, but
recording in local rock clubs I regularly see levels of
130 dB and above. Mics don't stop working.

But what about the quiet? Dynamic range is not a single
number - its a ratio, right?

At close range it is opssible to get higher sound
levels, but it isn't a piano sound anymore.

That's arguable. The pianist hears a very different
sound than the audience up in the third tier balcony.
Which one is the sound of the piano?

That's not my problem - I have a PZM inside a grand piano.
It's a loud place, but it never gets totally quiet in there,
either. Piano sounding boards are 2-way *streets*. They
receive sound from the room pretty well.

Arny Krueger wrote:

Since we've got recorders that on paper can easily handle 90
dB or more dynamic range, there should be some extant real
world recordings that manifest this kind of dynamic range.

Never seen any.

Well I don't know about 90 dB, but at least "a lot"... I attended[*]
(with my mics and ear plugs) a big band rehearsal last sunday. I have to
wonder what sound levels the saxplayers experienced with the trombone
bells just decimeters from their ears... and some rooms can be pretty
quiet on occasion. At least out on the country where traffic is sparse
and there is no AC or forced ventilation.

Lars
[*] I might get back to the forum with a couple of samples and a request
for advise later bt that would be another thread another time...

--
lars farm // http://www.farm.se
lars is also a mail-account on the server farm.se
aim:

Mr.T MrT@home wrote:

No. It's a fallacy to compare a 1Hz BW with a full band spec.
It's probably about 18 bits I'll bet.

If when listening to the device you could hear a signal that's 22 bits
below 0 dBFS, would you say that's an 18-bit device or a 22?

The 20 Hz to 22 KHz single-number noise spec doesn't give the whole
picture. The way we determine noise specs is lacking.

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

It seems to me that you are interpreting noise specs ala sony with Atrac
recording / compression.
--
Martin Harrington
www.lendanear-sound.com

"Len Moskowitz" wrote in message
...

Mr.T MrT@home wrote:

No. It's a fallacy to compare a 1Hz BW with a full band spec.
It's probably about 18 bits I'll bet.

If when listening to the device you could hear a signal that's 22 bits
below 0 dBFS, would you say that's an 18-bit device or a 22?

The 20 Hz to 22 KHz single-number noise spec doesn't give the whole
picture. The way we determine noise specs is lacking.

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

Martin Harrington wrote:

It seems to me that you are interpreting noise specs ala sony with Atrac
recording / compression.

ATRAC uses lossy compression while Mic2496 uses linear PCM -- they're
nothing alike.

And while Sony's work on perceptual coding does somewhat take into
account how we hear, it has little to do with noise specifications.

You usually see noise specifications reduced to a single number, and
that number is misleading.

Here's an example: A device that has a -108 dB noise specification might
really have more than 130 dB of dynamic range. Based on the single
number spec a -130 dBFS signal should be lost in the noise -- after all
it's below -108, right? But it's not -- you can hear it.

What does that tell you?

To me it says that the way we specify noise performance is wrong and
misleading.

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

What you say may be all well and good, BUT, there is a standard for
specifications, one that we all understand and adhere to.
Changing the goalposts doesn't help anybody but the manufacturers.
Interpretation is just that, interpretation, and different people interpret
differently.
A standard is a standard...not open to interpretation.--
Martin Harrington
www.lendanear-sound.com

"Len Moskowitz" wrote in message
...

Martin Harrington wrote:

It seems to me that you are interpreting noise specs ala sony with Atrac
recording / compression.

ATRAC uses lossy compression while Mic2496 uses linear PCM -- they're
nothing alike.

And while Sony's work on perceptual coding does somewhat take into
account how we hear, it has little to do with noise specifications.

You usually see noise specifications reduced to a single number, and
that number is misleading.

Here's an example: A device that has a -108 dB noise specification might
really have more than 130 dB of dynamic range. Based on the single
number spec a -130 dBFS signal should be lost in the noise -- after all
it's below -108, right? But it's not -- you can hear it.

What does that tell you?

To me it says that the way we specify noise performance is wrong and
misleading.

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

"Len Moskowitz" wrote in message
...
Mr.T MrT@home wrote:

No. It's a fallacy to compare a 1Hz BW with a full band spec.
It's probably about 18 bits I'll bet.

