Re: What's this 2.82MHz sample rate DSD crap?

[ScotFraser]

<< What's missing from the DSD rating is the fact that it represents only 1
bit. Hell, the 16/44.1 S/PDIF spec involves 2.82 million bits per second
(two 16-bit samples in a 64-bit word, 64 x 44100 = 2,822,400), does that
make it a 2.82MHz 1-bit sample rate too? >>

You're comparing delta-sigma modulation to pulse code modulation. The numbers
aren't equivalent for similar resolution.


Scott Fraser

[Leoaw3]

Scott Fraser wrote:
>You're comparing delta-sigma modulation to pulse code modulation. The numbers
>aren't equivalent for similar resolution.

Good observation. So....... as a rough cut, what WOULD be the equivalent
sample rate?

-lee-

[Arny Krueger]

"Leoaw3" > wrote in message

> Scott Fraser wrote:
>> You're comparing delta-sigma modulation to pulse code modulation.
>> The numbers aren't equivalent for similar resolution.
>
> Good observation. So....... as a rough cut, what WOULD be the
> equivalent sample rate?

The required sample rate would be 200 KHz because DSD has 100 KHz bandpass.

Because of the heavy noise shaping, the equivalent bits per sample would
probably only be around 18 bits.

18 * 200 = 3.6 MHz.

[Remixer]

According to Philips, 8Fs is transparent for DSD. But then there are those
who hold that 1Fs, others 2Fs, and still others, 4Fs is all we'll ever need
audio, so take your pick.

[Jay - atldigi]

In article >, "Arny Krueger"
> wrote:

> "Leoaw3" > wrote in message
>
> > Scott Fraser wrote:
> >> You're comparing delta-sigma modulation to pulse code modulation.
> >> The numbers aren't equivalent for similar resolution.
> >
> > Good observation. So....... as a rough cut, what WOULD be the
> > equivalent sample rate?
>
> The required sample rate would be 200 KHz because DSD has 100 KHz
> bandpass.
>
> Because of the heavy noise shaping, the equivalent bits per sample would
> probably only be around 18 bits.
>

The effective dynamic range of DSD changes wildly with frequency. In the
range of human hearing, the useable range is claimed to average around
120 dB (20 bit equivalent performance), but as you go higher, it shrinks
drastically. With the severe noise shaping required, all that noise has
to go somewhere, and above the commonly accepted range of human hearing
is where they put it.

--
Jay Frigoletto
Mastersuite
Los Angeles
promastering.com

[Arny Krueger]

"Jay - atldigi" > wrote in message

> In article >, "Arny Krueger"
> > wrote:
>
>> "Leoaw3" > wrote in message
>>
>>> Scott Fraser wrote:
>>>> You're comparing delta-sigma modulation to pulse code modulation.
>>>> The numbers aren't equivalent for similar resolution.
>>>
>>> Good observation. So....... as a rough cut, what WOULD be the
>>> equivalent sample rate?
>>
>> The required sample rate would be 200 KHz because DSD has 100 KHz
>> bandpass.
>>
>> Because of the heavy noise shaping, the equivalent bits per sample
>> would probably only be around 18 bits.
>>
>
> The effective dynamic range of DSD changes wildly with frequency.

Agreed, and "heavy noise shaping" is techy-speak for "dynamic range changes
wildly with frequency". "heavy noise shaping" is the means, and "dynamic
range changes wildly with frequency" is the results. We're on exactly the
same page.

> In the range of human hearing, the useable range is claimed to average
> around 120 dB (20 bit equivalent performance), but as you go higher,
> it shrinks drastically.

Since listening to most music at 120 dB will cause a very significant shift
in the threshold of hearing (as many if not most of us have personally
experienced!) saying the human ear has 120 dB of dynamic range is IMO
questionable. It's like saying a car can go 120 mph when the car's engine
always blows when you actually try to do it. The car goes 1 foot at 120 mph,
blows its engine, and never goes again without big repairs. The ear listens
to 120 dB for not very long at all, and then it loses its ability to hear
sounds at 0 dB, 10 dB, even 20 or 30 dB. This 120 dB dynamic range idea is
not only practically impractical for critical listening, under OSHA federal
law, it's illegal to impose on employees in the workplace.

