"The Incredibles" orchestra recorded analog? Does it make adifference?

[muzician21]

Someone posted part of a documentary lifted from the "bonus material"
from The Incredibles on YouTube. They make a big point of talking
about how the orchestra was recorded analog because they wanted "that
analog 60's sound" or something to that effect. They show a bank of
sync'd reel to reel machines.

Assuming one accepts that there's a profound difference in the sound
of an analog recorded orchestra, doesn't the fact that at some point
it has to be transferred to the digital realm make it a moot point?
It goes to digital for the DVD and any soundtrack album. I'm not
versed in sound for film but I assume these days by the time it makes
it to your local theater it's gone through a digital process - or is
that not correct?

[Laurence Payne[_2_]]

On Mon, 7 Sep 2009 14:57:35 -0700 (PDT), muzician21
> wrote:

>Someone posted part of a documentary lifted from the "bonus material"
>from The Incredibles on YouTube. They make a big point of talking
>about how the orchestra was recorded analog because they wanted "that
>analog 60's sound" or something to that effect. They show a bank of
>sync'd reel to reel machines.
>
>Assuming one accepts that there's a profound difference in the sound
>of an analog recorded orchestra, doesn't the fact that at some point
>it has to be transferred to the digital realm make it a moot point?
>It goes to digital for the DVD and any soundtrack album. I'm not
>versed in sound for film but I assume these days by the time it makes
>it to your local theater it's gone through a digital process - or is
>that not correct?

Tape has a "sound", digital doesn't. If you like the tape sound,
digital will capture and preserve it.

But they're wanking. There's other ways of getting tape-style
distortion if you really want it.

[Sean Conolly]

"muzician21" > wrote in message
...
> Someone posted part of a documentary lifted from the "bonus material"
> from The Incredibles on YouTube. They make a big point of talking
> about how the orchestra was recorded analog because they wanted "that
> analog 60's sound" or something to that effect. They show a bank of
> sync'd reel to reel machines.
>
> Assuming one accepts that there's a profound difference in the sound
> of an analog recorded orchestra, doesn't the fact that at some point
> it has to be transferred to the digital realm make it a moot point?
> It goes to digital for the DVD and any soundtrack album. I'm not
> versed in sound for film but I assume these days by the time it makes
> it to your local theater it's gone through a digital process - or is
> that not correct?

The direct answer is yes - it was transferred to digital, probably as the
next step after recording.

As for getting a 'tape' sound, I think whatever subtle differences that tape
provided were long gone by the time it reached the audience's ears.

JMHO,
Sean

[William Sommerwerck]

It depends on the equipment being used, which includes the mixing boards and
mics. If one listens to classic Living Stereo recordings (for example), part
of that sound comes from the "other" electronics in the chain.

I don't think that just using analog tape decks is going to produce the full
"retro" sound you want, and if someone thought it was enough, they were
probably wasting their time.

[Scott Dorsey]

muzician21 > wrote:
>Someone posted part of a documentary lifted from the "bonus material"
>from The Incredibles on YouTube. They make a big point of talking
>about how the orchestra was recorded analog because they wanted "that
>analog 60's sound" or something to that effect. They show a bank of
>sync'd reel to reel machines.

That's fine, although I'm not sure why you'd need to sync multiple
machines for that. Even spotmiking the crap out of everything, you
would be hardpressed to break 16 tracks on a film orchestra gig.

>Assuming one accepts that there's a profound difference in the sound
>of an analog recorded orchestra, doesn't the fact that at some point
>it has to be transferred to the digital realm make it a moot point?

No. These days the digital systems can be pretty transparent if you
are willing to pay the money for that transparency. Analogue systems
can be pretty damn transparent too, if you want them to be. If you
don't, you can get a variety of interesting and useful colorations
from them. A good digital dub should preserve those.

>It goes to digital for the DVD and any soundtrack album. I'm not
>versed in sound for film but I assume these days by the time it makes
>it to your local theater it's gone through a digital process - or is
>that not correct?

