[Tobiah]

Is it generally regarded as a better idea to record CD
source material at 88.2k rather than 96k becasue
88.2k is a multiple of 44100?

Thanks,

Tobiah

--
Posted via a free Usenet account from http://www.teranews.com

[David Morgan \(MAMS\)]

"Tobiah" wrote in message .. .
Is it generally regarded as a better idea to record CD
source material at 88.2k rather than 96k becasue
88.2k is a multiple of 44100?


Yes.




--
David Morgan (MAMS)
http://www.m-a-m-s DOT com
Morgan Audio Media Service
Dallas, Texas (214) 662-9901
_______________________________________
http://www.januarysound.com

[Mike Rivers]

On Oct 26, 4:58 pm, Tobiah wrote:
Is it generally regarded as a better idea to record CD
source material at 88.2k rather than 96k becasue
88.2k is a multiple of 44100?

It doesn't matter either way as long as you have a good sample rate
conversion program, and if you have a sloppy sample rate conversion
program, it won't work any better at one rate of the other.

Contrary to intuition, you actually have to recalculate every sample
even when converting by an integer ratio. Simply dropping every other
sample doesn't work very well. Nor is it any easier to do the math if
the ratio is an integer.

[Arny Krueger]

"Tobiah" wrote in message
.. .

Is it generally regarded as a better idea to record CD
source material at 88.2k rather than 96k becasue
88.2k is a multiple of 44100?

Explain why its a good idea to record so much data that you throw away in
the final product.

[David Morgan \(MAMS\)]

"Arny Krueger" wrote in message . ..

"Tobiah" wrote in message
.. .

Is it generally regarded as a better idea to record CD
source material at 88.2k rather than 96k becasue
88.2k is a multiple of 44100?

Explain why its a good idea to record so much data that you throw away in
the final product.

I wasn't going to go there... but the likely answer is, "Because I can."

DM

[Laurence Payne]

On Sat, 27 Oct 2007 00:27:42 GMT, "David Morgan \(MAMS\)"
/Odm wrote:

Explain why its a good idea to record so much data that you throw away in
the final product.

I wasn't going to go there... but the likely answer is, "Because I can."

There's some evidence that if you're going to submit your tracks to
digital effects, particularly those such as pitch-shifting where you
walk a tightrope between achieving the desired result and losing
quality, the higher sample rate gives better results. But make sure
your plug-ins support the higher rate :-)

If you're going to output to CD and only perform basic editing, it's
difficult to explain how the higher rate would be an advantage.
Recording into 24 bits relieves you from having to fuss over levels
quite so much. But all you're going to do with those extra Hz is
throw them away.

[William Sommerwerck]

"Arny Krueger" wrote in message
. ..

Explain why its a good idea to record so much data that you
throw away in the final product.

For one thing, it permits much less-abrupt filtration during recording --
though this advantage is partly lost when downsampling.

[Arny Krueger]

"William Sommerwerck" wrote in message
. ..
"Arny Krueger" wrote in message
. ..

Explain why its a good idea to record so much data that you
throw away in the final product.

For one thing, it permits much less-abrupt filtration during recording --
though this advantage is partly lost when downsampling.

No, its all lost.

[Laurence Payne]

On Fri, 26 Oct 2007 18:09:32 -0700, "William Sommerwerck"
wrote:

Explain why its a good idea to record so much data that you
throw away in the final product.

For one thing, it permits much less-abrupt filtration during recording --
though this advantage is partly lost when downsampling.

Explain "partly" please?

[Arny Krueger]

"Laurence Payne" NOSPAMlpayne1ATdsl.pipex.com wrote in message
...
On Fri, 26 Oct 2007 18:09:32 -0700, "William Sommerwerck"
wrote:

Explain why its a good idea to record so much data that you
throw away in the final product.

For one thing, it permits much less-abrupt filtration during recording --
though this advantage is partly lost when downsampling.

Explain "partly" please?

Yeah, really. To me, 90-100 dB down is not "partly lost". It's totally
gone for almost all practical purposes!

[Don Pearce]

On Fri, 26 Oct 2007 18:09:32 -0700, "William Sommerwerck"
wrote:

"Arny Krueger" wrote in message
...

Explain why its a good idea to record so much data that you
throw away in the final product.