Similarly, no practical recording studio, and no practical listening room
can accurately portray a sound at 0 dB. Most rooms have a background noise
level of 30 dB or more, particularly by the time people get in the room to
make and listen to music.

So music as loud as 120 dB is a bad idea, and music as quiet as 0 dB is an
impractical idea, making 120 dB the dynamic range between a bad idea and an
impractical idea.I'm in favor of talking about dynamic range as something
that is in the realm of good and practical ideas.

Since music varies so much in terms of its time/amplitude profiles, it is
difficult to say exactly what the ear's dynamic range is, but it appears
that its someplace between 60 and 80 dB. I think that Bob Katz's comments
about monitoring mastering with peak levels in the 85 dB range are
reasonable and meaningful.

In DBTs I've found that the ability to hear small differences tends to go
away when the levels are too high, just as surely the ability to hear small
differences goes away when the levels are too low.

> With the severe noise shaping required, all
> that noise has to go somewhere, and above the commonly accepted range
> of human hearing is where they put it.

Again agreed. The reason why maybe the information theory equivalent of 14
bits can provide ca. 120 dB dynamic range up to 20 KHz is all the garbage
they dump in the 80 KHz above 20 KHz. I've been taken to task for calling
DSD a perceptual coding technique, but I'm sticking to my guns even years
later. DSD is not the same kind of perceptual coding technology as is used
with MP3s, but the basic game plan is very much the same - hide the loss of
information where its not so objectionable to the ear, and/or put it in a
place can be dealt with some other way.

In a way, DSD is a gigantic perceptual shell game. DVD-A at least lays all
its cards on the table, even though its still a gigantic waste of bandwidth.
IMO, it would be a very good thing if people instead put this kind of
effort into the capture and recreation of the spatial properties of music,
current events and drama.

[Arny Krueger]

"Jay - atldigi" > wrote in message


> Of course, most PCM A/D converters
> these days use oversampling and initially sample at rates many times
> higher than the output rate, whether it be 44.1 or 96. But you do
> have a point about the marketing value of higher numbers, and how
> that doesn't necessarily tell the whole story.

Not only that, but some techniques used to dither down for production of CDs
at 16 bits, also use heavy noise shaping to get very impressive amounts of
dynamic range where the ear is most sensitive. The following paper shows
the clear technical benefits of this kind of technology:

http://audio.rightmark.org/lukin/dither/dither.pdf

The samples from this web page can be used to listen to the benefits of this
technology for audio signals whose effective dynamic range has been
artifically and substantially increases for the purpose of demonstration:

http://audio.rightmark.org/lukin/dither/#samples

The punch line is of course, how this works with recordings that are made
normally and naturally.

[Roger W. Norman]

"Jay - atldigi" > wrote in message
...
> The effective dynamic range of DSD changes wildly with frequency. In the
> range of human hearing, the useable range is claimed to average around
> 120 dB (20 bit equivalent performance), but as you go higher, it shrinks
> drastically. With the severe noise shaping required, all that noise has
> to go somewhere, and above the commonly accepted range of human hearing
> is where they put it.
>
> --
> Jay Frigoletto
> Mastersuite
> Los Angeles
> promastering.com

Just as Sony did with SBM on the 18 bit converter DATs. Putting noise where
one can't hear it is a pretty good idea (well, as close to where one
couldn't hear it as DAT gets). Didn't UV22 used to spread the noise
throughout the audible spectrum, somewhat akin to very low level tape hiss?

--


Roger W. Norman
SirMusic Studio
Purchase your copy of the Fifth of RAP CD set at www.recaudiopro.net.
See how far $20 really goes.

