The HDV format...Compressed audio?

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

>> The issue isn't 'can we hear what was taken away' - of course we can -
>
> No. Laboratory tests indicate that in normal listening conditions people CAN
> NOT hear the difference. It is based on the psychoacoustics of masking.

My point is, since HDV audio is starting with less than linear PCM, what
happens in further stages distribution and broadcast where additional data
compression algorithms are applied to audio that is already data compressed.

OTOH, would we be any better off if the audio was captured at 192 kHz and
24-bit before data reduction, or would squeezing 16-bit 48kHz to the same
final size be worse?

It might make sense to go with better (24-bit, 192kHz) A/D converters, might
it not?

Dunno.

Regards,

Ty Ford




-- Ty Ford's equipment reviews, audio samples, rates and other audiocentric
stuff are at www.tyford.com

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

On Sat, 5 Feb 2005 14:20:44 -0500, wolf wrote
(in article <fZWdndJR5YRfhpjfRVn-2w@comcast.com>):

> Subject: Re: The HDV format...Compressed audio?
> From: wolf <wolfvidREPLACEWITH_ATcomcast.net>
> Date: Yesterday 2:20 PM
> Newsgroups: rec.video.desktop, rec.video.production,
> rec.arts.movies.production.sound
>
> I heard and saw an HDV-Z1 camera tape projected and played on an excellent
> system by Sony at the DV expo in LA. I was blown away by it. They only used
> the on board stereo mike but it was extremely low noise and had decent phase

> accuracy in stereo that held rock solid. So I can assume that the audio
> compression is minimal ( phase accuracy is the first to get lost usually -
> listen to your I pod at low res.). Bandwidth was good and detail was good. So

> judging with just my ears there is nothing wrong with the audio from the
> camera even at the price - and it was certainly better than any other DV
> corder I have heard

Again,

I'm not so much concerned about the single, first generation sound. What
concerns me is what happens when that audio is distributed; sent to DVD,
and/or broadcast, where other compression algorithms are typically applied.

The Herb Squire's demonstrations at AES NY a number of years ago have made me
very aware that we need to think beyond the first generation...all the way to
distribution and satellite or HDTV broadcast, all of which have additional
compression schemes.

Regards,

Ty Ford


-- Ty Ford's equipment reviews, audio samples, rates and other audiocentric
stuff are at www.tyford.com

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

On Sat, 5 Feb 2005 21:07:38 -0800, "Alpha" <logos1@trip.net> wrote:

>20 kHz is the bandwidth of hearing. Even CD has headroom for anti-aliasing
>fall-off. So what is the point of 98 ks/sec sampling?

It is done so you don't get funny mirror-frequencies coming back into
your original audio, once you go back to the analogue domain. And,
audiophiles also say that higher sample-rates, also sample the
harmonics of the higher frequencies.

>The dynamic range of hearing is at best 120-130 dB. So what is the point of
>24 bit sampling?

More accuracy reduces rounding errors during editing, so the
mutilation of your original sound gets less.

cheers

-martin-

--
Can the terror of spam be included in the war on terror?

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

"Martin Heffels" <marybooks56@yahoo.com> wrote in message
news:vvbb01tog03tt2mou3mp22qng34qieuk3q@4ax.com...
> On Sat, 5 Feb 2005 21:07:38 -0800, "Alpha" <logos1@trip.net> wrote:
>
>>20 kHz is the bandwidth of hearing. Even CD has headroom for
>>anti-aliasing
>>fall-off. So what is the point of 98 ks/sec sampling?
>
> It is done so you don't get funny mirror-frequencies coming back into
> your original audio, once you go back to the analogue domain. And,
> audiophiles also say that higher sample-rates, also sample the
> harmonics of the higher frequencies.
>
>>The dynamic range of hearing is at best 120-130 dB. So what is the point
>>of
>>24 bit sampling?
>
> More accuracy reduces rounding errors during editing, so the
> mutilation of your original sound gets less.
>
> cheers
>
> -martin-
>
> --
> Can the terror of spam be included in the war on terror?

You can't hear higher than 20 Khz. Antialiasing is done with oversampling
and filtering.

