Sign in with
Sign up | Sign in
Your question

QUESTION: 16-Bit vs. 24-Bit

Last response: in Home Audio
Share
Anonymous
August 16, 2004 7:28:43 PM

Archived from groups: rec.audio.pro (More info?)

Now that I finally have a recording rig that can record in 24-bit audio with a
96mHz sample rate, I was wondering something.

Do I have to turn my final mixes into 16-bit, 44.1mHz files in order to burn
them onto a CD?

In other words, is it possible to retain the 24-bit, 96mHz sound when burning
onto CD?

More about : question bit bit

Anonymous
August 16, 2004 7:28:44 PM

Archived from groups: rec.audio.pro (More info?)

"HWBossHoss" <hwbosshoss@aol.com> wrote in message
news:20040816112843.20060.00001003@mb-m25.aol.com

> Now that I finally have a recording rig that can record in 24-bit
> audio with a 96mHz sample rate, I was wondering something.

96 KHz is generally worthless for audio, but the 24 bit thing is nice for
things like building in some reserves for headroom.

> Do I have to turn my final mixes into 16-bit, 44.1mHz files in order
> to burn them onto a CD?

That would be 16 bit, 44.1 KHz


> In other words, is it possible to retain the 24-bit, 96mHz sound when
> burning onto CD?

Not if you want to play them on a regular CD player.
Anonymous
August 16, 2004 7:28:44 PM

Archived from groups: rec.audio.pro (More info?)

HWBossHoss <hwbosshoss@aol.com> wrote:
>Now that I finally have a recording rig that can record in 24-bit audio with a
>96mHz sample rate, I was wondering something.
>
>Do I have to turn my final mixes into 16-bit, 44.1mHz files in order to burn
>them onto a CD?

Yes.

>In other words, is it possible to retain the 24-bit, 96mHz sound when burning
>onto CD?

The wider word is still a good thing because it allows you more slop in
setting levels, even though you are dithering down to 16-bit. But recording
at 44.1 is definitely a good idea, because you will lose quality in the
sample rate conversion from 96->44.1. Odds are you will get better sound
recording at 44.1 directly than taking the SRC route.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Related resources
Anonymous
August 16, 2004 9:12:51 PM

Archived from groups: rec.audio.pro (More info?)

hwbosshoss@aol.com (HWBossHoss) wrote in
news:20040816112843.20060.00001003@mb-m25.aol.com:

>
> Now that I finally have a recording rig that can record in 24-bit
> audio with a 96mHz sample rate, I was wondering something.
>
> Do I have to turn my final mixes into 16-bit, 44.1mHz files in order
> to burn them onto a CD?
>
> In other words, is it possible to retain the 24-bit, 96mHz sound when
> burning onto CD?

16 bit 44.1 kHz is required for CD's

You can burn audio DVD's (DVD-A format) at 24-bit 96 kHz that will play on
some DVD players but not on any CD player.
August 16, 2004 11:26:54 PM

Archived from groups: rec.audio.pro (More info?)

Nope. You have to turn them into 16-bit, 44. However, it is better to work
with 24/96 until that final burn process. You'll see. Or, you'll hear :) 

Kalle

"HWBossHoss" <hwbosshoss@aol.com> wrote in message
news:20040816112843.20060.00001003@mb-m25.aol.com...
>
> Now that I finally have a recording rig that can record in 24-bit audio
with a
> 96mHz sample rate, I was wondering something.
>
> Do I have to turn my final mixes into 16-bit, 44.1mHz files in order to
burn
> them onto a CD?
>
> In other words, is it possible to retain the 24-bit, 96mHz sound when
burning
> onto CD?
Anonymous
August 17, 2004 12:42:07 AM

Archived from groups: rec.audio.pro (More info?)

The down-sample loss is a given, seeing as how you'd be discarding well over
50% of your samples, but I have found it's always better to start with good
stuff and downsample. The downsample loss is only noticable between the 96
source and the 44.1 target. The difference between something downed for
mastering and something that's 44.1 straight through is barely noticeable.
It's really the same thing with the 24 to 16 downing but, by mixing with a
better file, any adjustments made will be more precise. If you have the
power to work reasonably with the 24x96 files, go for it.

-gran


--
Dave Schein II, CSO
Printergy, Inc. - Moving a Million Documents to the Web!
www.printergy.com
DOCHighway, Inc. - Wherever You Are! Wireless!
www.dochighway.com
CDs, DVDs, Scanning, Document Management, Knowledge Sharing...
...The Future!
2066 York. Rd.
Suite 205
Baltimore, MD 21093
410-561-8436 - TEL
410-561-1220 - FAX
443-803-2119 - Direct


"Scott Dorsey" <kludge@panix.com> wrote in message
news:cfqs32$qfo$1@panix2.panix.com...
> HWBossHoss <hwbosshoss@aol.com> wrote:
> >Now that I finally have a recording rig that can record in 24-bit audio
with a
> >96mHz sample rate, I was wondering something.
> >
> >Do I have to turn my final mixes into 16-bit, 44.1mHz files in order to
burn
> >them onto a CD?
>
> Yes.
>
> >In other words, is it possible to retain the 24-bit, 96mHz sound when
burning
> >onto CD?
>
> The wider word is still a good thing because it allows you more slop in
> setting levels, even though you are dithering down to 16-bit. But
recording
> at 44.1 is definitely a good idea, because you will lose quality in the
> sample rate conversion from 96->44.1. Odds are you will get better sound
> recording at 44.1 directly than taking the SRC route.
> --scott
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
August 17, 2004 1:25:41 AM

Archived from groups: rec.audio.pro (More info?)

Okay Dave says 24x96 , Scot says 44.1 Ed says "I'll take a double "
Question: How does one know what kind of music the poster is going to record
and does it matter regarding 24/96 or 24/96?

Thank you for the info
Ed Bridge
Brooklyn N.Y.
www.bridgeclassicalguitars.com
""Granma" Dave Schein II, CSO" <granmadave@yahoo.com> wrote in message
news:z_8Uc.7860$Kf4.6812@nwrddc02.gnilink.net...
> The down-sample loss is a given, seeing as how you'd be discarding well
over
> 50% of your samples, but I have found it's always better to start with
good
> stuff and downsample. The downsample loss is only noticable between the
96
> source and the 44.1 target. The difference between something downed for
> mastering and something that's 44.1 straight through is barely noticeable.
> It's really the same thing with the 24 to 16 downing but, by mixing with a
> better file, any adjustments made will be more precise. If you have the
> power to work reasonably with the 24x96 files, go for it.
>
> -gran
>
>
> --
> Dave Schein II, CSO
> Printergy, Inc. - Moving a Million Documents to the Web!
> www.printergy.com
> DOCHighway, Inc. - Wherever You Are! Wireless!
> www.dochighway.com
> CDs, DVDs, Scanning, Document Management, Knowledge Sharing...
> ...The Future!
> 2066 York. Rd.
> Suite 205
> Baltimore, MD 21093
> 410-561-8436 - TEL
> 410-561-1220 - FAX
> 443-803-2119 - Direct
>
>
> "Scott Dorsey" <kludge@panix.com> wrote in message
> news:cfqs32$qfo$1@panix2.panix.com...
> > HWBossHoss <hwbosshoss@aol.com> wrote:
> > >Now that I finally have a recording rig that can record in 24-bit audio
> with a
> > >96mHz sample rate, I was wondering something.
> > >
> > >Do I have to turn my final mixes into 16-bit, 44.1mHz files in order to
> burn
> > >them onto a CD?
> >
> > Yes.
> >
> > >In other words, is it possible to retain the 24-bit, 96mHz sound when
> burning
> > >onto CD?
> >
> > The wider word is still a good thing because it allows you more slop in
> > setting levels, even though you are dithering down to 16-bit. But
> recording
> > at 44.1 is definitely a good idea, because you will lose quality in the
> > sample rate conversion from 96->44.1. Odds are you will get better
sound
> > recording at 44.1 directly than taking the SRC route.
> > --scott
> > --
> > "C'est un Nagra. C'est suisse, et tres, tres precis."
>
>
Anonymous
August 17, 2004 1:25:42 AM

Archived from groups: rec.audio.pro (More info?)