Sadly, the majority of theatres today are still using analogue
optical soundtracks made with 1930s vintage Western Electric light
valve cameras. There are a lot of theatres with digital sound systems,
but they all fall back to the analogue track.

For the most part, the analogue film soundtrack is a rube goldberg
system; it's remarkable that it works as well as it does, and at 20KC
it doesn't work so well. Most of the reason that it has been kept in
spite of 80 years of improvements is that it's very cheap.
--scott

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

[Scott Dorsey]

Laurence Payne > wrote:
>
>Tape has a "sound", digital doesn't. If you like the tape sound,
>digital will capture and preserve it.

Tape has a lot of different sounds, and you have control over it.

>But they're wanking. There's other ways of getting tape-style
>distortion if you really want it.

I remain extremely skeptical on this point.
--scott

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

[Scott Dorsey]

William Sommerwerck > wrote:
>It depends on the equipment being used, which includes the mixing boards and
>mics. If one listens to classic Living Stereo recordings (for example), part
>of that sound comes from the "other" electronics in the chain.
>
>I don't think that just using analog tape decks is going to produce the full
>"retro" sound you want, and if someone thought it was enough, they were
>probably wasting their time.

Listening to the CD, they sure weren't wasting their time. And my guess
is that 90% of that sound comes from mike technique and performance, and
the folks they got doing the job were just spot-on. It's a great sounding
CD.

No, it doesn't sound realistic, it sounds like a film score.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[William Sommerwerck]

"Scott Dorsey" > wrote in message
...
> William Sommerwerck > wrote:

>> It depends on the equipment being used, which includes the mixing
>> boards and mics. If one listens to classic Living Stereo recordings
>> (for example), part of that sound comes from the "other" electronics
>> in the chain.

>> I don't think that using just analog tape decks is going to produce
>> the full "retro" sound you want, and if someone thought it was enough,
>> they were probably wasting their time.

> Listening to the CD, they sure weren't wasting their time. And my guess
> is that 90% of that sound comes from mike technique and performance,
> and the folks they got doing the job were just spot-on. It's a
great-sounding
> CD.

> No, it doesn't sound realistic, it sounds like a film score.

But that's true of digital recordings, as well. Film scores are heavily
multi-miked.

[William Sommerwerck]

> Sadly, the majority of theatres today are still using analog
> optical soundtracks made with 1930s vintage Western Electric
> light valve cameras. There are a lot of theatres with digital sound
> systems, but they all fall back to the analogue track.

Are they? Haven't most theaters switched to a digital disk locked to the
film?


> For the most part, the analogue film soundtrack is a Rube Goldberg
> system; it's remarkable that it works as well as it does, and at 20KC
> it doesn't work so well. Most of the reason that it has been kept in
> spite of 80 years of improvements is that it's very cheap.

I don't see where the system described above would be particularly
expensive.

By the way, you're not using "Rube Goldberg" in its correct sense. It means
an overly complex way of doing something relatively simple.

[Richard Crowley]

"Scott Dorsey" wrote ...
> muzician21 wrote:
>>Someone posted part of a documentary lifted from the "bonus material"
>>from The Incredibles on YouTube. They make a big point of talking
>>about how the orchestra was recorded analog because they wanted "that
>>analog 60's sound" or something to that effect. They show a bank of
>>sync'd reel to reel machines.
>
> That's fine, although I'm not sure why you'd need to sync multiple
> machines for that. Even spotmiking the crap out of everything, you
> would be hardpressed to break 16 tracks on a film orchestra gig.

Maybe they showed a bank of sync'd mag film recorders.
That would acount for needing several. :-)

> For the most part, the analogue film soundtrack is a rube goldberg
> system; it's remarkable that it works as well as it does, and at 20KC
> it doesn't work so well. Most of the reason that it has been kept in
> spite of 80 years of improvements is that it's very cheap.

And simple, and reliable, and degrades gracefully. And can
be repaired by the local TV repair guy in a pinch.

Few of which you can say for any digital system. Not that I am
defending them, mind you. I'm glad I don't have to use or maintain
any cine equipment.