Numerical accuracy in editing and effects is a different matter entirely.
It depends not a bit on the sampling rate.

 

[Wolf]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

I suggest following reading material: ( it will clear your mind)


PRODUCING GREAT SOUND FOR DIGITAL VIDEO

by Jay Rose
List Price: $39.95 Amazon Price: $31.96 (check BN.com often cheaper Paperback -
352 pages Book & CD edition (December1999) Miller Freeman Books; ISBN:
0879305975 ; http://www.dplay.com/book Make your video project sound better than
it looks. This book delivers solutions to specific problems throughout the
entire process of creating engaging audio for digital video. Written by Jay, a
CLEO and Emmy-winning sound designer, he explains hundreds of real-world
techniques you can use from pre-production through mix. You get how-tos, tips
and time-savers, plus tutorials on key skills such as dialog and music editing.
With an audio CD of sample tracks and diagnostic tools. I found the audio CD
very educational. I find smart things on every page of the book and cannot
imagine a professional soundmixer who has not wanted to write this book himself.
There are many many tips that carry good audio from location to post. There are
many hints for new folks and many clarifications for the experienced pro. Don't
go without it.

AUDIO POST PRODUCTION, also by Jay Rose.
Specializing in sound after the shoot, this new book has long chapters on
editing, equalizing, noise reduction, sound effects, using music effectively,
mixing for various media... as well as on building and wiring the perfect audio
set-up. This is a good introduction for location recordists. Have a look and
sound real smart when questions like; “ can you fix it in the mix” are raised.
430 pages List Price $ 44.95 http://www.dplay.com/book/ buy both books at a 30%
discount:
http://www.amazon.com/exec/obidos/ASIN/1578201160/ref%3Dnosim/wwwdplaycom/002-9309747-1188035




Sound for Film & Television

by Tomlinson Holman. Sound for Film and Television covers the broad field of
sound accompanying pictures, from the fundamentals through recording, editing,
and mixing for films, documentaries, and television shows. The book provides a
solid grounding in all aspects of the sound process. Basic principles are
presented with illustrations on how they affect the day-to-day activities on a
film or television set, in the editing room, and in the mix room. The
accompanying audio CD demonstrates the key concepts discussed in the book. ISBN:
0240802918 Publisher Pretty much the standard college text. Butterworth-Heinemann

http://www.amazon.com/exec/obidos/tg/detail/-/0240802918/002-2626545-0027220?vi=glance

dont buy my books they are too specific.

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

In article <110cuui23s6ej6e@corp.supernews.com>, "Alpha" <logos1@trip.net>
wrote:
....
> You can't hear higher than 20 Khz. Antialiasing is done with oversampling
> and filtering.

Statement A is up for debate, as is any categorical statement about the
limits of human capability. I'm sure neither God nor Darwin calibrated the
human ear in Hertz. Certainly there are plenty of mastering engineers who
claim to hear higher. There are plenty of double-blind tests showing a
difference between 48 kHz sampling and 96 kHz sampling (with good monitors
and trained listeners, of course).

Statement B implies that oversampling is necessary for antialiasing. It
isn't; oversampling is just a great way to eliminate the problems of
analog filters.

--
Correct address is spell out the letter j, AT dplaydahtcom
Clio- and Emmy-winning sound design
Learn audio for video at www.dplay.com

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

"Jay Rose CAS" <SEE-SIGFILE@rcn.com> wrote in message
news:SEE-SIGFILE-0602051844040001@192.168.1.101...
> In article <110cuui23s6ej6e@corp.supernews.com>, "Alpha" <logos1@trip.net>
> wrote:
> ...
>> You can't hear higher than 20 Khz. Antialiasing is done with
>> oversampling
>> and filtering.
>
> Statement A is up for debate, as is any categorical statement about the
> limits of human capability.

It is not at all open for debate. 20 khz is a laboratory based value...in
the real world it is less, in fact.

>I'm sure neither God nor Darwin calibrated the
> human ear in Hertz. Certainly there are plenty of mastering engineers who
> claim to hear higher. There are plenty of double-blind tests showing a
> difference between 48 kHz sampling and 96 kHz sampling (with good monitors
> and trained listeners, of course).