Edward Bridge <edbridgeNOSPAM@earthlink.net> wrote:
>Okay Dave says 24x96 , Scot says 44.1 Ed says "I'll take a double "
>Question: How does one know what kind of music the poster is going to record
>and does it matter regarding 24/96 or 24/96?

If your music has a lot of dynamic range, and if you are doing live work
where levels can be all over the place, the longer word is always a good
idea.

But to be honest, I have heard a lot of systems that sound worse at 96
ksamp/sec than they do at 44.1 Pick whichever sounds better with your
converters and go with it, but do expect substantial loss with SRC.

A lot of folks into the high speed conversion are recording at 88.2, since
the SRC down to 44.1 is much easier than a non-integral ratio SRC.

But the musical style shouldn't make any difference, because converters
are supposed to be accurate and produce the same output as input, rather
than euphonically colored.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
August 17, 2004 1:28:03 PM

Archived from groups: rec.audio.pro (More info?)

In article <pD9Uc.23969$nx2.20883@newsread2.news.atl.earthlink.net> edbridgeNOSPAM@earthlink.net writes:

> Okay Dave says 24x96 , Scot says 44.1 Ed says "I'll take a double "
> Question: How does one know what kind of music the poster is going to record
> and does it matter regarding 24/96 or 24/96?

If he's going to put it on a CD, it will end up at 16-bit anyway. The
advice is to record at higher resolution if there will be any
manipulation between the recording and the CD. It is unlikely that
anyone who needs to ask will be recording something for CD that might
actualy benefit from 96 kHz sample rate.

Some people can't just that "It's OK" for an answer.

--
I'm really Mike Rivers (mrivers@d-and-d.com)
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
August 17, 2004 2:55:03 PM

Archived from groups: rec.audio.pro (More info?)

you're too new at this to worry about 96K. that will just add
complexity and waste hard drive space/performance for you.

24 bit is very much a good idea.

what you do is record at 24 bit, and mix without thinking about it too
much. then when your mix sounds good and you want to "print it", put
a dithering program on the stereo master. put it on the very last
insert on the stereo master.

then when you "bounce to disk" your files will be 16 bit.

dither shrinks 24 bit down to 16 bit in a way that preserves most of
the information. some dithers are better than others.

do a google search on dither and read up on it. learn it, live it.
August 17, 2004 2:56:06 PM

Archived from groups: rec.audio.pro (More info?)

you're too new at this to worry about 96K. that will just add
complexity and waste hard drive space/performance for you.

but 24 bit is very much a good idea.

what you do is record at 24 bit, and mix without thinking about it too
much. then when your mix sounds good and you want to "print it", put
a dithering program on the stereo master. put it on the very last
insert on the stereo master.

then when you "bounce to disk" your files will be 16 bit.

dither shrinks 24 bit down to 16 bit in a way that preserves most of
the information. some dithers are better than others.

do a google search on dither and read up on it. learn it, live it.
Anonymous
August 17, 2004 5:12:28 PM

Archived from groups: rec.audio.pro (More info?)

<< A lot of folks into the high speed conversion are recording at 88.2, since
the SRC down to 44.1 is much easier than a non-integral ratio SRC. >>



This makes sense, logically and mathematically. But PLEASE forgive this newbie
for the next dumb question: What is "SRC"? Sample Rate Clock?
Anonymous
August 17, 2004 5:12:29 PM

Archived from groups: rec.audio.pro (More info?)

"HWBossHoss" <hwbosshoss@aol.com> wrote in message
news:20040817091228.19209.00003580@mb-m04.aol.com

> << A lot of folks into the high speed conversion are recording at
> 88.2, since the SRC down to 44.1 is much easier than a non-integral
> ratio SRC. >>



In the current context, the *ease* factor is a non-issue. Generally today,
downsampling is done with an algorithm that works equally well and the same,
for all sample sample rates in the desired range, whether integer multiples
or not. Hardware SRCs where the input is essentially arbitrary and the
output may be clocked by some other device are arguably the most complex
because the rates being converted don't necessrily have a fixed quotient.
You may think its 88.1 to 44.1 but its really 88.009 to 43.9996, right now,
but it shifts 0.0001% up or down a little while later.

> This makes sense, logically and mathematically. But PLEASE forgive
> this newbie for the next dumb question: What is "SRC"? Sample Rate
> Clock?

SRC = Sample Rate Conversion or Sample Rate Converter, etc.
Anonymous
August 17, 2004 5:12:29 PM

Archived from groups: rec.audio.pro (More info?)

HWBossHoss <hwbosshoss@aol.com> wrote:
><< A lot of folks into the high speed conversion are recording at 88.2, since
>the SRC down to 44.1 is much easier than a non-integral ratio SRC. >>


>
>This makes sense, logically and mathematically. But PLEASE forgive this newbie
>for the next dumb question: What is "SRC"? Sample Rate Clock?

Sample Rate Conversion or Sample Rate Converter.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
August 17, 2004 5:55:45 PM

Archived from groups: rec.audio.pro (More info?)

"Mike Rivers" <mrivers@d-and-d.com> wrote in message
news:znr1092743567k@trad...
>
.. The
> advice is to record at higher resolution if there will be any
> manipulation between the recording and the CD.

Like for films, plugin's or editing?

It is unlikely that
> anyone who needs to ask will be recording something for CD that might
> actualy benefit from 96 kHz sample rate.

I just got digi002 so at least I now _can_ record at 24/96 .lol. . So
what is the benefit from 96 khz?
Peace,
Ed Bridge
Brooklyn N.Y.
www.bridgeclassicalguitars.com
Anonymous
August 17, 2004 6:50:34 PM

Archived from groups: rec.audio.pro (More info?)

Also, if you are recording synths and samplers, there's little point to
using 96k, or even 24 bit, since the instruments will most likely be
outputting 16/44.1.

Albert

> If your music has a lot of dynamic range, and if you are doing live work
> where levels can be all over the place, the longer word is always a good
> idea.
>
> But to be honest, I have heard a lot of systems that sound worse at 96
> ksamp/sec than they do at 44.1 Pick whichever sounds better with your
> converters and go with it, but do expect substantial loss with SRC.
>
> A lot of folks into the high speed conversion are recording at 88.2, since
> the SRC down to 44.1 is much easier than a non-integral ratio SRC.
>
> But the musical style shouldn't make any difference, because converters
> are supposed to be accurate and produce the same output as input, rather
> than euphonically colored.
> --scott
Anonymous
August 17, 2004 8:31:06 PM

Archived from groups: rec.audio.pro (More info?)