[Al, Cambridge, UK]

On Sep 7, 10:57*pm, muzician21 > wrote:
> Someone posted part of a documentary lifted from the "bonus material"
> from The Incredibles on YouTube. They make a big point of talking
> about how the orchestra was recorded analog because they wanted "that
> analog 60's sound" or something to that effect. They show a bank of
> sync'd reel to reel machines.
>
> Assuming one accepts that there's a profound difference in the sound
> of an analog recorded orchestra, doesn't the fact that at some point
> it has to be transferred to the digital realm make it a moot point?
> It goes to digital for the DVD and any soundtrack album. I'm not
> versed in sound for film but I assume these days by the time it makes
> it to your local theater it's gone through a digital process - or is
> that not correct?

Perhaps if the *musicians* know they're recording to analog, they
might feel different about the session as they watch the tapes go
round and play subtly differently?
Al

[Scott Dorsey]

William Sommerwerck > wrote:
>"Scott Dorsey" > wrote in message
...
>> William Sommerwerck > wrote:
>
>>> It depends on the equipment being used, which includes the mixing
>>> boards and mics. If one listens to classic Living Stereo recordings
>>> (for example), part of that sound comes from the "other" electronics
>>> in the chain.
>
>>> I don't think that using just analog tape decks is going to produce
>>> the full "retro" sound you want, and if someone thought it was enough,
>>> they were probably wasting their time.
>
>> Listening to the CD, they sure weren't wasting their time. And my guess
>> is that 90% of that sound comes from mike technique and performance,
>> and the folks they got doing the job were just spot-on. It's a
>great-sounding
>> CD.
>
>> No, it doesn't sound realistic, it sounds like a film score.
>
>But that's true of digital recordings, as well. Film scores are heavily
>multi-miked.

Absolutely, but these days they tend to be multi-miked with somewhat
different technique than you'd hear on fifties and sixties soundtracks.
Some of that is because today we tend to double parts a lot in order to
keep costs down.

I think the miking has more to do with the classic sound of that album
than the recording technology does. And the performance and arrangement
has more to do with it than the miking. People just don't _do_ that kind
of arrangement any more.
--scott

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

[Scott Dorsey]

William Sommerwerck > wrote:
>> Sadly, the majority of theatres today are still using analog
>> optical soundtracks made with 1930s vintage Western Electric
>> light valve cameras. There are a lot of theatres with digital sound
>> systems, but they all fall back to the analogue track.
>
>Are they? Haven't most theaters switched to a digital disk locked to the
>film?

That's the DTS system.

A lot of theatres today use DTS, which uses a CD-ROM that contains compressed
audio data, and a timecode on the film. And a lot of distributors lose or
forget to send the DTS discs along with the film too.

A lot of other theatres used the Dolby Digital system, which has compressed
digital data stored as a bunch of little dots in-between the perforations
of the film. DD doesn't require additional media, but the error rate is
very high so it tends to fall back to the analogue track very often.

But if you go into your local multiplex, you'll find a couple halls
equiped with some digital format, and a whole lot more of them with
analogue sound. In fact, you'll find a remarkable number of them today
running mono.

>> For the most part, the analogue film soundtrack is a Rube Goldberg
>> system; it's remarkable that it works as well as it does, and at 20KC
>> it doesn't work so well. Most of the reason that it has been kept in
>> spite of 80 years of improvements is that it's very cheap.
>
>I don't see where the system described above would be particularly
>expensive.

The digital systems are expensive to install and expensive to operate, but
they are a huge improvement in that regard over the magnetic soundtrack
systems of the seventies. Commag just sounded great, but the prints were
easily damaged by magnetic tools and there was a substantial additional cost
in striking the prints. Also distributors had to keep two sets of prints
around.

The one big demand with the digital formats was that it NOT require separate
sets of prints, and that has helped a lot.

>By the way, you're not using "Rube Goldberg" in its correct sense. It means
>an overly complex way of doing something relatively simple.