Give me the scientific links to such double-blind tests. Further, show me
real-world data. You can discriminate what you can't cognate.

>
> Statement B implies that oversampling is necessary for antialiasing. It
> isn't; oversampling is just a great way to eliminate the problems of
> analog filters.

There was no such implication.

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

On Sun, 6 Feb 2005 12:19:27 -0800, "Alpha" <logos1@trip.net> wrote:

>You can't hear higher than 20 Khz. Antialiasing is done with oversampling
>and filtering.

My dog can! (just kidding)

>Numerical accuracy in editing and effects is a different matter entirely.
>It depends not a bit on the sampling rate.

The higher the amount of samples, the more accurate you get a
reflection of the original analogue soundwave. If you look at the
sound-signal on a scope, you can see it's shape is very very complex.
Garbage in = garbage out.

cheers

-martin-

--
Can the terror of spam be included in the war on terror?

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

On a sunny day (Mon, 07 Feb 2005 08:12:39 +1100) it happened Martin Heffels
<richardmiller114@zwallet.com> wrote in
<ms1d01t5mhvb9pjl23464fo2qfj08cnn66@4ax.com>:

>>Numerical accuracy in editing and effects is a different matter entirely.
>>It depends not a bit on the sampling rate.
>
>The higher the amount of samples, the more accurate you get a
>reflection of the original analogue soundwave. If you look at the
>sound-signal on a scope, you can see it's shape is very very complex.
Given a preselected frequency range, whatever it is, say for example
20 Hz to 20000 Hz, there is a distinct limit to how many samples per
second you'd want to take.
Niquist's theorem tells us we should sample at least 2 x the maximum
audio frequency we want.
So that would be a little over 40 kHz.
That is how 44.1 kHz for CD came about.
The steepness of the filters used BEFORE the sampling forces you
to sample a bit faster then 2 x.
Very steep filter can cause all sorts of problems.
These days (20 years later or so) digital filters can be made to overcome
many of those problems, oversampling... etc..
If however you sample MUCH faster, as you seem to suggest, you only
add high frequency noise (if you did not change the preemphasis filter).

When doing mathematics on the digitized signal (mixing, volume, etc..), then
it does make a difference what accuray you use (double precision in C, or
many many bits wide resgisters in hardware, say FPGA).
After all multiplying 2 24 bit signals can give an 48 bit result.

>Garbage in = garbage out.
That makes no sense to me in relation to what was discussed.

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

In article <ms1d01t5mhvb9pjl23464fo2qfj08cnn66@4ax.com>, Martin Heffels
<richardmiller114@zwallet.com> wrote:

> On Sun, 6 Feb 2005 12:19:27 -0800, "Alpha" <logos1@trip.net> wrote:
>
> >Numerical accuracy in editing and effects is a different matter entirely.
> >It depends not a bit on the sampling rate.
>
> The higher the amount of samples, the more accurate you get a
> reflection of the original analogue soundwave. If you look at the
> sound-signal on a scope, you can see it's shape is very very complex.

You're talking about two different things. Numerical accuracy refers to
eliminating rounding errors _on_each_sample_ by using a higher bit depth
(and much higher precision calculations). This is necessary because every
processing step - eq, compression, rev, even simple fades - requires doing
multiplication or division on the individual sample values.

Raising the sample rate lets you record and recreate more of the high
frequency components - primarily harmonics - of the original signal.

"Accurate" reproduction requires both a high sample rate and a high bit
depth. "High", in this context, is in the eye of the user (or audiophile,
or marketing department...)

---

Much as it pains me to give any support to anything Alpha wrote, after he
so blindly missed the point about the uselessness of using phase inversion
to tell how transparent a psychoacoustic coding scheme is, and assumed
because I stated that there is no standard algorithm for modeling the
complete mechanical/neuro/psychological listening chain that "I no nothing
about acoustics"...

--
Correct address is spell out the letter j, AT dplaydahtcom
Clio- and Emmy-winning sound design
Learn audio for video at www.dplay.com

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

You might want to know the guys you are calling stupid and blurting out
"nonsense!" to have decades of experience at the top of the field, something
that, with luck, you may live long enough to have a shot at.