In article <B7oUc.282$2L3.99@newsread3.news.atl.earthlink.net> edbridgeNOSPAM@earthlink.net writes:

> I just got digi002 so at least I now _can_ record at 24/96 .lol. . So
> what is the benefit from 96 khz?

As far as I can tell, extended frequency response beyond audibility.
But whether it's for this reason or some other, with very good
equipment, a 96 kHz recording actually seems to sound a little better
than a 48 kHz recording when compared to the original source (or what
comes straight out of the mic preamp). But I emphasize "very good
equipment." Nor have I taken this a step further and reduced both down
to "CD quality" and made another comparison.

I'd save your disk drive money and record at 44.1 or 48 kHz until
you're able to actually tell a difference yourself.


--
I'm really Mike Rivers (mrivers@d-and-d.com)
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
Anonymous
August 17, 2004 10:02:06 PM

Archived from groups: rec.audio.pro (More info?)

>
>I just got digi002 so at least I now _can_ record at 24/96 .lol. . So
>what is the benefit from 96 khz?
>Peace,
>Ed Bridge
>Brooklyn N.Y.
>www.bridgeclassicalguitars.com
>




You can support the economy by having to buy bigger hard drives. <G>



Richard H. Kuschel
"I canna change the law of physics."-----Scotty
August 17, 2004 10:23:42 PM

Archived from groups: rec.audio.pro (More info?)

"Arny Krueger" <arnyk@hotpop.com> writes:

>"HWBossHoss" <hwbosshoss@aol.com> wrote in message
>> this newbie for the next dumb question: What is "SRC"? Sample Rate
>> Clock?

>SRC = Sample Rate Conversion or Sample Rate Converter, etc.

.... not to be confused with The Scot Richard Case for those old enough
to recall.
Anonymous
August 18, 2004 9:17:13 PM

Archived from groups: rec.audio.pro (More info?)

"Carey Carlan" <gulfjoe@hotmail.com> wrote in message
news:Xns95478680F4FC8gulfjoehotmailcom@207.69.154.203...
> hwbosshoss@aol.com (HWBossHoss) wrote in
> news:20040816112843.20060.00001003@mb-m25.aol.com:

> You can burn audio DVD's (DVD-A format) at 24-bit 96 kHz that will play on
> some DVD players but not on any CD player.

Not quite true. There are lots of CD players now that will play DVDA or
SACD.
You should qualify by saying any *standard* CD player.

TonyP.
Anonymous
August 18, 2004 9:17:14 PM

Archived from groups: rec.audio.pro (More info?)

In article <41230295$0$9658$afc38c87@news.optusnet.com.au> TonyP@optus.net.com.au writes:

> There are lots of CD players now that will play DVDA or
> SACD.
> You should qualify by saying any *standard* CD player.

One that's at least five years old.

--
I'm really Mike Rivers (mrivers@d-and-d.com)
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
Anonymous
August 18, 2004 9:17:15 PM

Archived from groups: rec.audio.pro (More info?)

In article <znr1092828890k@trad>, Mike Rivers <mrivers@d-and-d.com> wrote:
>
>In article <41230295$0$9658$afc38c87@news.optusnet.com.au> TonyP@optus.net.com.au writes:
>
>> There are lots of CD players now that will play DVDA or
>> SACD.
>> You should qualify by saying any *standard* CD player.
>
>One that's at least five years old.

If it's got an IR laser, it's a CD player. If it has a blue or UV laser,
it's something different altogether.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
August 19, 2004 12:38:38 PM

Archived from groups: rec.audio.pro (More info?)

On 16 Aug 2004 15:28:43 GMT, hwbosshoss@aol.com (HWBossHoss) wrote:

>
>Now that I finally have a recording rig that can record in 24-bit audio with a
>96mHz sample rate, I was wondering something.
>
>Do I have to turn my final mixes into 16-bit, 44.1mHz files in order to burn
>them onto a CD?
>
>In other words, is it possible to retain the 24-bit, 96mHz sound when burning
>onto CD?
Anyone remember Nyquist? The theory is that you define your highest
frequency by 2 samples. At 44.1 the highest 22.05k. That is
basically a triangle wave regardless of its real shape. At 96k you
get approx. 4 samples. At 20k or less you get a truer sound. If you
stay at 24/96 until ready to burn, any manipulations you do (EQ, comp,
reverb, etc) in software will introduce less digital error. There are
aliasing considerations and other problems that are subtle, but can
add up, that improve when the processing is done at higher rates.
Anonymous
August 19, 2004 12:38:39 PM

Archived from groups: rec.audio.pro (More info?)

"MrCatnip" <mrcatnip@hotmail.com> wrote in message
news:fuo8i05qoea0g762psvciqe1b0o4k46hp2@4ax.com

> If you
> stay at 24/96 until ready to burn, any manipulations you do (EQ, comp,
> reverb, etc) in software will introduce less digital error.

This seems to vastly underestimate how profound downsampling to 44 from 96
actually is.

I can think of one advantage to high sample rates - the controls on some
digital equalizers (hardware or software) opearate a little stranglely near
the top frequency range, If you use a higher sample rate, they still act
strangely near the top frequency range, but that range is generally not
being considered. OTOH, if you run at 96 you have to contemplate how you'r
going to equalize the range from 20 to 44, and some people might get it
wrong.

> There are aliasing considerations and other problems that are subtle,
> but can
> add up, that improve when the processing is done at higher rates.

Aliasing is only an issue when downsampling, or at initial conversion. It is
not an issue during digital processing where the input and output are at the
same sample rate, and so is everything in-between.
Anonymous
August 21, 2004 10:05:59 AM

Archived from groups: rec.audio.pro (More info?)

"MrCatnip" <mrcatnip@hotmail.com> wrote in message
news:fuo8i05qoea0g762psvciqe1b0o4k46hp2@4ax.com

> Anyone remember Nyquist?

The question is, do you?

> The theory is that you define your highest
> frequency by 2 samples. At 44.1 the highest 22.05k.

So far so good.

> That is basically a triangle wave regardless of its real shape.

Not after passage through the reconstruction filter.

> At 96k you get approx. 4 samples.

So far so good.

> At 20k or less you get a truer sound.

In fact you get the same sine wave out of an appropriate digital signal,
either way.

> If you stay at 24/96 until ready to burn, any manipulations you do (EQ,
> comp,
> reverb, etc) in software will introduce less digital error.

Not at all. The final downsample have very profound effects and basically
defines the error.

> There are aliasing considerations and other problems that are subtle, but
> can
> add up, that improve when the processing is done at higher rates.

You can't have aliasing without downsampling or the equivalent. If you track
at 44.1 and burn at 44.1 no aliasing can be introduced by the processing
in-between, as long as it happens at 44.1 .
Anonymous
August 22, 2004 11:22:16 AM

Archived from groups: rec.audio.pro (More info?)