If when listening to the device you could hear a signal that's 22 bits
below 0 dBFS, would you say that's an 18-bit device or a 22?

But you can hear a narrow band signal less than -96 dB with a CD player, so
what do YOU call that, 16 bits?
Nobody else does!


The 20 Hz to 22 KHz single-number noise spec doesn't give the whole
picture. The way we determine noise specs is lacking.

Not really. They are roughly comparable *IF* you are consistent
bandwidth and weighting curve, if used.
You are just confusing the requirements for comparison.

MrT.

"Len Moskowitz" wrote in message
...
You usually see noise specifications reduced to a single number, and
that number is misleading.

Here's an example: A device that has a -108 dB noise specification might
really have more than 130 dB of dynamic range. Based on the single
number spec a -130 dBFS signal should be lost in the noise -- after all
it's below -108, right? But it's not -- you can hear it.

What does that tell you?

It tells me you don't understand the difference between wide band and narrow
band noise measurements.
You probably have a problem with weighting curves as well then.

MrT.

"Len Moskowitz" wrote in message

Mr.T MrT@home wrote:

No. It's a fallacy to compare a 1Hz BW with a full band
spec. It's probably about 18 bits I'll bet.

If when listening to the device you could hear a signal
that's 22 bits below 0 dBFS, would you say that's an
18-bit device or a 22?

This would be a trick question. There is no single-number
relationship between bits and audibility.

The 20 Hz to 22 KHz single-number noise spec doesn't give
the whole picture.

To say the least.

The way we determine noise specs is lacking.

Not necessarily. The purpose of specs for most consumers is
to predict how the specified equipment will perform in
actual use.

When noise specs were used to characterize the difference
beween one machine with -55 dB noise and another with 65 dB
noise, then there was probably (almost certainly in critical
applications!) some difference in noise performance that
would be noticable in actual use.

Ironically, specs being what they were in the days of -55 dB
noise, the noticable difference was not absolutely certain,
and the difference observed in actual use might not even
follow the same ordering as the specs.

OTOH, if you compared two devices with equal headroom and 55
dB versus 65 dB A-weighted noise, the device with the better
A-weighted noise would probably be perceptibly less noisy in
many circumstances.

Of course if one device was specified in terms of 20-20 KHz
noise with flat weighting, and the other was specified in
terms of A-weighted noise, then predicting real world
performance could be tricky.

Fast forward to today, where we are comparing equipment
with -95 dB noise versus say, -105 dB noise. With reasonably
careful use, either device is so much quieter than its
operational environment that the 10 dB difference has
questionable value, even if all other things are equal.

Martin Harrington wrote:

What you say may be all well and good, BUT, there is a standard for
specifications, one that we all understand and adhere to.
Changing the goalposts doesn't help anybody but the manufacturers.

Bad standards should be changed. I'm surprised that you'd want a bad
standard to be continued.

In this case, the bad specification procedure helps the manufacturers
and hurts the consumer. Using that single number spec lets
manufacturers tailor their design to give a good test number, and it
doesn't tell us what the real noise performance is.

If manufacturer's would post noise level versus frequency graphs for a
few standard condition, we'd all be better off.

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

Mr.T MrT@home wrote:

Here's an example: A device that has a -108 dB noise specification might
really have more than 130 dB of dynamic range. Based on the single
number spec a -130 dBFS signal should be lost in the noise -- after all
it's below -108, right? But it's not -- you can hear it.

What does that tell you?

It tells me you don't understand the difference between wide band and narrow
band noise measurements.
You probably have a problem with weighting curves as well then.

If you take that position, it tells me that you don't understand how we
hear. And that you'd prefer to stick with a bad standard than correct
it.

Weighting curves can be good (as in the Fletcher-Munson loudness
curves), or they can be used as marketing tools to cover up bad
performance.

If manufacturers would publish their noise floor versus frequency
response curves at minimum gain and at a few other gain settings, with a
well-disclosed set of conditions (e.g., input terminations, input
voltage references) we all would be better off. Why let 'em hide behind
a one number specification that's misleading at best?