Alpha wrote:

> PS
>
> 20 kHz is the bandwidth of hearing. Even CD has headroom for anti-aliasing
> fall-off. So what is the point of 98 ks/sec sampling?

Nobody records to CD. We are making master recordings and higer sample rates
place aliasing products above the passband of human hearing.
>
> The dynamic range of hearing is at best 120-130 dB. So what is the point of
> 24 bit sampling?

The point is to begin with the best recording possible. Do you have an analog
cassette player? Why don't we just record on cassettes?


Put a name in your post or we start killfiling you.

John

 

[frank]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

On Sat, 5 Feb 2005 11:26:05 -0500, in 'rec.video.desktop',
in article <The HDV format...Compressed audio?>,
Ty Ford <tyreeford@comcast.net> wrote:

>As I'm going over the specs of the new Sony HDV camcorder, I notice audio is
>recorded as compressed audio, not as linear audio.

Yes, indeed, it's compressed, lossy compressed in fact. MPEG-1 Layer 2
(.mp2) at a 384 kbps data rate. That's a 4:1 compression ratio or to
put it another way, a 75 percent reduction (the compressed data set is
25 percent of the size of the original data set). The original LPCM
would have had a data rate of 1536 kbps.

It's not just the Sony camcorders (the HDR-FX1, HDR-FX1E, HVR-Z1U,
HVR-Z1J, and HVR-Z1E), but the HDV format itself. The audio is always
16-bit (word length), 48 kHz (sampling rate), 2-channel stereo in HDV.
This is the same format which is used on some DVD-Video discs, so if
you're heard any discs like this, then you know, at least potentially,
what sort of sound quality level to expect.

I know that when I first read the HDV specs back in May of last year,
I was half expecting to see use of a lossless (2:1) compression
scheme, but alas, this was not to be.

>I realize that video has been compressed rather massively in lots of formats
>without a huge outcry, but I think audio is more sensitive to compression,
>especially if one compression scheme is followed by others during
>distribution.

Very good point, which is why I think that if one is editing the HDV
directly, doing cuts-only editing, and burning the final output of the
project to DVD-Video disc, it might be best to leave the audio in its
original format all of the way through the process, including burning
it to disc as MPEG-1 Layer II instead of transcoding to Dolby Digital
(yet another lossy format) as so many folks do, or if available disc
space permits, burning it to disc as unencoded LPCM.

Of course, this assumes that you're not editing the audio track in any
way as well, including even simple level changes (normalization, for
example).

>You have an ISDN V/O, a DVD, a broadcast. At the lowest posssible, that's
>three different compression algorithms (not counting more if you store the
>program on a networ or TV station's hard drive. If the station's transmitter
>is not located near master control, you might have a digital STL that uses
>it's own compression.

Transcoding from one lossy format to another lossy format will
certainly degrade the overall sound quality. It's not much different
in this respect than say, transcoding a .wma (Windows Media Audio)
file or a .ra (RealAudio) file to .mp3 (MPEG Layer III). There will be
a loss.

>Are we walking down a dangerous path here?

As a bit of an audiophile in a former life, I was extremely upset to
see the use of lossy compressed audio in the HDV format. To my way of
thinking, that's _not_ progress, but I can see where it would be a
natural choice considering the use of MPEG-2 compression for the video
track.

OTOH, if we consider that the general public, especially younger
people (a demographic of which I am no longer a member), seem for the
most part to find MP3 at 128 kbps to be of acceptable quality, well,
what can I say. Personally, I wouldn't trade a single $50,000 home
audio system for a truckload of iPods even if that meant that I could
only listen to music (and watch television) in my living room.

An alternative, for quality-oriented videographers, is to record audio
to both the HDV tape *and* a separate recorder as well. There are a
number of reasonably priced technologies available today including
portable DAT recorders (HHB, for example), portable CD-R burners
(Marantz Pro), and several forms of flash memory devices, including
the new Roland/Edirol R-1 CompactFlash-based recorder which does
24-bit 44.1 kHz (CD quality) LPCM .wav file recording (125 minutes on
a 2 GB CF card).