When it comes to the 24 VS the 16 bit debate, I doubt you'll find
anyone who disagrees that 24 bit is better. Where most of the debate
is at least for the past 8 years is about sample rates. Back in the
80's I remember it was that 44.1khz wasnt a high enough rate to
acuratley capture all the music lost between each sample.
Now that we have higher sampling rates and we CAN have a better
"picture" of our source (in our hearing range). The argument seems to
have shifted to the highest frequency we can reproduce and the
harmonics over 20khz we need.
Wasnt the quest originally to get as close as possible to analog
recording and achive that perfectly smooth "sine wave"? And then
enjoying enjoying all the benefits that digital recording gives like
dynamic range, better transient reproduction, and no need noise
reduction.
Have DAW companies and their suggestive marketing ploys clouded the
issue you think...?

-Dean



"Arny Krueger" <arnyk@hotpop.com> wrote in message news:<L5qdnRZRB7zkg7rcRVn-uA@comcast.com>...
> "MrCatnip" <mrcatnip@hotmail.com> wrote in message
> news:fuo8i05qoea0g762psvciqe1b0o4k46hp2@4ax.com
>
> > Anyone remember Nyquist?
>
> The question is, do you?
>
> > The theory is that you define your highest
> > frequency by 2 samples. At 44.1 the highest 22.05k.
>
> So far so good.
>
> > That is basically a triangle wave regardless of its real shape.
>
> Not after passage through the reconstruction filter.
>
> > At 96k you get approx. 4 samples.
>
> So far so good.
>
> > At 20k or less you get a truer sound.
>
> In fact you get the same sine wave out of an appropriate digital signal,
> either way.
>
> > If you stay at 24/96 until ready to burn, any manipulations you do (EQ,
> > comp,
> > reverb, etc) in software will introduce less digital error.
>
> Not at all. The final downsample have very profound effects and basically
> defines the error.
>
> > There are aliasing considerations and other problems that are subtle, but
> > can
> > add up, that improve when the processing is done at higher rates.
>
> You can't have aliasing without downsampling or the equivalent. If you track
> at 44.1 and burn at 44.1 no aliasing can be introduced by the processing
> in-between, as long as it happens at 44.1 .
Anonymous
August 22, 2004 8:46:32 PM

Archived from groups: rec.audio.pro (More info?)

"Dean Dydekl" <dydekd@bellsouth.net> wrote in message
news:faf20d88.0408220622.1d05cfe9@posting.google.com
> When it comes to the 24 VS the 16 bit debate, I doubt you'll find
> anyone who disagrees that 24 bit is better. Where most of the debate
> is at least for the past 8 years is about sample rates.

This area is pretty easy to study, since we have a wide selection of sample
rates to work with. even in fairly inexpensive equipment.

> Back in the 80's I remember it was that 44.1khz wasnt a high enough rate
> to
> acuratley capture all the music lost between each sample.

Except of course, it is well known that there ain't no such thing.

> Now that we have higher sampling rates and we CAN have a better
> "picture" of our source (in our hearing range).

There's no doubt that 192 KHz is way outside our hearing range, so that
better picture is well-within our existing selection of sample rates.

>The argument seems to
> have shifted to the highest frequency we can reproduce and the
> harmonics over 20khz we need.

Why argue, when its so easy to just listen?

> Wasnt the quest originally to get as close as possible to analog
> recording and achive that perfectly smooth "sine wave"?

The goal is sonic transparency, nothing more, nothing less.

> Have DAW companies and their suggestive marketing ploys clouded the
> issue you think...?

It's easy to uncloud. Here's some worked-out examples for your listening
pleasure:

http://www.pcabx.com/technical/sample_rates/index.htm
Anonymous
August 24, 2004 4:59:28 AM

Archived from groups: rec.audio.pro (More info?)

On Mon, 16 Aug 2004 17:28:43 +0200, HWBossHoss wrote:


> Now that I finally have a recording rig that can record in 24-bit audio
> with a 96mHz sample rate, I was wondering something.
>
> Do I have to turn my final mixes into 16-bit, 44.1mHz files in order to
> burn them onto a CD?

Yes as a last step

> In other words, is it possible to retain the 24-bit, 96mHz sound when
> burning onto CD?

No, but you can save your data on a data-CD or DVD. 24bits has advantages
for recording and processing. 96 khz is the best choice if you want the
freedom to go to 44.1, 48 and 96 khz without too much losses. New media
are often based on 48khz and multiples, CD's are 44.1 khz.

--
Chel van Gennip
Visit Serg van Gennip's site http://www.serg.vangennip.com
Anonymous
August 24, 2004 12:42:10 PM

Archived from groups: rec.audio.pro (More info?)

MrCatnip <mrcatnip@hotmail.com> wrote:
>Exactly and at 44.1 you have to filter them out and force it into a
>sine. If you would take the time to read, instead of making stupid
>insults, you would realize that I am not talking about what the "box"
>puts out after heavy filtering. You stated my case. If you create a
>22k triangular wave and sample it at 44.1 the triangle is out as a
>sine. Hense the actual wave is distorted from a triangle into "a
>really nice sine
>>wave. Smooth, rounded and actually one of the very nicest sine waves you've
>>ever seen in your life!"
>but it does represent the original. Sampling at 96 or 128 gives a
>more accurate picture.

Right, but if you can hear the 66 KHz third harmonic of a 22 KHz triangle
wave, I'll buy you a case of beer. It's possible, but nobody has yet given
any real evidence (and the Kanagawa Institute studies are NOT real evidence).

Not to mention, of course, that your speakers probably can't reproduce
any of that stuff anyway. Most tweeters drop like a rock above 20 KHz,
many of them drop like a rock before they even get that high.

Try playing back a 15 KHz square wave and a 15 KHz sine wave with the
fundamental at the same level. Sound similar?
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
August 25, 2004 12:13:16 AM

Archived from groups: rec.audio.pro (More info?)

Chel van Gennip wrote:
> On Mon, 16 Aug 2004 17:28:43 +0200, HWBossHoss wrote:
>
>
>> Now that I finally have a recording rig that can record in 24-bit
>> audio with a 96mHz sample rate, I was wondering something.
>>
>> Do I have to turn my final mixes into 16-bit, 44.1mHz files in order
>> to burn them onto a CD?
>
> Yes as a last step
>
>> In other words, is it possible to retain the 24-bit, 96mHz sound when
>> burning onto CD?
>
> No, but you can save your data on a data-CD or DVD. 24bits has
> advantages for recording and processing.

Recording in 24 bits allows program to be recorded at significantly lower
levels than 0dBFS, without getting into sonically-impaired territory.
Recording a signal peaking at -12dB in a 16 bit system and you are listening
to way fewer bits of reslution.


geoff
Anonymous
August 25, 2004 12:13:17 AM

Archived from groups: rec.audio.pro (More info?)

"Geoff Wood" <geoff@paf.co.nz-nospam> wrote in message news:<WMCWc.80$mZ2.4271@news02.tsnz.net>...
>
> Recording a signal peaking at -12dB in a 16 bit system and you are listening
> to way fewer bits of reslution.
>


Mmmm...reslution. Gots to get me some bits of that, not enough sluts
in my control room.
Anonymous
August 25, 2004 12:13:17 AM

Archived from groups: rec.audio.pro (More info?)

Geoff Wood wrote:

>Recording in 24 bits allows program to be recorded at significantly lower
>levels than 0dBFS, without getting into sonically-impaired territory.

This is correct in many, but not necessarily all, cases.

>Recording a signal peaking at -12dB in a 16 bit system and you are listening
>to way fewer bits of reslution.