For one thing, it permits much less-abrupt filtration during recording --
though this advantage is partly lost when downsampling.


If you make a recording at 44.1kHz, then in all probability your ADC
is sampling at 2.8MHz (64 times oversampling is common). That is
followed by a very steep digital filter that passes 20kHz, and rejects
completely (more or less) 24.1 kHz - the alias frequency. This high
frequency sample stream is then converted (by simply ignoring most of
the words) into 44.1kHz data.

All of that happens entirely beyond your control in the front end of
every audio ADC, so the idea that you can avoid steep filtering during
later processing is entirely false. It has happened, and there is
nothing you can do about it.

Making a recording at 82 or 96kHz and downsampling again later simply
means that this process has had to happen twice instead of once. That
cannot possibly make things better, and if implemented poorly will
certainly make them worse.

d

--
Pearce Consulting
http://www.pearce.uk.com

[anahata]

Arny Krueger wrote:
"Tobiah" wrote in message
.. .


Is it generally regarded as a better idea to record CD
source material at 88.2k rather than 96k becasue
88.2k is a multiple of 44100?

Explain why its a good idea to record so much data that you throw away in
the final product.

That's a diffent issue from 88.2k vs. 96k.

But a possible reason for higher sampling rates is that if your digital
processing should ever run to clipping you'll get fewer in-band aliasing
artefacts.
(The obvious reply to that is that you shouldn't let it clip)

--
Anahata
-+- http://www.treewind.co.uk
Home: 01638 720444 Mob: 07976 263827

[Tobiah]

Arny Krueger wrote:
"Tobiah" wrote in message
.. .

Is it generally regarded as a better idea to record CD
source material at 88.2k rather than 96k becasue
88.2k is a multiple of 44100?

Explain why its a good idea to record so much data that you throw away in
the final product.


If I were to shoot a picture with a digital camera for use on a web
page at 100x100 pixels, I would still shoot it at the maximum resolution
of which the camera was capable; there may be a more demanding use for
the image at a later time.

Mike Rivers had the most lucid response. I think that I will choose
96k, because it's enough, standard, and a reasonable compromise between
resources and quality.

--
Posted via a free Usenet account from http://www.teranews.com

[Laurence Payne]

On Sat, 27 Oct 2007 09:03:00 -0400, Tobiah wrote:

Is it generally regarded as a better idea to record CD
source material at 88.2k rather than 96k becasue
88.2k is a multiple of 44100?

Explain why its a good idea to record so much data that you throw away in
the final product.


If I were to shoot a picture with a digital camera for use on a web
page at 100x100 pixels, I would still shoot it at the maximum resolution
of which the camera was capable; there may be a more demanding use for
the image at a later time.

Mike Rivers had the most lucid response. I think that I will choose
96k, because it's enough, standard, and a reasonable compromise between
resources and quality.

Actually, you're probably recording at the high rate because you have
a gut feeling that it's somehow "better" and you only want to hear
facts and opinions that support that. But no matter. If you have a
soundcard that sounds good at the high rates, and adequate processing
power and storage, it'll do you no harm :-)

[Tobiah]

Laurence Payne wrote:
On Sat, 27 Oct 2007 09:03:00 -0400, Tobiah wrote:

Is it generally regarded as a better idea to record CD
source material at 88.2k rather than 96k becasue
88.2k is a multiple of 44100?
Explain why its a good idea to record so much data that you throw away in
the final product.

If I were to shoot a picture with a digital camera for use on a web
page at 100x100 pixels, I would still shoot it at the maximum resolution
of which the camera was capable; there may be a more demanding use for
the image at a later time.

Mike Rivers had the most lucid response. I think that I will choose
96k, because it's enough, standard, and a reasonable compromise between
resources and quality.

Actually, you're probably recording at the high rate because you have
a gut feeling that it's somehow "better" and you only want to hear
facts and opinions that support that. But no matter. If you have a
soundcard that sounds good at the high rates, and adequate processing
power and storage, it'll do you no harm :-)

I can pick out 24/96k over CD audio in a blind taste test with some
material. I would have to test further to find out whether it is
the resolution or rate that is creating the difference. I know that
I want the resolution, because I generate audio with great dynamic
range. As far as the sample rate goes, I suppose there is some
'more is better' mentality working there.