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

Mr.T MrT@home wrote:

Not really. They are roughly comparable *IF* you are consistent
bandwidth and weighting curve, if used.
You are just confusing the requirements for comparison.

Ever listen to the difference in noise characteristics between two mics
that have the same self-noise spec?

Try it sometime -- it might change your mind.

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

Arny Krueger wrote:

Fast forward to today, where we are comparing equipment
with -95 dB noise versus say, -105 dB noise. With reasonably
careful use, either device is so much quieter than its
operational environment that the 10 dB difference has
questionable value, even if all other things are equal.

Try comparing how self-noise sounds for a few microphones from different
manufacturers that have similar single-number noise specs. It's a
relatively easy test and will probably surprise you.

By your standards those self-noise numbers (typically 5 to 15 dB) should
reflect noise levels way down in the mud, but in real-life recording
situations they're not -- they're audible. And the noise
characteristics are very, very different.

The noise spec doesn't give any insight into those differences.

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

--

"Len Moskowitz" wrote in message
...

Martin Harrington wrote:

What you say may be all well and good, BUT, there is a standard for
specifications, one that we all understand and adhere to.
Changing the goalposts doesn't help anybody but the manufacturers.

Bad standards should be changed. I'm surprised that you'd want a bad
standard to be continued.


In your opinion, and why do you feel that you have the right to change those
standards?

Martin Harrington
www.lendanear-sound.com


In this case, the bad specification procedure helps the manufacturers
and hurts the consumer. Using that single number spec lets
manufacturers tailor their design to give a good test number, and it
doesn't tell us what the real noise performance is.

If manufacturer's would post noise level versus frequency graphs for a
few standard condition, we'd all be better off.

--
Len Moskowitz PDAudio, Binaural Mics, Cables, DPA, M-Audio
Core Sound http://www.core-sound.com
Teaneck, New Jersey USA Tel: 201-801-0812, FAX: 201-801-0912

"Len Moskowitz" wrote in message

Arny Krueger wrote:

Fast forward to today, where we are comparing equipment
with -95 dB noise versus say, -105 dB noise. With
reasonably careful use, either device is so much quieter
than its operational environment that the 10 dB
difference has questionable value, even if all other
things are equal.

Try comparing how self-noise sounds for a few microphones
from different manufacturers that have similar
single-number noise specs. It's a relatively easy test
and will probably surprise you.

Been there, done that. What's your point Len other than
taking advantage of the situation to characterize me as
being inexperienced?

By your standards those self-noise numbers (typically 5
to 15 dB) should reflect noise levels way down in the
mud, but in real-life recording situations they're not --
they're audible. And the noise characteristics are very,
very different.

I didn't say that at all, Len. What's your point Len other
than taking advantage of the situation to misrepresent what
I said?

The noise spec doesn't give any insight into those
differences.

I said that as well Len but you deleted it. What's your
point Len other than taking advantage of the situation to
twist what I said into something dumb?

I think Len Moskowit and Glenn Sanders should get together...they certainly
seem to have a lot in common.
--
Martin Harrington
www.lendanear-sound.com

"Arny Krueger" wrote in message
...
"Len Moskowitz" wrote in message

Arny Krueger wrote:

Fast forward to today, where we are comparing equipment
with -95 dB noise versus say, -105 dB noise. With
reasonably careful use, either device is so much quieter
than its operational environment that the 10 dB
difference has questionable value, even if all other
things are equal.

Try comparing how self-noise sounds for a few microphones
from different manufacturers that have similar
single-number noise specs. It's a relatively easy test
and will probably surprise you.

Been there, done that. What's your point Len other than taking advantage
of the situation to characterize me as being inexperienced?

By your standards those self-noise numbers (typically 5
to 15 dB) should reflect noise levels way down in the
mud, but in real-life recording situations they're not --
they're audible. And the noise characteristics are very,
very different.

I didn't say that at all, Len. What's your point Len other than taking
advantage of the situation to misrepresent what I said?

The noise spec doesn't give any insight into those differences.

I said that as well Len but you deleted it. What's your point Len other
than taking advantage of the situation to twist what I said into something
dumb?

"Len Moskowitz" wrote in message
...
Not really. They are roughly comparable *IF* you are consistent
bandwidth and weighting curve, if used.
You are just confusing the requirements for comparison.