This is wrong, on two counts.

First, at 6 dB per bit, recording a -12 dBfs means that there are two bits
that never change state. In a 16-bit system this means that 12.5% of the
bits go unused. 12.5% is rarely "way fewer" in my book.

Second, a 16-bit system provides 16-bits of resolution regardless of the
level to which it is driven. Resolution is the ability of a system's
ability to accurately measure its input signal. A 16-bit system can
measure its input single with an accuracy of one part in 65,536 or about
0.01038 dB. A 16-bit converter therefore resolves the difference between
-11.98962 dBfs and -12.00000 dBfs as well as it measures the difference
between -0.01038 dBfs and 0.00000 dBfs.

There are other reasons that 16-bit systems sound better than 24-bit
systems when not driven to nearly full scale. A designer is likely, for
example, to have put more effort into the analog portion of a 24-bit ADC
than that of a 16-bitter and the parts used in today's 24-bit converters
have the benefit of performance improvements over those used in yesterday's
16-bitters. The difference is not, however, due to "listening to way
fewer bits of resolution."

One needs no more dynamic range in a digital system than that of the
signals one is recording. A 16-bit ADC provides slightly more than 96 dB
of dynamic range. If one allocates that range for a nominal level of
-20 dBfs, the remaining 76 dB to between the noise floor and nominal
level probably exceeds that of even the finest studio environments and
their analog signal chains. Any additional resolution is simply unused.

You may ask, then, why the manufacturers don't market better 16-bit
converters rather than 24-bitters. The answer to this question lies in
the same kind of thinking that results in jacked up pickup trucks with
huge tires and that fills my email box with advertisements for "male
enhancement" products.

--
========================================================================
Michael Kesti | "And like, one and one don't make
| two, one and one make one."
mkesti@gv.net | - The Who, Bargain
Anonymous
August 25, 2004 1:07:12 AM

Archived from groups: rec.audio.pro (More info?)

On Thu, 19 Aug 2004 08:38:38 GMT, MrCatnip <mrcatnip@hotmail.com>
wrote:

>On 16 Aug 2004 15:28:43 GMT, hwbosshoss@aol.com (HWBossHoss) wrote:
>
>>
>>Now that I finally have a recording rig that can record in 24-bit audio with a
>>96mHz sample rate, I was wondering something.
>>
>>Do I have to turn my final mixes into 16-bit, 44.1mHz files in order to burn
>>them onto a CD?
>>
>>In other words, is it possible to retain the 24-bit, 96mHz sound when burning
>>onto CD?
>Anyone remember Nyquist? The theory is that you define your highest
>frequency by 2 samples. At 44.1 the highest 22.05k. That is
>basically a triangle wave regardless of its real shape. At 96k you
>get approx. 4 samples. At 20k or less you get a truer sound. If you
>stay at 24/96 until ready to burn, any manipulations you do (EQ, comp,
>reverb, etc) in software will introduce less digital error. There are
>aliasing considerations and other problems that are subtle, but can
>add up, that improve when the processing is done at higher rates.

Most of the arguments center around my statement that 2 points per
hertz can only be drawn as a triangular wave. The hue and cry went up
that it defines a sine wave. No, it does not. It becomes a sine wave
because that is the desired state and the signal is manipulated to
look like one. If I gave you a graph of the points and said tell me
what kind of signal produced it, you couldn't. Those of you who are
mired in the present would most likely say sine wave. Those of you
who are thinkers, but maybe not adventuresome, would say that you
can't really know. Finally, those of you who are thinkers and
adventuresome will come up with all kinds of signals from a 1000KV, 1
usec pulse at each dot to a multimegahertz signal that is a harmonic
to a digital pulse stream.

My statement that it could ONLY be drawn as a Triangle was too narrow
a statement. It could be drawn as anything your heart desires.
Anonymous
August 25, 2004 1:07:13 AM

Archived from groups: rec.audio.pro (More info?)

MrCatnip wrote:

>Most of the arguments center around my statement that 2 points per
>hertz can only be drawn as a triangular wave. The hue and cry went up
>that it defines a sine wave. No, it does not. It becomes a sine wave
>because that is the desired state and the signal is manipulated to
>look like one. If I gave you a graph of the points and said tell me
>what kind of signal produced it, you couldn't. Those of you who are
>mired in the present would most likely say sine wave. Those of you
>who are thinkers, but maybe not adventuresome, would say that you
>can't really know. Finally, those of you who are thinkers and
>adventuresome will come up with all kinds of signals from a 1000KV, 1
>usec pulse at each dot to a multimegahertz signal that is a harmonic
>to a digital pulse stream.

Those who understand band-limited digital sampling systems would know that
all in-band aspects of the original signal can be reproduced as long as more
than two samples per cycle of the highest in-band frequency were recorded.

>My statement that it could ONLY be drawn as a Triangle was too narrow
>a statement. It could be drawn as anything your heart desires.

Yes, and those hearts' desires would represent the original signal only
after being properly filtered.

Digital audio's goal is accurate reproduction of recorded sound, isn't it?

--
========================================================================
Michael Kesti | "And like, one and one don't make
| two, one and one make one."
mkesti@gv.net | - The Who, Bargain
Anonymous
August 25, 2004 1:07:13 AM

Archived from groups: rec.audio.pro (More info?)

In article <leani0p5qao48u4nsh457gh1l6os0rl69g@4ax.com> mrcatnip@hotmail.com writes:

> Most of the arguments center around my statement that 2 points per
> hertz can only be drawn as a triangular wave. The hue and cry went up
> that it defines a sine wave. No, it does not. It becomes a sine wave
> because that is the desired state and the signal is manipulated to
> look like one.

If you draw a line between two samples, and there are only two samples
(in round numbers) within one clock cycle, then the frequency of the
"triangle" waveform you are defining is is 1/2 the sample rate. Do you
understand that much?

Good.

Now if your system cuts off all frequencies higher than half the
sample rate, frequencies that are harmonics of the fundamental
frequency that comprises that triangle wave, what's left? A single
frequency at half the sample rate. And what's a single frequency?

A sine wave.

If you want to reproduce a 20 kHz triangle wave, or square wave, or
half-wave rectified pulses, or any waveforme that you can think of,
you have to be able to accommodate a wider bandwidth than the
fundamental frequency, because anything but a sine wave contains
frequencies that are higher than the fundamental.

> If I gave you a graph of the points and said tell me
> what kind of signal produced it, you couldn't.

That's true, if you know that it's at half the sample rate. But if
it's any lower in frequency, then you can completely reconstruct the
waveform. The trick is to assure that there are no frequencies greater
than half the sample rate. That makes a pretty big assumption - that
the waveform repeats, started at the beginning of time, and goes on
until the end of time. This is why there's a little "wiggle room" and
the working high frequency limit is a little below half the sample
rate.

> My statement that it could ONLY be drawn as a Triangle was too narrow
> a statement. It could be drawn as anything your heart desires.

Actualy what comes out of the system with such a waveform is only one
thing - a sine wave. And in fact, whatever went in would be a sine
wave after it went through the filter that cut off everything above
its fundamental frequency.

It's not just a good idea, it's the law!

--
I'm really Mike Rivers (mrivers@d-and-d.com)
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
Anonymous
August 25, 2004 4:56:33 AM

Archived from groups: rec.audio.pro (More info?)