[Roger W. Norman]

"Arny Krueger" > wrote in message
...
>
>
> Since listening to most music at 120 dB will cause a very significant
shift
> in the threshold of hearing (as many if not most of us have personally
> experienced!) saying the human ear has 120 dB of dynamic range is IMO
> questionable. It's like saying a car can go 120 mph when the car's engine
> always blows when you actually try to do it. The car goes 1 foot at 120
mph,
> blows its engine, and never goes again without big repairs. The ear
listens
> to 120 dB for not very long at all, and then it loses its ability to hear
> sounds at 0 dB, 10 dB, even 20 or 30 dB. This 120 dB dynamic range idea is
> not only practically impractical for critical listening, under OSHA
federal
> law, it's illegal to impose on employees in the workplace.


That's a rather questionable argument, Arny. Not all music is all 120 dB
all the time, although I admit I wouldn't want to listen, unprotected, to
ANY music at 120 dB. But Jay wasn't talking about SPL, he was talking about
dynamic range, and even if you turn the volume down, the dynamic range is
still 120 dB.

--


Roger W. Norman
SirMusic Studio
Purchase your copy of the Fifth of RAP CD set at www.recaudiopro.net.
See how far $20 really goes.




>
> So music as loud as 120 dB is a bad idea, and music as quiet as 0 dB is an
> impractical idea, making 120 dB the dynamic range between a bad idea and
an
> impractical idea.I'm in favor of talking about dynamic range as something
> that is in the realm of good and practical ideas.
>
> Since music varies so much in terms of its time/amplitude profiles, it is
> difficult to say exactly what the ear's dynamic range is, but it appears
> that its someplace between 60 and 80 dB. I think that Bob Katz's comments
> about monitoring mastering with peak levels in the 85 dB range are
> reasonable and meaningful.
>
> In DBTs I've found that the ability to hear small differences tends to go
> away when the levels are too high, just as surely the ability to hear
small
> differences goes away when the levels are too low.
>
> > With the severe noise shaping required, all
> > that noise has to go somewhere, and above the commonly accepted range
> > of human hearing is where they put it.
>
> Again agreed. The reason why maybe the information theory equivalent of 14
> bits can provide ca. 120 dB dynamic range up to 20 KHz is all the garbage
> they dump in the 80 KHz above 20 KHz. I've been taken to task for calling
> DSD a perceptual coding technique, but I'm sticking to my guns even years
> later. DSD is not the same kind of perceptual coding technology as is used
> with MP3s, but the basic game plan is very much the same - hide the loss
of
> information where its not so objectionable to the ear, and/or put it in a
> place can be dealt with some other way.
>
> In a way, DSD is a gigantic perceptual shell game. DVD-A at least lays
all
> its cards on the table, even though its still a gigantic waste of
bandwidth.
> IMO, it would be a very good thing if people instead put this kind of
> effort into the capture and recreation of the spatial properties of music,
> current events and drama.
>
>
>

[Arny Krueger]

"Roger W. Norman" > wrote in message

> "Arny Krueger" > wrote in message
> ...
>>
>>
>> Since listening to most music at 120 dB will cause a very significant
shift
>> in the threshold of hearing (as many if not most of us have
>> personally experienced!) saying the human ear has 120 dB of dynamic
>> range is IMO questionable. It's like saying a car can go 120 mph
>> when the car's engine always blows when you actually try to do it.
>> The car goes 1 foot at 120 mph, blows its engine, and never goes
>> again without big repairs. The ear listens
>> to 120 dB for not very long at all, and then it loses its ability to
>> hear sounds at 0 dB, 10 dB, even 20 or 30 dB. This 120 dB dynamic
>> range idea is not only practically impractical for critical
>> listening, under OSHA federal
>> law, it's illegal to impose on employees in the workplace.

> That's a rather questionable argument, Arny.

It's controversial, but it's also very pragmatic, I think.

> Not all music is all
> 120 dB all the time, although I admit I wouldn't want to listen,
> unprotected, to ANY music at 120 dB.

Seems like we're in agreement at this point.

> But Jay wasn't talking about SPL, he was talking about dynamic range, and
even if you turn the
> volume down, the dynamic range is still 120 dB.

Except as a practical matter, the dynamic range of music is never as much as
120 dB. Dynamic range isn't based on one measurement, it's based on two. One
measurement is how loud things get, and the other is how quiet things get.
Dynamic range is the difference between the two.