Compare optical sound with a magnetic track. It's very, very complex, and
it doesn't really sound all that great but it's what we got.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[William Sommerwerck]

>> By the way, you're not using "Rube Goldberg" in its correct sense.
>> It means an overly complex way of doing something relatively simple.

> Compare optical sound with a magnetic track. It's very, very complex,
> and it doesn't really sound all that great but it's what we got.


Yes, but... Magnetic recording didn't exist in the 1920s. And magnetic
tracks are not as physically robust as optical tracks (analog or digital).

I don't see variable-width recording as being particularly complex, as the
complex part of it is at the originating end (where "complex parts" always
should be).

Variable-density optical is pretty simple (compared to variable-width), but
it has linearity problems.

[Scott Dorsey]

William Sommerwerck > wrote:
>Yes, but... Magnetic recording didn't exist in the 1920s. And magnetic
>tracks are not as physically robust as optical tracks (analog or digital).

Right, but we have had plenty of opportunities to replace optical tracks
over the past 90 years, but they still keep going on and on.

>I don't see variable-width recording as being particularly complex, as the
>complex part of it is at the originating end (where "complex parts" always
>should be).

That's true. But there is a lot of very ugly stuff at the originating end.

>Variable-density optical is pretty simple (compared to variable-width), but
>it has linearity problems.

That's true, but don't forget that variable-area systems have some pretty
bad linearity issues themselves. I'll be happy if I never see another
cross-modulation test ever again.

The playback side is interesting, and with the recent introduction of dye
soundtracks in place of silver soundtracks, there have been some changes
in soundheads. Some of these changes introduce some other nonlinearity issues
that either aren't very well understood or aren't being talked about by the
people who do understand them. Look on any of the film forums for a
discussion on "reverse scan" soundheads. But that's a whole other kettle of
fish.

The dye tracks are coming in, once again, because they are cheaper to make.
Labs don't want to bother going through the process of seperate soundtrack
development, which has been required since color processes came in.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[William Sommerwerck]

>> Yes, but... Magnetic recording didn't exist in the 1920s. And magnetic
>> tracks are not as physically robust as optical tracks (analog or
digital).

> Right, but we have had plenty of opportunities to replace optical tracks
> over the past 90 years, but they still keep going on and on.

Other than a separate optical disk, what _would_ you replace them with?


> That's true, but don't forget that variable-area systems have some pretty
> bad linearity issues themselves. I'll be happy if I never see another
> cross-modulation test ever again.

It might be my imagination, but I believe the soundtracks for a lot of
classic Disney DVDs have been derived from the original stems. They
definitely seen cleaner. A single generation of optical sound isn't
necessarily "awful".


> The playback side is interesting, and with the recent introduction of dye
> soundtracks in place of silver soundtracks, there have been some changes
> in soundheads. Some of these changes introduce some other nonlinearity
> issues that either aren't very well understood or aren't being talked
about
> by the people who do understand them. Look on any of the film forums for a
> discussion on "reverse scan" soundheads. But that's a whole other kettle
of
> fish.

> The dye tracks are coming in, once again, because they are cheaper to
make.
> Labs don't want to bother going through the process of seperate soundtrack
> development, which has been required since color processes came in.

Some years ago I noticed that two-strip Technicolor films (eg, "Wax Museum")
had cleaner, more "ballsy" sound than silver prints. I wonder if there's any
connection.

[nebulax]

On Sep 7, 5:57*pm, muzician21 > wrote:
> Someone posted part of a documentary lifted from the "bonus material"
> from The Incredibles on YouTube. They make a big point of talking
> about how the orchestra was recorded analog because they wanted "that
> analog 60's sound" or something to that effect. They show a bank of
> sync'd reel to reel machines.
>
> Assuming one accepts that there's a profound difference in the sound
> of an analog recorded orchestra, doesn't the fact that at some point
> it has to be transferred to the digital realm make it a moot point?
> It goes to digital for the DVD and any soundtrack album. I'm not
> versed in sound for film but I assume these days by the time it makes
> it to your local theater it's gone through a digital process - or is
> that not correct?