>Regards,
>
>Ty Ford

--
Frank, Independent Consultant
New York, NY

[Please remove 'nojunkmail.' from address to reply via e-mail.]

Read Frank's thoughts on HDV at http://www.humanvalues.net/hdv/

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

Frank wrote:
>
> It's not just the Sony camcorders (the HDR-FX1, HDR-FX1E, HVR-Z1U,
> HVR-Z1J, and HVR-Z1E), but the HDV format itself. The audio is always
> 16-bit (word length), 48 kHz (sampling rate), 2-channel stereo in HDV.
> This is the same format which is used on some DVD-Video discs

Close. DVD-V discs use AC-3 encoding for their primary audio channels (though MPEG-2 is/was one of many optional formats which can additionally be provided.) It decodes to 48k but at depths of 16-24 bits.



> As a bit of an audiophile in a former life, I was extremely upset to
> see the use of lossy compressed audio in the HDV format. To my way of
> thinking, that's _not_ progress, but I can see where it would be a
> natural choice considering the use of MPEG-2 compression for the video
> track.
>
> OTOH, if we consider that the general public, especially younger
> people (a demographic of which I am no longer a member), seem for the
> most part to find MP3 at 128 kbps to be of acceptable quality, well,
> what can I say.

I see no problem with HDV's audio spec as a release format--it's an ideal match. As an acquisition format, however, it leaves much to be desired.

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

"wolf" <wolfvidREPLACEWITH_ATcomcast.net> wrote in message
news:MLedndorcf0cEpvfRVn-vg@comcast.com...
>I suggest following reading material: ( it will clear your mind)
>
>
> PRODUCING GREAT SOUND FOR DIGITAL VIDEO
>
> by Jay Rose
> List Price: $39.95 Amazon Price: $31.96 (check BN.com often cheaper
> Paperback - 352 pages Book & CD edition (December1999) Miller Freeman
> Books; ISBN: 0879305975 ; http://www.dplay.com/book Make your video
> project sound better than it looks. This book delivers solutions to
> specific problems throughout the entire process of creating engaging audio
> for digital video. Written by Jay, a CLEO and Emmy-winning sound designer,
> he explains hundreds of real-world techniques you can use from
> pre-production through mix. You get how-tos, tips and time-savers, plus
> tutorials on key skills such as dialog and music editing. With an audio CD
> of sample tracks and diagnostic tools. I found the audio CD very
> educational. I find smart things on every page of the book and cannot
> imagine a professional soundmixer who has not wanted to write this book
> himself. There are many many tips that carry good audio from location to
> post. There are many hints for new folks and many clarifications for the
> experienced pro. Don't go without it.
>
> AUDIO POST PRODUCTION, also by Jay Rose.
> Specializing in sound after the shoot, this new book has long chapters on
> editing, equalizing, noise reduction, sound effects, using music
> effectively, mixing for various media... as well as on building and wiring
> the perfect audio set-up. This is a good introduction for location
> recordists. Have a look and sound real smart when questions like; “ can
> you fix it in the mix” are raised. 430 pages List Price $ 44.95
> http://www.dplay.com/book/ buy both books at a 30% discount:
> http://www.amazon.com/exec/obidos/ASIN/1578201160/ref%3Dnosim/wwwdplaycom/002-9309747-1188035
>


OH, boy..








>
>
> Sound for Film & Television
>
> by Tomlinson Holman. Sound for Film and Television covers the broad field
> of sound accompanying pictures, from the fundamentals through recording,
> editing, and mixing for films, documentaries, and television shows. The
> book provides a solid grounding in all aspects of the sound process. Basic
> principles are presented with illustrations on how they affect the
> day-to-day activities on a film or television set, in the editing room,
> and in the mix room. The accompanying audio CD demonstrates the key
> concepts discussed in the book. ISBN: 0240802918 Publisher Pretty much
> the standard college text. Butterworth-Heinemann
>
> http://www.amazon.com/exec/obidos/tg/detail/-/0240802918/002-2626545-0027220?vi=glance
>
> dont buy my books they are too specific.
>

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

On Mon, 7 Feb 2005 00:54:58 -0800, "Alpha" <logos1@trip.net> wrote:

>It is not at all open for debate. 20 khz is a laboratory based value...in
>the real world it is less, in fact.