"Geoff Wood" <geoff@paf.co.nz-nospam> wrote in message
news:WMCWc.80$mZ2.4271@news02.tsnz.net...
..
> Recording a signal peaking at -12dB in a 16 bit system and you are
listening
> to way fewer bits of reslution.

Loss of 2 bits regardless of recording. Now an interesting thing happens,
some soundcards have 96 dB DNR at 16 or 24 bits. (simply limited by the
analog stages and converters)
Record at -12dB, and you get 14 bits of data in either case, regardless of
how much hand waving and shouting that you do.

TonyP.
Anonymous
August 25, 2004 5:06:46 AM

Archived from groups: rec.audio.pro (More info?)

"MrCatnip" <mrcatnip@hotmail.com> wrote in message
news:vh4mi01g7eqdf1qtq6qpr2gd98ehm4qqp7@4ax.com...
> On Sun, 22 Aug 2004 07:08:51 -0400, "Arny Krueger" <arnyk@hotpop.com>
> >"MrCatnip" <mrcatnip@hotmail.com> wrote in message
> >news:7okgi0hi9p69m71joanmmrtjelord45adi@4ax.com
> >> On Sat, 21 Aug 2004 18:30:48 +1000, "TonyP" <TonyP@optus.net.com.au>
> >>> "MrCatnip" <mrcatnip@hotmail.com> wrote in message
> >>> news:fuo8i05qoea0g762psvciqe1b0o4k46hp2@4ax.com...
> >>>> Anyone remember Nyquist? The theory is that you define your highest
> >>>> frequency by 2 samples. At 44.1 the highest 22.05k. That is
> >>>> basically a triangle wave regardless of its real shape.

> >>> Actually NO. The ONLY 22kHz waveshape that can be described by a
> >>> 22.05 sampling rate is a SINE wave. Why, because a triangular wave
> >>> MUST have harmonics which are above the passband.

> Exactly and at 44.1 you have to filter them out and force it into a
> sine. If you would take the time to read, instead of making stupid
> insults, you would realize that I am not talking about what the "box"
> puts out after heavy filtering. You stated my case. If you create a
> 22k triangular wave and sample it at 44.1 the triangle is out as a
> sine. Hense the actual wave is distorted from a triangle into "a
> really nice sine


That's NOT what you wrote above, but your new argument is correct. NOW prove
to me that you can hear the difference between a 22kHz sine wave and a 22
kHz triangular wave. I've never met anyone that could when the fundamental
is matched for level.

TonyP.
Anonymous
August 25, 2004 2:41:12 PM

Archived from groups: rec.audio.pro (More info?)

On Tue, 24 Aug 2004 10:13:16 +0200, Geoff Wood wrote:

> Chel van Gennip wrote:
>> On Mon, 16 Aug 2004 17:28:43 +0200, HWBossHoss wrote: No, but you can
>> save your data on a data-CD or DVD. 24bits has advantages for recording
>> and processing.
>
> Recording in 24 bits allows program to be recorded at significantly
> lower levels than 0dBFS, without getting into sonically-impaired
> territory. Recording a signal peaking at -12dB in a 16 bit system and
> you are listening to way fewer bits of reslution.

As the discussion here gets a bit confusing about certain benefits, I
think it is wise to highlight the two points of discussion here:

Although not every soundcard trully implements a total 24 bits, 24 bits
has advantages while recording and processing. The introduced problems,
higher bandwidth, storage and processing load, have been overcome by
hardware developments.

Although different points of view exists around the question if the
difference between 44.1, 48 or 96khz are noticable, it is clear that your
recordings, if they are important, wil have to be converted to 44.1, 48,
and maybe 96khz now or in the future, because that are the frequencies
used in current media standards. Therefore 96khz is a better starting
point. Conversion from 44.1 to 48 khz, and visa versa, is a problem.
Conversion from 96khz to either 44.1 or 48 khz, does not impose serious
problems, if you are downsampling with a factor 2 or higher the most
important factor influencing the result is the target frequency. Again:
The introduced problems, higher bandwidth, storage and processing load,
have been overcome by hardware developments.

--
Chel van Gennip
Visit Serg van Gennip's site http://www.serg.vangennip.com
Anonymous
August 25, 2004 2:41:13 PM

Archived from groups: rec.audio.pro (More info?)

"Chel van Gennip" <chel@vangennip.nl> wrote in message
news:2p33b8FfpusmU1@uni-berlin.de
>
> Although different points of view exists around the question if the
> difference between 44.1, 48 or 96khz are noticable,

Viewpoints are like butt-holes - everybody has at least one. However, there
is only one physical universe, and in that physical universe there are no
audible differences between good implementations any of these sample rates.
Given that consumers have given broad acceptance to formats that do
introduce well-known audible compromises such as MP3, AAV, DTS and AC3 (DD)
it is clear that should be some marginal audible benefit that has escaped 10
or more years of careful investigation, it has no practical advantage in the
marketplace.

IOW, when the market has had one highly-popular *sonic overkill* format
called CD audio for over 20 years, and it is in fact the marketplace's
benchmark for audio quality, and it is sold every street corner and being
reproduced by consumers in many of the homes between the street corners,
there is just no widespread perceived need for addtional *sonic overkill*
formats.

>it is clear that
> your recordings, if they are important, wil have to be converted to
> 44.1, 48, and maybe 96khz now or in the future, because that are the
> frequencies used in current media standards.

In fact the so-called high resolution formats are failing in the
marketplace. SACD was slated for cancellation earlier this year, but somehow
it got a reprive. How long will they last?

Closer to the topic, it is also true that a goodly proportion of existing
narrowly-distributed, slow-selling so-called high resolution format
recordings are simply remasterings of earlier low-resolution formats. Thus,
the claim that upsampled recordings are problematical in the marketplace for
so-called high resolution formats, no matter the bogus technology this
actually represents, is false. We have to understand that so-called high
resolution recordings are about perceived value in the absense of any
discernable technical value. They are numbers for the sake of numbers. The
weakest links in the production chain are elsewhere and they defines the
ultimate sound quality, not sample rates > 44.1 KHz.

> Therefore 96khz is a better starting point. Conversion from 44.1 to 48
> khz, and visa
> versa, is a problem.

This is a false claim, technologically speaking. Modern upsampling
algorithms do not require or even show a discernable benefit from claims
based on the abstract study of numerology.


> Conversion from 96khz to either 44.1 or 48 khz,
> does not impose serious problems, if you are downsampling with a
> factor 2 or higher the most important factor influencing the result
> is the target frequency.

Yet another false claim based on what appears to be yet another intuitive
study of numerology. In fact the most delicate processing step in
downsampling is the digital filtering, not the decimation.

>Again: The introduced problems, higher bandwidth, storage and processing
>load, have been overcome by
> hardware developments.

Now that I agree with. However in practical use downsampling from 96 KHz can
be a major time-waster. Given that high sample rates have no practical
audible benefit, its just time wasted.
Anonymous
August 25, 2004 3:27:09 PM

Archived from groups: rec.audio.pro (More info?)

MrCatnip <mrcatnip@hotmail.com> wrote:
>I'm sorry my phrasing confused you, I meant that you have to jam the
>signal through a brick wall to get anything that is not badly messed
>up by aliasing.

Why is this a bad thing?