SPLs are the reasonable way to measure acoustic dynamic range. You measure
two SPLs and you take the difference. Voila, a number for dynamic range.

I think that anybody who has been at a loud rock concert will agree that 0
dB SPL is very foreign to what happens at a loud rock concert.

But this is very far afield from the issue of the actual dynamic range of
the human ear. My point is that if you try to show that the human ear has a
dynamic range of 120 dB in a practical sense, you're going to fail. To do
the experiment you've got to have people who can hear two sounds that are
120 dB apart, both of which are audible.

We know a priori that the smallest sound that can be heard under the most
ideal conditions is about -4 dB SPL. That pretty well sets what the loudest
sound is going to be. It's going to be at least 120 dB over -4 dB SPL or
about 116 dB SPL.

We also know from practical experience that listening to musical sounds that
peak at 116 dB SPL for any practical amount of time will naturally preclude
hearing a sound at -4 dB SPL for hours, or even days.

Therefore, the idea that recordings have to have 120 dB dynamic range to
match the capabilities of the human ear does not agree with practical,
empirical observations. Oh, we can go out and hear a -4 dB tone in a cave,
and we can then find live musical performances with 120 dB peaks. But, if
you spend any time listening to that musical performance at 120 dB, we're
not going to hear that -4 dB tone in the cave for hours or days.

Furthermore listening to -4 dB tones in a cave is not part of any practical
musical experience that much of anybody is interested in. So why should we
subject ourselves to the pretense that we need to capture a 120 dB dynamic
range to match the capabilities of the human ear?

Interestingly enough we have the capability to come pretty close to making
and distributing recordings with 120 dB dynamic range and there are people
with audio systems that can come pretty close to reproducing them
accurately. However, the maximum dynamic range that anybody has ever found
on a commercial recording is still only about 73 dB aside from artificial
fades.

[Arny Krueger]

"Roger W. Norman" > wrote in message

> "Jay - atldigi" > wrote in message
> ...
>> The effective dynamic range of DSD changes wildly with frequency. In
>> the range of human hearing, the useable range is claimed to average
>> around 120 dB (20 bit equivalent performance), but as you go higher,
>> it shrinks drastically. With the severe noise shaping required, all
>> that noise has to go somewhere, and above the commonly accepted
>> range of human hearing is where they put it.
>>
>> --
>> Jay Frigoletto
>> Mastersuite
>> Los Angeles
>> promastering.com
>
> Just as Sony did with SBM on the 18 bit converter DATs. Putting
> noise where one can't hear it is a pretty good idea (well, as close
> to where one couldn't hear it as DAT gets). Didn't UV22 used to
> spread the noise throughout the audible spectrum, somewhat akin to
> very low level tape hiss?

Please see:

http://audio.rightmark.org/lukin/dither/dither.htm

more specifically:

http://audio.rightmark.org/lukin/dither/images/uv22.png

[Justin Ulysses Morse]

Roger W. Norman > wrote:

> That's a rather questionable argument, Arny. Not all music is all 120 dB
> all the time, although I admit I wouldn't want to listen, unprotected, to
> ANY music at 120 dB. But Jay wasn't talking about SPL, he was talking about
> dynamic range, and even if you turn the volume down, the dynamic range is
> still 120 dB.

How do you figure? If you turn down your stereo so that the full scale
loud parts are now a comfortable 85dBSPL instead of 120, then how loud
are the quietest (-120dBFS) parts? -35dBSPL? You can't really go
below 0dBSPL. I mean you could, but it wouldn't be any more silent
than silence. The -85dBFS passages would be equally silent. You've
lost dynamic range. You now have a theoretical 85dB of dynamic range.

Except whoops we forgot that there's environmental noise in the
listening room. So there's really only 50dB of dynamic range.

ulysses

[Rob Adelman]

Justin Ulysses Morse wrote:


> How do you figure? If you turn down your stereo so that the full scale
> loud parts are now a comfortable 85dBSPL instead of 120, then how loud
> are the quietest (-120dBFS) parts? -35dBSPL? You can't really go
> below 0dBSPL. I mean you could, but it wouldn't be any more silent
> than silence. The -85dBFS passages would be equally silent. You've
> lost dynamic range. You now have a theoretical 85dB of dynamic range.