Ever listen to the difference in noise characteristics between two mics
that have the same self-noise spec?

Try it sometime -- it might change your mind.

We all know advertised specs are not necessarily accurate, and very often
provide no mention of test conditions at all.
However that is a *different* argument entirely.

MrT.

On Mon, 17 Oct 2005 23:25:24 GMT, "Martin Harrington"
schreef:

I think Len Moskowit and Glenn Sanders should get together...they certainly
seem to have a lot in common.

Hahahahahha !


R






--
Http://www.xs4all.nl/~tuig/index.html

Jonny Durango wrote:

I see your point, and for most people I think the "photographer
method" would work. But how then, would you record a 5 hour
NON-STOP symphony

Same procedure as with any other uninteruptible recording: two
recorders.

Jonny Durango


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************

much simpler is to bring better recorder that not stop , not to mention
you can run dissent mikes straight in :-)
12 gb cf card of course

--
Oleg Kaizerman (gebe) Hollyland

"Peter Larsen" wrote in message
...
Jonny Durango wrote:

I see your point, and for most people I think the "photographer
method" would work. But how then, would you record a 5 hour
NON-STOP symphony

Same procedure as with any other uninteruptible recording: two
recorders.

Jonny Durango


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************

Lorin David Schultz wrote:

If you record with condenser mics, the Microtrack is probably
not a good choice.

External mic pre? - my DAT recently lost the last third of a very good
Brahms Requiem due to a flake from the name brand computer backup tape.
Fortunately it was not the only recording device, a FR2 was also
present. I can see the Microtrack being quite relevant for me as
secondary recording device. And the phantom voltage issue isn't one with
my AKG's anyway, at least not at the SPL I prefer to record.

- Lorin David Schultz


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************

Martin Harrington wrote:

You're loosing the plot.
this group, (R.A.M.P.S), is predominantly a professional location recording
forum...using mixing equipment every day.

..pro and .tech are also present in this thread.

Martin Harrington


Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************

"R¤ €b€rt°•" wrote:

On Tue, 04 Oct 2005 10:50:44 GMT, "Lorin David Schultz"
schreef:

I'm saying that the MT is not really all that useful to an audio pro as
it presently stands, but that it could be handy in certain circumstances
if it worked well without peripherals.

This is the same sort of gear that likes to flirt with the
''professional'' market as FCP, or a Edirol 4 track.

From the preliminary noises from an owner that I have heard it is better
than you think.

Great for Budget Bob & Discountboy, but from what i read in various
forums, more buggy then Windows ME.

Hmmm ... I had better read on in this thread .... thanks.

R

Kind regards

Peter Larsen

--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************

Scott Dorsey wrote:

In article ,
Martin Harrington wrote:
You see, that's the difference between RAMPS members, (location recordists),
and the other groups.
We are often in situations where we are likely to have the recorder and/or
mixer fall and be dangling by the cables, or worse.

Of course. That's why the Tuchel connectors on my Nagra lock into place
with a screw-down ring.

I will rather have the plug pulled out than the equipment damaged or the
"displaced person" falling in the mic cable because it did not get
pulled out get hurt by falling.

--scott


Kind regards

Peter Larsen


--
*******************************************
* My site is at: http://www.muyiovatki.dk *
*******************************************

On Thu, 20 Oct 2005 21:51:05 +0200, Peter Larsen
schreef:


From the preliminary noises from an owner that I have heard it is better
than you think.

Great for Budget Bob & Discountboy, but from what i read in various
forums, more buggy then Windows ME.

Hmmm ... I had better read on in this thread .... thanks.

Buy it if you like it, and tell us the results. My clients are a bit
bored by buggy hardware, in whatever form. ( at least that's what they
tell me )



R





--
Http://www.xs4all.nl/~tuig/index.html

Len, back in the early 80s I tested my Sony PCM-F1 in just this same
way. I could hear signals 19 bits below 0 dBFS, and that was only a
16-bit device. That's what dither's all about--it lets you hear signals
that are well below the noise floor.

So it's all very well that your device passes that test, since in a
general way it verifies your use of dither. But you certainly can't use
that kind of test to estimate the bit depth of a recording device.

--best regards