In labs people could hear up to 28 kHz. 20kHz is a value based on
calculations given the characteristics of the ear. But indeed, most
people's limit is 16 kHz.

> >I'm sure neither God nor Darwin calibrated the
>> human ear in Hertz. Certainly there are plenty of mastering engineers who
>> claim to hear higher. There are plenty of double-blind tests showing a
>> difference between 48 kHz sampling and 96 kHz sampling (with good monitors
>> and trained listeners, of course).
>
>Give me the scientific links to such double-blind tests. Further, show me
>real-world data. You can discriminate what you can't cognate.

Just like us you can Google around a bit yourself, can't you?

cheers

-martin-

--
Can the terror of spam be included in the war on terror?

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

Didn't Rupert Neves consoles go up to 100,000 Hz? Didn't he do that
because he felt you lost harmonics from the highs at lower frequencies
if you cut off at 20,000 Hz?
Isn't analog 75 db of dynamic range? Isn't DAT (anything 16 bit 48K) 45
db of dynamic range?

Is that why 24 bit 96K is "important"?



Martin Heffels wrote:
> On Mon, 7 Feb 2005 00:54:58 -0800, "Alpha" <logos1@trip.net> wrote:
>
>
>>It is not at all open for debate. 20 khz is a laboratory based value...in
>>the real world it is less, in fact.
>
>
> In labs people could hear up to 28 kHz. 20kHz is a value based on
> calculations given the characteristics of the ear. But indeed, most
> people's limit is 16 kHz.
>
>
>>>I'm sure neither God nor Darwin calibrated the
>>>human ear in Hertz. Certainly there are plenty of mastering engineers who
>>>claim to hear higher. There are plenty of double-blind tests showing a
>>>difference between 48 kHz sampling and 96 kHz sampling (with good monitors
>>>and trained listeners, of course).
>>
>>Give me the scientific links to such double-blind tests. Further, show me
>>real-world data. You can discriminate what you can't cognate.
>
>
> Just like us you can Google around a bit yourself, can't you?
>
> cheers
>
> -martin-
>

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

"Martin Heffels" <richardmiller114@zwallet.com> wrote in message
news:r2me01lod3i5ihtdj0frsmsnoi1hoba3l1@4ax.com...
> On Mon, 7 Feb 2005 00:54:58 -0800, "Alpha" <logos1@trip.net> wrote:
>
>>It is not at all open for debate. 20 khz is a laboratory based value...in
>>the real world it is less, in fact.
>
> In labs people could hear up to 28 kHz. 20kHz is a value based on
> calculations given the characteristics of the ear. But indeed, most
> people's limit is 16 kHz.
>
>> >I'm sure neither God nor Darwin calibrated the
>>> human ear in Hertz. Certainly there are plenty of mastering engineers
>>> who
>>> claim to hear higher. There are plenty of double-blind tests showing a
>>> difference between 48 kHz sampling and 96 kHz sampling (with good
>>> monitors
>>> and trained listeners, of course).
>>
>>Give me the scientific links to such double-blind tests. Further, show me
>>real-world data. You can discriminate what you can't cognate.
>
> Just like us you can Google around a bit yourself, can't you?
>
> cheers
>
> -martin-
>
> --
> Can the terror of spam be included in the war on terror?

I can google, and I find no adequate science that you mentioned. 28 Khz was
found in a single individual. Sure, people live to be 108 etc., but you
would not base a retirement system on that figure!

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

On a sunny day (Mon, 07 Feb 2005 07:37:16 -0800) it happened Raymond Collins
<rcol@intergate.bc.ca> wrote in <420789fa_1@dowco.com>:

>
>
>Didn't Rupert Neves consoles go up to 100,000 Hz? Didn't he do that
>because he felt you lost harmonics from the highs at lower frequencies
maybe he was designing Sonar?