Something I have been wondering about for a while now. This is a bit
confusing. Isn't any dynamic range specification actually a potential
dynamic range? Depending on listening environment, speaker system,
amplifier capacity, and most importantly the position of the volume
control?

So a system and recording of minimal dynamic range turned up very loud
could still have a potential of huge range between a quiet passage (or
silence) and full on?

Still wondering.

-Rob

[Rob Adelman]

Rob Adelman wrote:


> So a system and recording of minimal dynamic range turned up very loud
> could still have a potential of huge range between a quiet passage (or
> silence) and full on?


Oh, and vise-versa, a system and recording with huge dynamic range
listened to at very low level would really have very little range?

[Arny Krueger]

"Justin Ulysses Morse" > wrote in message
m

>If you turn down your stereo so that the full
> scale loud parts are now a comfortable 85dBSPL instead of 120, then
> how loud are the quietest (-120dBFS) parts? -35dBSPL? You can't
> really go below 0dBSPL. I mean you could, but it wouldn't be any
> more silent than silence. The -85dBFS passages would be equally
> silent. You've lost dynamic range. You now have a theoretical 85dB
> of dynamic range.

> Except whoops we forgot that there's environmental noise in the
> listening room. So there's really only 50dB of dynamic range.

I'm sure that a lot of illusions got ruined when people discovered the
statistical analysis tool in Cool Edit. You highlight a passage, click an
icon with a magnifying glass and red squiggleys, and there it is, "Average
RMS power" for the highlighted area.

In a recording of mine I just popped up, I highlighted a passage that was
mic-on but nobody was talking, playing or singing for about a second.
"Average RMS power" was about -59 dB. Next I highlighted the biggest organ
crescendo, same mic. -17 dB. The dynamic range for that mic was a whopping
42 dB.

[Joakim Wendel]

In article >,
Justin Ulysses Morse > wrote:

> Roger W. Norman > wrote:
>
> > That's a rather questionable argument, Arny. Not all music is all 120 dB
> > all the time, although I admit I wouldn't want to listen, unprotected, to
> > ANY music at 120 dB. But Jay wasn't talking about SPL, he was talking about
> > dynamic range, and even if you turn the volume down, the dynamic range is
> > still 120 dB.
>
> How do you figure? If you turn down your stereo so that the full scale
> loud parts are now a comfortable 85dBSPL instead of 120, then how loud
> are the quietest (-120dBFS) parts? -35dBSPL? You can't really go
> below 0dBSPL. I mean you could, but it wouldn't be any more silent
> than silence. The -85dBFS passages would be equally silent. You've
> lost dynamic range. You now have a theoretical 85dB of dynamic range.
>
> Except whoops we forgot that there's environmental noise in the
> listening room. So there's really only 50dB of dynamic range.
>
> ulysses

I work a lot in a symphony orchestra and we get above 120 dB quite
frequently on stage and the quietest is not very quiet at all ... (part
of it ofcourse due to the fact that our ears can't really hear really
soft things after having the 120+ roar going on everyday etc.)

I thunk the 120 was the 'actual technically possible dynamic range' and
not the thing you'd use as a finished product ?
This is a dynamic range that is totally worthless in the consumers end.
(IMHO)
The more dynamic range you can work with i thought was better though ?

--
Joakim Wendel
Remove obvious mail JUNK block for mail reply.

My homepage : http://violinist.nu

[hollywood_steve]

>
> The effective dynamic range of DSD changes wildly with frequency. In the
> range of human hearing, the useable range is claimed to average around
> 120 dB (20 bit equivalent performance), but as you go higher, it shrinks
> drastically. With the severe noise shaping required, all that noise has
> to go somewhere, and above the commonly accepted range of human hearing
> is where they put it.