It used to be a bad thing in the days before oversampling, but for the
last 15 years, phase shift at the end of the passband has not really
been an issue.

>Why are you stating the obvious? I don't remember anybody arguing
>that tape and vinyl was better than CD's. I WILL say that DVD audio
>exceeds the capabilities of CD's recorded at 16/44 though.

I thought you were arguing that wide bandwidth was tied to accuracy?
My analogue tape machine has a -3 dB point at 35 KHz. Certainly wider
bandwidth than CDs have.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
August 25, 2004 4:42:38 PM

Archived from groups: rec.audio.pro (More info?)

On Wed, 25 Aug 2004 11:48:00 +0200, Arny Krueger wrote:

> "Chel van Gennip" <chel@vangennip.nl> wrote in message
> news:2p33b8FfpusmU1@uni-berlin.de
> IOW, when the market has had one highly-popular *sonic overkill* format
> called CD audio for over 20 years, and it is in fact the marketplace's
> benchmark for audio quality, and it is sold every street corner and
> being reproduced by consumers in many of the homes between the street
> corners, there is just no widespread perceived need for addtional *sonic
> overkill* formats.

Agreed. There is indeed a lot of overkill in technological fuzz in the
market. Dynamics of sound is better served bij leaving the compression of
dynamic range out of the production cycle than by more bits of resolution
in the mediumformat. I think 96 dB is more than sufficient for sound
reproduction. In the sound production process a bit more resolution does
simplify the process. If you want to compress the dynamic range in the end
result and want to have a 96 dB dynamic range in the end result, you need
more dynamic range in the production process.

>>it is clear that
>> your recordings, if they are important, wil have to be converted to
>> 44.1, 48, and maybe 96khz now or in the future, because that are the
>> frequencies used in current media standards.
>
> In fact the so-called high resolution formats are failing in the
> marketplace. SACD was slated for cancellation earlier this year, but
> somehow it got a reprive. How long will they last?

The DVD audio channel normally is 48khz or a multitude, I think that will
last.

>> Therefore 96khz is a better starting point. Conversion from 44.1 to 48
>> khz, and visa
>> versa, is a problem.
>
> This is a false claim, technologically speaking. Modern upsampling
> algorithms do not require or even show a discernable benefit from claims
> based on the abstract study of numerology.
>
>
>> Conversion from 96khz to either 44.1 or 48 khz, does not impose serious
>> problems, if you are downsampling with a factor 2 or higher the most
>> important factor influencing the result is the target frequency.
>
> Yet another false claim based on what appears to be yet another
> intuitive study of numerology. In fact the most delicate processing step
> in downsampling is the digital filtering, not the decimation.

Digital filtering too is less complex and more accurate at 1/4 of the
samplerate. (As most other processing done on the tracks done during the
production process) If all has been done well, the ultimate limitation is
the samplerate of the final result.

>>Again: The introduced problems, higher bandwidth, storage and
>>processing load, have been overcome by
>> hardware developments.
>
> Now that I agree with. However in practical use downsampling from 96 KHz
> can be a major time-waster. Given that high sample rates have no
> practical audible benefit, its just time wasted.

It cost some time, but it is not in the interactive part, it is part of a
batch process. BTW, a good MP3 conversion, that is part of the same batch
run for me, takes a lot more time.

--
Chel van Gennip
Visit Serg van Gennip's site http://www.serg.vangennip.com
Anonymous
August 25, 2004 7:32:58 PM

Archived from groups: rec.audio.pro (More info?)

"Chel van Gennip" <chel@vangennip.nl> wrote in message
news:2p3amqFg6ldhU1@uni-berlin.de

> I think 96 dB is more than
> sufficient for sound reproduction.

I know that anybody who mentions 96 dB and "audio production" in the same
sentence is very short on practical experience with actually measuring the
dynamic range of music recorded in the real world.

The widest dynamic range I've ever found in a commercial recording was 73-75
dB. A typical live recording with audience present will drop to the high
50s or low 60s.
Anonymous
August 26, 2004 4:46:22 AM

Archived from groups: rec.audio.pro (More info?)

"MrCatnip" <mrcatnip@hotmail.com> wrote in message
news:497ni059h8a12f0vv062bs0g7u4pnlgpt6@4ax.com...
> On Wed, 25 Aug 2004 01:20:36 +1000, "TonyP" <TonyP@optus.net.com.au>
> wrote:
> My humble appollogies, I did misplace my reply, I meant it for the
> statement just below here and it is your statement. So I will repeat
> it in the proper place so you might be able to figure it out.
> >
> >> >Why would anyone DOWN sample FIRST ??????
>
> I wasn't suggesting that you should, but since you asked I suspect
> you
> are not sure yourself. Let me assure you it is not a good idea. Do
> you know why it is not? Probably you don't or you would not have
> have
> made the other useless statements. I suspect you are a technician and
> go by experience rather than knowledge since you simply echoed what
> others have said without seeming to understand

In that case you are quite wrong. Please spell out where I have
misunderstood anything other than your poor attempts at justifying yourself?

TonyP.
Anonymous
August 26, 2004 4:48:13 AM

Archived from groups: rec.audio.pro (More info?)

On Wed, 25 Aug 2004 12:42:38 +0200, Chel van Gennip
<chel@vangennip.nl> wrote:

>On Wed, 25 Aug 2004 11:48:00 +0200, Arny Krueger wrote:
>
>> "Chel van Gennip" <chel@vangennip.nl> wrote in message
>> news:2p33b8FfpusmU1@uni-berlin.de
>> IOW, when the market has had one highly-popular *sonic overkill* format
>> called CD audio for over 20 years, and it is in fact the marketplace's
>> benchmark for audio quality, and it is sold every street corner and
>> being reproduced by consumers in many of the homes between the street
>> corners, there is just no widespread perceived need for addtional *sonic
>> overkill* formats.
>
>Agreed. There is indeed a lot of overkill in technological fuzz in the
>market. Dynamics of sound is better served bij leaving the compression of
>dynamic range out of the production cycle than by more bits of resolution
>in the mediumformat. I think 96 dB is more than sufficient for sound
>reproduction. In the sound production process a bit more resolution does
>simplify the process. If you want to compress the dynamic range in the end
>result and want to have a 96 dB dynamic range in the end result, you need
>more dynamic range in the production process.
>
>>>it is clear that
>>> your recordings, if they are important, wil have to be converted to
>>> 44.1, 48, and maybe 96khz now or in the future, because that are the
>>> frequencies used in current media standards.
>>
>> In fact the so-called high resolution formats are failing in the
>> marketplace. SACD was slated for cancellation earlier this year, but
>> somehow it got a reprive. How long will they last?
>
>The DVD audio channel normally is 48khz or a multitude, I think that will
>last.
>
>>> Therefore 96khz is a better starting point. Conversion from 44.1 to 48
>>> khz, and visa
>>> versa, is a problem.
>>
>> This is a false claim, technologically speaking. Modern upsampling
>> algorithms do not require or even show a discernable benefit from claims
>> based on the abstract study of numerology.
>>
>>
>>> Conversion from 96khz to either 44.1 or 48 khz, does not impose serious
>>> problems, if you are downsampling with a factor 2 or higher the most
>>> important factor influencing the result is the target frequency.
>>
>> Yet another false claim based on what appears to be yet another
>> intuitive study of numerology. In fact the most delicate processing step
>> in downsampling is the digital filtering, not the decimation.
>
>Digital filtering too is less complex and more accurate at 1/4 of the
>samplerate. (As most other processing done on the tracks done during the
>production process) If all has been done well, the ultimate limitation is
>the samplerate of the final result.
>
>>>Again: The introduced problems, higher bandwidth, storage and
>>>processing load, have been overcome by
>>> hardware developments.
>>
>> Now that I agree with. However in practical use downsampling from 96 KHz
>> can be a major time-waster. Given that high sample rates have no
>> practical audible benefit, its just time wasted.
>
>It cost some time, but it is not in the interactive part, it is part of a
>batch process. BTW, a good MP3 conversion, that is part of the same batch
>run for me, takes a lot more time.
>--