Music recorded in the analog domain can retain most of it's character
even after being transferred to a digital format. And there definitely
is an "analog sound" (in lots of flavors), because plug-in makers are
always trying to come up with new ways to emulate it. I think the
makers of the film were inspired by a lot of 50's and 60's era sci-fi
and cartoons, so it totally makes sense that they would want the sound
to be representative of that era, as well as the visuals.

-Neb

[Scott Dorsey]

William Sommerwerck > wrote:
>>> Yes, but... Magnetic recording didn't exist in the 1920s. And magnetic
>>> tracks are not as physically robust as optical tracks (analog or
>digital).
>
>> Right, but we have had plenty of opportunities to replace optical tracks
>> over the past 90 years, but they still keep going on and on.
>
>Other than a separate optical disk, what _would_ you replace them with?

Magstripes to begin with. Twentieth Century Fox was pushing it very big
starting in 1953, and it went over very well. For the most part, though,
mag only wound up being used for 70mm and for special venue releases once
the first burst of the Cinemascope craze died down.

It requires considerable additional cost to strike a print, and that's more
of a problem today than it was in 1953 because today we strike so many damn
prints on a single run. A film today premieres in thousands of cities at
the same time.

>> That's true, but don't forget that variable-area systems have some pretty
>> bad linearity issues themselves. I'll be happy if I never see another
>> cross-modulation test ever again.
>
>It might be my imagination, but I believe the soundtracks for a lot of
>classic Disney DVDs have been derived from the original stems. They
>definitely seen cleaner. A single generation of optical sound isn't
>necessarily "awful".

Anything post-1950 or so had the original stems being done on magfilm,
and dubbed to optical only for release. Hollywood went to magfilm for
original recording and editing and quickly as they possibly could, but
optical stuck around for release prints because it was cheap and rugged.

Fantasia is an example of a typical release today done from the optical
originals. I wouldn't call it "awful" but it sure isn't up to the quality
that was possible by 1953.

>Some years ago I noticed that two-strip Technicolor films (eg, "Wax Museum")
>had cleaner, more "ballsy" sound than silver prints. I wonder if there's any
>connection.

I'm not sure how the sound was done on two-strip. On the three-strip prints,
it started out with a reel of "receiver film" which had a B&W emulsion on it.
The soundtrack and frame lines would be printed on it using conventional
B&W exposure and processing and then the color images would be imbibition
printed onto that. So the soundtrack itself was a conventional silver
soundtrack like on any B&W or color print. However, the receiver film could
be developed for the best possible soundtrack quality without having to
worry about a continuous-tone image on the film as well, which may have helped.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[William Sommerwerck]

>> Other than a separate optical disk, what _would_ you replace them with?

> Magstripes to begin with. Twentieth Century Fox was pushing it very big
> starting in 1953, and it went over very well. For the most part, though,
> mag only wound up being used for 70mm and for special venue releases
> once the first burst of the CinemaScope craze died down.

In fact, the first CinemaScope films were _obliged_ to have stereo sound.

I heard stories -- from projectionists -- that the stripes wore very
quickly, and the heads had to be cleaned after each projection. Certainly
coatings have greatly improved, and you ought to be able to get excellent
sound at 2' per second, but still...


> Fantasia is an example of a typical release today done from the
> optical originals. I wouldn't call it "awful" but it sure isn't up to the
> quality that was possible by 1953.

Would that it were. The nitrate optical stems disintegrated somewhere around
1954.

It's worth noting that, although the current soundtrack is derived from the
magnetic transfer (made more than 50 years ago over wideband phone lines),
the CD and DVD sound rather different.

[Scott Dorsey]

William Sommerwerck > wrote:
>>> Other than a separate optical disk, what _would_ you replace them with?
>
>> Magstripes to begin with. Twentieth Century Fox was pushing it very big
>> starting in 1953, and it went over very well. For the most part, though,
>> mag only wound up being used for 70mm and for special venue releases
>> once the first burst of the CinemaScope craze died down.
>
>In fact, the first CinemaScope films were _obliged_ to have stereo sound.