OK, we've all read the arguements about higher sample rates in the PCM
world (96kHz vs 192kHz, etc.), but what about 2.8mHz vs the proposed
5.6mHz DSD sample rate? Does the increase in sample rate from 2.8m to
5.6m really matter in the DSD world (like 96k vs 44k PCM) or is it
considered a load of hooey? (like 192k vs 96k PCM)

From a practical standpoint, does someone with a 2.8mHz DSD machine
need to be concerned about the proposed 5.6Mhz rate or does the
existing sample rate give you all the quality you are gonna get from
DSD technology?

thanks

Steve

[Arny Krueger]

"Rob Adelman" > wrote in message

> Justin Ulysses Morse wrote:
>
>
>> How do you figure? If you turn down your stereo so that the full
>> scale loud parts are now a comfortable 85dBSPL instead of 120, then
>> how loud are the quietest (-120dBFS) parts? -35dBSPL? You can't
>> really go below 0dBSPL. I mean you could, but it wouldn't be any
>> more silent than silence. The -85dBFS passages would be equally
>> silent. You've lost dynamic range. You now have a theoretical 85dB
>> of dynamic range.
>
> Something I have been wondering about for a while now. This is a bit
> confusing.


> Isn't any dynamic range specification actually a potential
> dynamic range? Depending on listening environment, speaker system,
> amplifier capacity, and most importantly the position of the volume
> control?

Presuming that the recording actually has an appreciable amount of dynamic
range...

Presuming that the dynamic range of the recording is not somehow
artificially expanded during playback using some kind of dynamics processor
or manual gain riding...

At low volumes a system is operating closer to the noise floor of the
playback system which includes the room. At higher volumes the maximum and
minimum output of the system increase, but the minimum output of the system
is buried in the noise floor of the room. At the highest volumes, the lowest
level passages of the recording may break out of the noise floor of the
room, or not.

The dynamic range of the recording sets the highest possible dynamic range
that anybody will hear while listening to it. The dynamic range perceived by
the listener will always be less than the dynamic range of the recording.

> So a system and recording of minimal dynamic range turned up very loud
> could still have a potential of huge range between a quiet passage (or
> silence) and full on?

The dynamic range of the recording limits the maximum dynamic range that
will ever be heard while playing it. There's no way to get more dynamic
range out of recording than what it has, aside from dynamics processing
during playback. If a recording has minimal dynamic range, turning it up
won't help. You just end up with playback that is loud all the time.

[Arny Krueger]

"Rob Adelman" > wrote in message

> Rob Adelman wrote:
>
>
>> So a system and recording of minimal dynamic range turned up very
>> loud could still have a potential of huge range between a quiet
>> passage (or silence) and full on?
>
>
> Oh, and vise-versa, a system and recording with huge dynamic range
> listened to at very low level would really have very little range?

This can happen, because the background noise of the playback environment
will swamp out and masks the low level passages on the recording.

[Arny Krueger]

"Joakim Wendel" > wrote in message


> I work a lot in a symphony orchestra and we get above 120 dB quite
> frequently on stage and the quietest is not very quiet at all ...

Right. The room when empty will still read above +15 dB due to ambient noise
and air conditioning noise. Add a few thousand people sitting quietly, and
you might be well above +30 dB.

> (part of it ofcourse due to the fact that our ears can't really hear
> really soft things after having the 120+ roar going on everyday etc.)

Right, temporary and permanent threshold shift is a strong effect.

> I thunk the 120 was the 'actual technically possible dynamic range'
> and not the thing you'd use as a finished product ?

I think it's based on the myth that you could possibly have 0 dB in the same
room as you have 120 dB.

> This is a dynamic range that is totally worthless in the consumers end.
(IMHO)

Very few consumers have the resources to possibly reproduce it, aside from
the limitations of the ear.

> The more dynamic range you can work with i thought was better though ?

When you can't hear the quietest passages because the dynamic range is to
great, you're missing part of the music, right?