You might as well give it up, particularly with these two. Arny is
mired in what he knew yesterday and Tony is a "me too" guy. These are
the same kind that swore that cassettes will kill the recording
industry and automobiles will never replace the horse

None of us ever talked about upsampling old recordings, of course
there is no real improvement, some are even worse. SACD is failing
because there is no push behind it and there are better systems coming
out. Color TV took forever to catch on, it only made it because the
industry backed it fully. If you don't believe me about the
importance of industry support, look at BETA vs VHS. BETA is clearly
a superior format, but VHS got most of the industry support.
>Chel van Gennip
>Visit Serg van Gennip's site http://www.serg.vangennip.com
Anonymous
August 26, 2004 5:01:34 AM

Archived from groups: rec.audio.pro (More info?)

"MrCatnip" <mrcatnip@hotmail.com> wrote in message
news:leani0p5qao48u4nsh457gh1l6os0rl69g@4ax.com...
> My statement that it could ONLY be drawn as a Triangle was too narrow
> a statement. It could be drawn as anything your heart desires.

Keep digging yourself deeper! The newsgroup is rec.audio.pro not
rec.modern.art.

TonyP.
Anonymous
August 26, 2004 5:15:18 AM

Archived from groups: rec.audio.pro (More info?)

"Chel van Gennip" <chel@vangennip.nl> wrote in message
news:2p3amqFg6ldhU1@uni-berlin.de...
> If you want to compress the dynamic range in the end
> result and want to have a 96 dB dynamic range in the end result, you need
> more dynamic range in the production process.

True, but no one really wants 96 dB dynamic range in the final product
anyway. I think a bit more than the currently fashionable 10 dB would be
nice, but 96 dB is unusable to the vast majority of people in the vast
majority of listening environments. Certainly at any normal listening
levels.
I'm all for overkill in the production process though, now that it costs so
little.


TonyP.
Anonymous
August 26, 2004 6:28:38 AM

Archived from groups: rec.audio.pro (More info?)

"Scott Dorsey" <kludge@panix.com> wrote in message
news:cgib4d$ptf$1@panix2.panix.com...
> I thought you were arguing that wide bandwidth was tied to accuracy?
> My analogue tape machine has a -3 dB point at 35 KHz. Certainly wider
> bandwidth than CDs have.

Not necessarily so! Bandwidth includes *BOTH* HF and LF. You tape machine
has nearly an extra octave on top, whilst CD has a couple of octaves or more
at the bottom end.

But less than the vast majority of 96 kHz digital recorders in any case.

TonyP.
Anonymous
August 26, 2004 6:28:39 AM

Archived from groups: rec.audio.pro (More info?)

"TonyP" <TonyP@optus.net.com.au> wrote in message
news:412cbe60$0$2481$afc38c87@news.optusnet.com.au
> "Scott Dorsey" <kludge@panix.com> wrote in message
> news:cgib4d$ptf$1@panix2.panix.com...

>> I thought you were arguing that wide bandwidth was tied to accuracy?
>> My analogue tape machine has a -3 dB point at 35 KHz. Certainly
>> wider bandwidth than CDs have.

> Not necessarily so! Bandwidth includes *BOTH* HF and LF. You tape
> machine has nearly an extra octave on top, whilst CD has a couple of
> octaves or more at the bottom end.

That gets to be a matter of tolerances. Due to head bumps, response below 50
Hz, particularly at the higher speeds like 15 or 30 ips gets really iffy.
OTOH, the lower frequency limit of a digital recording is the inverse of its
length. For example, a 180 second recording can have response down to
approximately 0.005 Hz

> But less than the vast majority of 96 kHz digital recorders in any case.

Now that 96 KHz recording is commonly availble, people can experiment with
the effects of high sample rate on their own. "Starter" recordings for doing
this can be found at http://www.pcabx.com/technical/sample_rates/index.htm
Anonymous
August 26, 2004 6:28:40 AM

Archived from groups: rec.audio.pro (More info?)

Arny Krueger <arnyk@hotpop.com> wrote:
>"TonyP" <TonyP@optus.net.com.au> wrote in message
>news:412cbe60$0$2481$afc38c87@news.optusnet.com.au
>> "Scott Dorsey" <kludge@panix.com> wrote in message
>> news:cgib4d$ptf$1@panix2.panix.com...
>
>>> I thought you were arguing that wide bandwidth was tied to accuracy?
>>> My analogue tape machine has a -3 dB point at 35 KHz. Certainly
>>> wider bandwidth than CDs have.
>
>> Not necessarily so! Bandwidth includes *BOTH* HF and LF. You tape
>> machine has nearly an extra octave on top, whilst CD has a couple of
>> octaves or more at the bottom end.
>
>That gets to be a matter of tolerances. Due to head bumps, response below 50
>Hz, particularly at the higher speeds like 15 or 30 ips gets really iffy.
>OTOH, the lower frequency limit of a digital recording is the inverse of its
>length. For example, a 180 second recording can have response down to
>approximately 0.005 Hz

No, the Ampex is very flat down to 20 Hz. But it's true that below that
it's really bizarre... the sub-bass response is extremely spiky. It does
drop below the head bump, but it also gets very jagged as it drops. The
Saki heads are definitely worse about this than the Flux Magnetics heads.

>> But less than the vast majority of 96 kHz digital recorders in any case.
>
>Now that 96 KHz recording is commonly availble, people can experiment with
>the effects of high sample rate on their own. "Starter" recordings for doing
>this can be found at http://www.pcabx.com/technical/sample_rates/index.htm

BUT, to make it a fair comparison, you really need for the rest of your
chain to be wideband as well.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Anonymous
August 26, 2004 6:28:41 AM

Archived from groups: rec.audio.pro (More info?)

"Scott Dorsey" <kludge@panix.com> wrote in message
news:cgiqjf$i9h$1@panix2.panix.com

> Arny Krueger <arnyk@hotpop.com> wrote:

>> Now that 96 KHz recording is commonly availble, people can
>> experiment with the effects of high sample rate on their own.
>> "Starter" recordings for doing this can be found at
>> http://www.pcabx.com/technical/sample_rates/index.htm
>
> BUT, to make it a fair comparison, you really need for the rest of
> your chain to be wideband as well.

Agreed. Which gets us to the next relevant point - most microphones and
monitor speakers in professional use, even equipment bought in the last
year, doesn't have anything like flat response past 20 KHz. Many fine
monitors get to 15 KHz and take a big dive. The most common microphones in
professional use start taking a dive in the 11-13 Khz range, if not lower.
!