>if you cut off at 20,000 Hz?
>Isn't analog 75 db of dynamic range? Isn't DAT (anything 16 bit 48K) 45
>db of dynamic range?
sample rate has nothing to do with dynamic range.
16 bits ranges from 0 to 65535

So a ratio of say 1 / 65536 between smallest and largest signal.
We talk voltage ratios,
20log(65536) = 96.323 dB
http://www.google.nl/search?hl=nl&lr=&oi=defmore&q=define:decibel+(dB)

>Is that why 24 bit 96K is "important"?
Dunno, when I was young I could hear 16kHz, 18 even.
I have trouble hearing 15625 these days, in the old days I could
tell you if the TV was in sync without looking.

But if your dog also listens to your music, yes, then it could be a
requirement.

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

Raymond Collins wrote:
> Didn't Rupert Neves consoles go up to 100,000 Hz? Didn't he do that
> because he felt you lost harmonics from the highs at lower frequencies
> if you cut off at 20,000 Hz?

Most all analog audio stages are designed to pass 100-300k, since analog filters (particularly when cascaded) create audible band phase anomalies if you put them much lower.

Digital filters can be designed not to do this, but are much easier to design for such when they operate on higher samplerate audio.

 

[frank]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

On Mon, 07 Feb 2005 10:17:30 -0800, in 'rec.video.desktop',
in article <Re: The HDV format...Compressed audio?>,
Kurt Albershardt <kurt@nv.net> wrote:

>Frank wrote:
>>
>> It's not just the Sony camcorders (the HDR-FX1, HDR-FX1E, HVR-Z1U,
>> HVR-Z1J, and HVR-Z1E), but the HDV format itself. The audio is always
>> 16-bit (word length), 48 kHz (sampling rate), 2-channel stereo in HDV.
>> This is the same format which is used on some DVD-Video discs
>
>Close. DVD-V discs use AC-3 encoding for their primary audio channels
>(though MPEG-2 is/was one of many optional formats which can
>additionally be provided.) It decodes to 48k but at depths of 16-24 bits.

Sorry, Kurt, I wasn't purposely trying to be vague or imprecise, just
trying to point out that MPEG Layer II audio is sometimes used for
encoding the stereo audio tracks on DVD-Video discs and that some
folks might have heard such tracks at one time or another without even
being aware of it.

Virtually all DVD-V discs these days, whether PAL or NTSC and
regardless of Region coding, have Dolby Digital AC-3 tracks, but in
the olden days (six or seven years ago) it wasn't all that uncommon
(at least for me) to encounter discs, especially Region 2 PAL discs,
which had only MPEG audio tracks. I believe that the audio portion of
the PAL DVD Video spec was revised sometime back in late '97 or early
'98 to permit the use of AC-3 tracks only - no MPEG or LPCM. The
original audio portion of the PAL DVD-Video spec mandated use of MPEG
and/or LPCM tracks.

I also seem to recall that a certain provider of consumer-level DVD
authoring software (Ulead Systems, if memory serves) once came out
with an authoring package which produced discs which had only MPEG
audio tracks. I believe that this situation was rectified in a
subsequent release of their software although I suppose that a number
of people purchased the original version without being cognizant of
this limitation.

MPEG audio (at 16/44.1 and usually at a data rate of 224 kbps) is also
used on Green Book and White Book CDi and VCD 1.1 and 2.0 discs.

I'm in New York (Region 1 NTSC) but deal with discs from all over the
world. I'm fortunate in having an all-region (region-free)
multi-standard (PAL/NTSC) DVD player (a Marantz Pro PMD930) and a
multi-standard (PAL/NTSC) television monitor (a Sony PVM-14M4U). With
this combination I've been able to play any DVD-Video disc that I've
ever encountered, regardless of source.

>I see no problem with HDV's audio spec as a release format--it's an
>ideal match. As an acquisition format, however, it leaves much to be
>desired.

I almost screamed when I first looked at the HDV spec and saw MPEG-2
Video and MPEG-1 Layer II audio. I have DTV cable television service
and therefore see MPEG-2 video used as a distribution format all the
time. I strongly believe that MPEG-2 video is a distribution format
and was never intended to be used as an acquisition format, especially
in situations where it's going to be edited.