[Roger W. Norman]

"Justin Ulysses Morse" > wrote in message
m...
> How do you figure? If you turn down your stereo so that the full scale
> loud parts are now a comfortable 85dBSPL instead of 120, then how loud
> are the quietest (-120dBFS) parts? -35dBSPL? You can't really go
> below 0dBSPL. I mean you could, but it wouldn't be any more silent
> than silence. The -85dBFS passages would be equally silent. You've
> lost dynamic range. You now have a theoretical 85dB of dynamic range.
>
> Except whoops we forgot that there's environmental noise in the
> listening room. So there's really only 50dB of dynamic range.


One's ability to hear 120 dB of dynamic range has nothing to do with the
material amplified to hear that 120 dB of dynamic range (which may be
quieter or louder). The factor involved is that, tested and proven, humans
have a tendency to compress the dynamic range of what they are hearing if
it's too loud. That point has been proven to be 120 dB of dynamic range,
whether those tests were done at the same time or at different times.
Certainly one could argue that -120 dB of dynamic range could be almost
impossible if someone were first subjected to the loudest possible noises
one could listen to. But if one measures where another can hear starting
from a clean slate and unadulterated ears, then that's the low point. And
the point where your brain starts to shut down your ears is the high point.
It's measurable, has been measurable for some time. That's how OSHA's
standards for noise level references have come about.

Would Jay bring it up if it were argueable? What I said was that the
dynamic range of the format, even at 120 dB, wouldn't be altered by simply
turning down the volume. Equates to moving farther back in the room, which
brings the inverse square law into play. Where you sit in a concert says
how loud it is. Certainly it's louder up front.

When I watched Focus doing Hocus Pocus onstage from 5 rows back, and the
guitar player broke a string that went "boing" rather quietly, did I not
hear it because those behind me didn't?

So I'm not trying to argue that 85 dBFS isn't 85 dBFS at 1 m/1 watt. Again,
it's the medium I was talking about and it's relationship to human hearing.
And somewhat towards human's developing technology that matches their
ability to perceive certain things, like hearing and seeing. One can view
the heavens with what one's eyes will do (telescope included) but one
doesn't necessarily understand as much as when the information becomes color
coded so you know what you're looking at that's outside of one's ability to
perceive. In the realm of music, there is no finer resolution than what you
can't hear and what you can. Unless you want to slow down or speed up the
tape.

And environmental noise has no effect on what people can ACTUALLY hear, just
on what they hear at the time. Non-argument.

--


Roger W. Norman
SirMusic Studio
Purchase your copy of the Fifth of RAP CD set at www.recaudiopro.net.
See how far $20 really goes.

[Roger W. Norman]

"Rob Adelman" > wrote in message
...
>
> > So a system and recording of minimal dynamic range turned up very loud
> > could still have a potential of huge range between a quiet passage (or
> > silence) and full on?
>
> Oh, and vise-versa, a system and recording with huge dynamic range
> listened to at very low level would really have very little range?
>

That's a point of perception, not a realistic measurement. Yes, turning up
the volume represents the same thing as moving some distance closer to the
event, and turning down the volume does the reverse, which is related to the
inverse square law. The concept behind a rating like X at 1 m/1 watt means
that there's something tangible in terms of a human's overall ability to
hear, meaning a testable result.

But you place the onus on turning up a quiet passage, or lowering a loud
one, than that's not a function of dynamic range. Listen again to Scott
Dorsey's submission to the Fifth of RAP CD. It will grab you at the depth
of a human's ability to hear. And, being recorded correctly, it left
nothing but to transfer it. Still not a true representation of a human's
ability to hear 120 dB of dynamic range, just an example of how much dynamic
range is on the media and how it is represented to you. A) If you turn the
volume up, does that mean the music still doesn't have it's range? No,
that's dependant upon your amps and speaker system. B) If you turn the
music down, does that diminish the perceived dynamic range? Yes. Does
question B suggest that question A is flawed? No. Two different things,
both based on what a human's extremes of audible range are.

--


Roger W. Norman
SirMusic Studio
Purchase your copy of the Fifth of RAP CD set at www.recaudiopro.net.
See how far $20 really goes.

[Mike Rivers]

In article > writes:

> Didn't UV22 used to spread the noise
> throughout the audible spectrum, somewhat akin to very low level tape hiss?