--
Posted via a free Usenet account from http://www.teranews.com

[Laurence Payne]

On Sat, 27 Oct 2007 09:33:42 -0400, Tobiah wrote:

I can pick out 24/96k over CD audio in a blind taste test with some
material. I would have to test further to find out whether it is
the resolution or rate that is creating the difference. I know that
I want the resolution, because I generate audio with great dynamic
range. As far as the sample rate goes, I suppose there is some
'more is better' mentality working there.

I very much doubt you've got a source or a playback system that can
provide 16 bits of resolution, let alone 24. Work out what the
voltage swing would need to be if you don't believe me :-) So it's
not the bit-depth as such that you're hearing. You might be hearing
differences in your soundcard's perfomance at the different settings.

[Arny Krueger]

"Tobiah" wrote in message
.. .

I can pick out 24/96k over CD audio in a blind taste test with some
material.

Then you're the first to do that!

Of course there are ways to trick the experiment up. For example, if you
attenuate your 24 bit music by 60 dB, and then reduce it to 16 bits, and
then amplify it by 60 dB, then you might hear a difference.

Or, if your blind test was not precisely level-matched and time-synched.

Do it right, and you'll get the same results everybody else gets when they
do it right. ;-)

[Mike Rivers]

On Oct 27, 9:33 am, Tobiah wrote:

I can pick out 24/96k over CD audio in a blind taste test with some
material. I would have to test further to find out whether it is
the resolution or rate that is creating the difference.

It's almost certainly the resolution.

I did attend a demonstration where I was able to hear a difference
between 24-bit 48 kHz and 96 kHz, but:

- It was sponsored by (among others) Sennheiser, who was showing us
why we should buy their new MKH-800 mic which has usable frequency
response above 30 kHz (so we can justify recording at higher sample
rates)

- The playback equipment and monitoring environment were very high
quality - better than what I'd have in my own studio, even if I bought
some MKH-800 mics.

- The signal path was mic - preamp - A/D converter - D/A converter -
monitor system. Even though the same converter pair was switched
between 48 and 96 kHz, nobody could assure that nothing other than the
high frequency response was different at the two sample rates.

24/96 is sort of the current high resolution standard, so you might as
well use it if you have the disk space and as long as your converters
actually sound better at 96 kHz than at 48. Most new ones do, many
older ones, even good ones, don't. At least that way you can preserve
the original mic preamp output as accurately as is reasonable. There
are DSD converters now that can do a little better, but you ALWAYS
have to convert that.

[anahata]

Tobiah wrote:
I know that
I want the resolution, because I generate audio with great dynamic
range. As far as the sample rate goes, I suppose there is some
'more is better' mentality working there.

There are very good practical reasons for using high resolution in the
recording and production process. However, I think it's agreed that 16
bits is enough for the end product: higher resolution simply means
lowering the level of noise that is already inaudible.

On sample rates: it's debatable whether more is better when more only
means more signals at frequencies you can't hear. Those signals need to
be handled accurately by equipment like power amplifiers to prevent
intermodulation products that could appear in the audio band, so their
presence could be a disadvantage.

--
Anahata
-+- http://www.treewind.co.uk
Home: 01638 720444 Mob: 07976 263827

[Arny Krueger]

"Tobiah" wrote in message
.. .
Arny Krueger wrote:
"Tobiah" wrote in message
.. .

Is it generally regarded as a better idea to record CD
source material at 88.2k rather than 96k becasue
88.2k is a multiple of 44100?

Explain why its a good idea to record so much data that you throw away in
the final product.

If I were to shoot a picture with a digital camera for use on a web
page at 100x100 pixels, I would still shoot it at the maximum resolution
of which the camera was capable; there may be a more demanding use for
the image at a later time.

So would I. However photography isn't identically the same as sound.

I can reliably see the difference between the max possible resolution of say
a Canon Digital Rebel, and a picture with say half the resolution, as
enlarged a reasonable amount.

But, no way can anybody hear the difference between your recording when
sampled at 24/96 and 24/48. If you think I'm pulling your leg, then do your
own listening test, just make sure your test is blind, level-matched, and
time-synched. The step-by-step instructions for do this with zero
out-of-pocket expense have been posted here many times, and I'd do it again
if I thought it would make a difference.