Yes, that was part of the point of the system. The perforations were reduced
in size and a magstripe slightly wider than the normal optical track was
used.

>I heard stories -- from projectionists -- that the stripes wore very
>quickly, and the heads had to be cleaned after each projection. Certainly
>coatings have greatly improved, and you ought to be able to get excellent
>sound at 2' per second, but still...

For the first generation of the coating, this was the case. After a few
years it go to be pretty stable, but even today the stripe still wears out
before the print does, if the print is carefully kept. You send it back
to Technicolor Magnecraft for restriping after a couple thousand shows.

>> Fantasia is an example of a typical release today done from the
>> optical originals. I wouldn't call it "awful" but it sure isn't up to the
>> quality that was possible by 1953.
>
>Would that it were. The nitrate optical stems disintegrated somewhere around
>1954.

Yup. The thing is, the coloration of the optical track probably swamps
the coloration resulting from the succeeding mag generations.

>It's worth noting that, although the current soundtrack is derived from the
>magnetic transfer (made more than 50 years ago over wideband phone lines),
>the CD and DVD sound rather different.

Welcome to the modern mastering world where people just can't keep their
damn fingers off of things.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Laurence Payne[_2_]]

On Tue, 8 Sep 2009 10:44:18 -0700 (PDT), nebulax
> wrote:

>Music recorded in the analog domain can retain most of it's character
>even after being transferred to a digital format.

Why only "most"?

[nebulax]

On Sep 8, 3:29*pm, Laurence Payne > wrote:
> On Tue, 8 Sep 2009 10:44:18 -0700 (PDT), nebulax
>
> > wrote:
> >Music recorded in the analog domain can retain most of it's character
> >even after being transferred to a digital format.
>
> Why only "most"?


Back in the 16 bit era (oh wait... it's still with us), analog to
digital transfers tended to lose a lot of their ambiance, spatial
depth, high frequency detail, "warmth", soundstage width, etc etc.
This is probably not so true in the 24 bit (and above) digital domain,
but I don't own a player that can do those bit rates yet, so for me
that remains to be seen, or rather, heard.

-Neb

[Les Cargill[_2_]]

Scott Dorsey wrote:
> William Sommerwerck > wrote:
>> "Scott Dorsey" > wrote in message
>> ...
>>> William Sommerwerck > wrote:
>>>> It depends on the equipment being used, which includes the mixing
>>>> boards and mics. If one listens to classic Living Stereo recordings
>>>> (for example), part of that sound comes from the "other" electronics
>>>> in the chain.
>>>> I don't think that using just analog tape decks is going to produce
>>>> the full "retro" sound you want, and if someone thought it was enough,
>>>> they were probably wasting their time.
>>> Listening to the CD, they sure weren't wasting their time. And my guess
>>> is that 90% of that sound comes from mike technique and performance,
>>> and the folks they got doing the job were just spot-on. It's a
>> great-sounding
>>> CD.
>>> No, it doesn't sound realistic, it sounds like a film score.
>> But that's true of digital recordings, as well. Film scores are heavily
>> multi-miked.
>
> Absolutely, but these days they tend to be multi-miked with somewhat
> different technique than you'd hear on fifties and sixties soundtracks.
> Some of that is because today we tend to double parts a lot in order to
> keep costs down.
>
> I think the miking has more to do with the classic sound of that album
> than the recording technology does. And the performance and arrangement
> has more to do with it than the miking. People just don't _do_ that kind
> of arrangement any more.
> --scott
>

Seth Macfarlane of "The Family Guy" does.

This was the last summer our youngest was likely to be living
at home, so I got a bunch of old musicals to watch with her. The
scores were uniformly great, even on lossy old DVD.