No. The 22 in UV22 refers to 22 kHz, where they pile as much of the
noise as they can.



--
I'm really Mike Rivers - )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

[Justin Ulysses Morse]

Rob Adelman > wrote:

> Something I have been wondering about for a while now. This is a bit
> confusing. Isn't any dynamic range specification actually a potential
> dynamic range? Depending on listening environment, speaker system,
> amplifier capacity, and most importantly the position of the volume
> control?
>
> So a system and recording of minimal dynamic range turned up very loud
> could still have a potential of huge range between a quiet passage (or
> silence) and full on?

Your next post, suggesting the opposite, is spot-on. But if you take a
system with limited dynamic range and crank up the volume, you don't
get any more dynamic range because the noise floor comes up too. You
don't have the "huge range between (silence) and full on" because there
is no silence. You can lose dynamic range, but you can't really invent
more once it's gone.

ulysses

[Justin Ulysses Morse]

Roger W. Norman > wrote:

> One's ability to hear 120 dB of dynamic range has nothing to do with the
> material amplified to hear that 120 dB of dynamic range (which may be
> quieter or louder).

We're not talking about one's ability to hear 120dB of dynamic range.
We're talking about listening to recordings with 120dB of dynamic
range. You can't have 120dB of dynamic range unless the recording is
reproduced with 120dBSPL passages. Or if you're listening with
mechanical ears that are more sensitive than human ears, in a room with
negative-SPL background noise.

<snip the totally irrelevant>

> Would Jay bring it up if it were argueable?

Maybe, but he didn't bring up any of this stuff you're talking about.
Don't blame him.

> What I said was that the dynamic range of the format, even at 120 dB,
> wouldn't be altered by simply turning down the volume.

Yes, you said that. But it still isn't true. "The Format" includes
the playback equipment. Even theoretically ideal playback equipment
can't pull this off.

> Equates to moving farther back in the room, which brings the inverse
> square law into play. Where you sit in a concert says how loud it
> is. Certainly it's louder up front.

So what? You move back from the stage, the quiet parts settle into the
noise floor. Less dynamic range.

<snip more of the totally irrelevant>

> And environmental noise has no effect on what people can ACTUALLY hear, just
> on what they hear at the time.

We're not talking about human physiology here. It doesn't matter if
you can hear 0dBSPL sounds in an anechoic chamber as well as 120dBSPL
sounds on the tarmac because you'll never get the two together in the
same place.

> Non-argument.

Says you.


ulysses

[Bob Cain]

Arny Krueger wrote:
>
> In DBTs I've found that the ability to hear small differences tends to go
> away when the levels are too high, just as surely the ability to hear small
> differences goes away when the levels are too low.

Part of the reason is that the eardrum has a mechanical
mechanism attached to it which physically constrains its
compliance and dynamic range in the presence of high SPL
sound fields.


Bob
--

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

A. Einstein

[Rob Adelman]

Bob Cain wrote:


>
>
> Part of the reason is that the eardrum has a mechanical
> mechanism attached to it which physically constrains its
> compliance and dynamic range in the presence of high SPL
> sound fields.

Ahhh, so that's why that Aerosmith concert sounded so compressed...

[Kalman Rubinson]

On Thu, 20 Nov 2003 14:18:38 -0600, Rob Adelman
> wrote:

>Bob Cain wrote:

>> Part of the reason is that the eardrum has a mechanical
>> mechanism attached to it which physically constrains its
>> compliance and dynamic range in the presence of high SPL
>> sound fields.
>
>Ahhh, so that's why that Aerosmith concert sounded so compressed...

That has been my experience at every rock concert I've been to.
(Admittedly, they are few in number.)

Kal

[Arny Krueger]

"Rob Adelman" > wrote in message

> Bob Cain wrote:
>
>
>>
>>
>> Part of the reason is that the eardrum has a mechanical
>> mechanism attached to it which physically constrains its
>> compliance and dynamic range in the presence of high SPL
>> sound fields.
>
> Ahhh, so that's why that Aerosmith concert sounded so compressed...

Shame on you for not bringing ear protection...