--
Frank, Independent Consultant
New York, NY

[Please remove 'nojunkmail.' from address to reply via e-mail.]

Read Frank's thoughts on HDV at http://www.humanvalues.net/hdv/

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

On Mon, 07 Feb 2005 07:37:16 -0800, Raymond Collins
<rcol@intergate.bc.ca> wrote:

>Isn't analog 75 db of dynamic range? Isn't DAT (anything 16 bit 48K) 45
>db of dynamic range?

Every bit in digital gives 6dB dynamic range, which makes 16 bit to 96
dB.

>Is that why 24 bit 96K is "important"?

Probably My ears got damaged which cost me a lot of the high
frequency range, so I won't hear the difference 🙁

cheers

-martin-

--
Can the terror of spam be included in the war on terror?

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

On Mon, 7 Feb 2005 13:17:30 -0500, Kurt Albershardt wrote
(in article <36pploF51eamcU1@individual.net>):

> Frank wrote:
>>
>> It's not just the Sony camcorders (the HDR-FX1, HDR-FX1E, HVR-Z1U,
>> HVR-Z1J, and HVR-Z1E), but the HDV format itself. The audio is always
>> 16-bit (word length), 48 kHz (sampling rate), 2-channel stereo in HDV.
>> This is the same format which is used on some DVD-Video discs
>
> Close. DVD-V discs use AC-3 encoding for their primary audio channels
> (though MPEG-2 is/was one of many optional formats which can additionally be
> provided.) It decodes to 48k but at depths of 16-24 bits.

NOT close. The Sony HDV camera is also able to record in DV mode and in THAT
mode it's 16-bit 48kHz. In HDV it's data reduced.


>> As a bit of an audiophile in a former life, I was extremely upset to
>> see the use of lossy compressed audio in the HDV format. To my way of
>> thinking, that's _not_ progress, but I can see where it would be a
>> natural choice considering the use of MPEG-2 compression for the video
>> track.
>>
>> OTOH, if we consider that the general public, especially younger
>> people (a demographic of which I am no longer a member), seem for the
>> most part to find MP3 at 128 kbps to be of acceptable quality, well,
>> what can I say.
>
> I see no problem with HDV's audio spec as a release format--it's an ideal
> match. As an acquisition format, however, it leaves much to be desired.

That would be me agreeing.

Regards,

Ty



-- Ty Ford's equipment reviews, audio samples, rates and other audiocentric
stuff are at www.tyford.com

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

In a fit of passion to prove he's right even when the facts prove
otherwise, Alpha wrote:
> I can google, and I find no adequate science that you mentioned. 28 Khz was
> found in a single individual. Sure, people live to be 108 etc., but you
> would not base a retirement system on that figure!

I would if I were that person.

John Blankenship,
Indianapolis
(email: my initials at mw daht net)

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

> Given the traditional level of importance that camcorder makers
> have placed on audio quality, is a decline even possible? Maybe
> this indicates that, despite the VX2000 furor, Sony (and others
> probably) are still as clueless as ever about audio in video.
>
>

This is why all serious recording work that I do does not rely on the camera
audio for anything but sync purposes. I use the MOTU 896 and eight large
diaphragm condenser mics to capture both a 5.1 surround mix and a separate
stereo mix from alternate locations on-stage. Camera audio is horrific and
only useful in emergency ENG situations, IMHO.


--
Best Regards,

Mark A. Weiss, P.E.
www.mwcomms.com
-

 

[Guest]

Archived from groups: rec.video.desktop,rec.video.production,rec.arts.movies.production.sound (More info?)

I'd like to hear more on this. I was under the impression that if a DV
camera was provided with an excellent sound source, as opposed to the
built-in mics, that the sound on most camcorders is actually quite good,
at least at 16 bit.

I can't believe I'm going to have to start hauling a Nagra around with
me....

Mark & Mary Ann Weiss wrote:
>>Given the traditional level of importance that camcorder makers

> stereo mix from alternate locations on-stage. Camera audio is horrific and
> only useful in emergency ENG situations, IMHO.
>
>
> --
> Best Regards,
>
> Mark A. Weiss, P.E.
> www.mwcomms.com
> -
>
>
>