--
Les Cargill

[Les Cargill[_2_]]

Al, Cambridge, UK wrote:
> On Sep 7, 10:57 pm, muzician21 > wrote:
>> Someone posted part of a documentary lifted from the "bonus material"
>> from The Incredibles on YouTube. They make a big point of talking
>> about how the orchestra was recorded analog because they wanted "that
>> analog 60's sound" or something to that effect. They show a bank of
>> sync'd reel to reel machines.
>>
>> Assuming one accepts that there's a profound difference in the sound
>> of an analog recorded orchestra, doesn't the fact that at some point
>> it has to be transferred to the digital realm make it a moot point?
>> It goes to digital for the DVD and any soundtrack album. I'm not
>> versed in sound for film but I assume these days by the time it makes
>> it to your local theater it's gone through a digital process - or is
>> that not correct?
>
> Perhaps if the *musicians* know they're recording to analog, they
> might feel different about the session as they watch the tapes go
> round and play subtly differently?
> Al

I doubt it. Might be true for pop groups, but not for
sheet-music folks. Then again, the conductor might be able
to riff on it to improve morale.

--
Les Cargill

[Les Cargill[_2_]]

nebulax wrote:
> On Sep 8, 3:29 pm, Laurence Payne > wrote:
>> On Tue, 8 Sep 2009 10:44:18 -0700 (PDT), nebulax
>>
>> > wrote:
>>> Music recorded in the analog domain can retain most of it's character
>>> even after being transferred to a digital format.
>> Why only "most"?
>
>
> Back in the 16 bit era (oh wait... it's still with us), analog to
> digital transfers tended to lose a lot of their ambiance, spatial
> depth, high frequency detail, "warmth", soundstage width, etc etc.
> This is probably not so true in the 24 bit (and above) digital domain,
> but I don't own a player that can do those bit rates yet, so for me
> that remains to be seen, or rather, heard.
>
> -Neb

http://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem

"If a function x(t) contains no frequencies higher than B hertz, it is
completely determined by giving its ordinates at a series of points
spaced 1/(2B) seconds apart."


--
Les Cargill

[nebulax]

On Sep 11, 11:17*pm, Les Cargill > wrote:
> nebulax wrote:
> > On Sep 8, 3:29 pm, Laurence Payne > wrote:
> >> On Tue, 8 Sep 2009 10:44:18 -0700 (PDT), nebulax
>
> >> > wrote:
> >>> Music recorded in the analog domain can retain most of it's character
> >>> even after being transferred to a digital format.
> >> Why only "most"?
>
> > Back in the 16 bit era (oh wait... it's still with us), analog to
> > digital transfers tended to lose a lot of their ambiance, spatial
> > depth, high frequency detail, "warmth", soundstage width, etc etc.
> > This is probably not so true in the 24 bit (and above) digital domain,
> > but I don't own a player that can do those bit rates yet, so for me
> > that remains to be seen, or rather, heard.
>
> > -Neb
>
> http://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem
>
> "If a function x(t) contains no frequencies higher than B hertz, it is
> completely determined by giving its ordinates at a series of points
> spaced 1/(2B) seconds apart."
>
> --
> Les Cargill


Yes, that's the Nyquist theorem. And your point is...?

-Neb

[William Sommerwerck]

> "If a function x(t) contains no frequencies higher than B hertz, it is
> completely determined by giving its ordinates at a series of points
> spaced 1/(2B) seconds apart."


That should be...

"If a function x(t) contains no frequencies higher than B hertz, it is
completely determined by giving its ordinates at a regularly spaced
series of points less than 1/(2B) seconds apart."

Have I missed anything?

[Les Cargill[_2_]]

nebulax wrote:
> On Sep 11, 11:17 pm, Les Cargill > wrote:
>> nebulax wrote:
>>> On Sep 8, 3:29 pm, Laurence Payne > wrote:
>>>> On Tue, 8 Sep 2009 10:44:18 -0700 (PDT), nebulax
>>>> > wrote:
>>>>> Music recorded in the analog domain can retain most of it's character
>>>>> even after being transferred to a digital format.
>>>> Why only "most"?
>>> Back in the 16 bit era (oh wait... it's still with us), analog to
>>> digital transfers tended to lose a lot of their ambiance, spatial
>>> depth, high frequency detail, "warmth", soundstage width, etc etc.
>>> This is probably not so true in the 24 bit (and above) digital domain,
>>> but I don't own a player that can do those bit rates yet, so for me
>>> that remains to be seen, or rather, heard.
>>> -Neb
>> http://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem
>>
>> "If a function x(t) contains no frequencies higher than B hertz, it is
>> completely determined by giving its ordinates at a series of points
>> spaced 1/(2B) seconds apart."
>>
>> --
>> Les Cargill
>
>
> Yes, that's the Nyquist theorem. And your point is...?
>
> -Neb

"Completely determined". This does not mean there were not bad
implementations.

--
Les Cargill

[Arny Krueger]

"nebulax" > wrote in message

> On Sep 8, 3:29 pm, Laurence Payne
> > wrote:
>> On Tue, 8 Sep 2009 10:44:18 -0700 (PDT), nebulax
>>
>> > wrote:
>>> Music recorded in the analog domain can retain most of
>>> it's character even after being transferred to a
>>> digital format.
>>
>> Why only "most"?

> Back in the 16 bit era (oh wait... it's still with us),
> analog to digital transfers tended to lose a lot of their
> ambiance, spatial depth, high frequency detail, "warmth",
> soundstage width, etc etc.

It's easy to show that such things need not happen.

My approach is to find a great-sounding 24 bit recording, and bounce it down
to 16 and then upsample it back to 24. Obviously the damage done in the 16
bit domain is not *undone* by the upsampling. Actually, the upsampling even
hurts a bit more. The upsampling simply gives us 2 24 bit recordings to
compare, one of which has a 16-bit history while the otther does not. Since
the recordings are identical in format, we can compare them with the
identical same playback chain.

Bottom line, we're comparing formats which are with us for a long time, not
equipment which comes and goes.

In proper listening tests, nobody hears a difference. Of course you can
listen to fade-to-zero with 40 dB added gain and hear a difference, but for
the duration of a regular live recording with live endings, reverb tails
etc., no difference.

> This is probably not so true
> in the 24 bit (and above) digital domain, but I don't own
> a player that can do those bit rates yet, so for me that
> remains to be seen, or rather, heard.

I've had 24/96 equipment for almost a decade, and 24/192 (better grade stuff
like Lynx) for about 5 years. I've also done my homework.

[nebulax]

On Sep 12, 6:20*pm, "Arny Krueger" > wrote:
> "nebulax" > wrote in message
>
>
>
> > On Sep 8, 3:29 pm, Laurence Payne
> > > wrote:
> >> On Tue, 8 Sep 2009 10:44:18 -0700 (PDT), nebulax
>
> >> > wrote:
> >>> Music recorded in the analog domain can retain most of
> >>> it's character even after being transferred to a
> >>> digital format.
>
> >> Why only "most"?
> > Back in the 16 bit era (oh wait... it's still with us),
> > analog to digital transfers tended to lose a lot of their
> > ambiance, spatial depth, high frequency detail, "warmth",
> > soundstage width, etc etc.
>
> It's easy to show that such things need not happen.
>
> My approach is to find a great-sounding 24 bit recording, and bounce it down
> to 16 and then upsample it back to 24. Obviously the damage done in the 16
> bit domain is not *undone* by the upsampling. Actually, the upsampling even
> hurts a bit more. The upsampling simply gives us 2 24 bit recordings to
> compare, one of which has a 16-bit history while the otther does not. Since
> the recordings are identical in format, we can compare them with the
> identical same playback chain.
>
> Bottom line, we're comparing formats which are with us for a long time, not
> equipment which comes and goes.
>
> In proper listening tests, nobody hears a difference. Of course you can
> listen to fade-to-zero with 40 dB added gain and hear a difference, but for
> the duration of a regular live recording with live endings, reverb tails
> etc., no difference.
>
> > This is probably not so true
> > in the 24 bit (and above) digital domain, but I don't own
> > a player that can do those bit rates yet, so for me that
> > remains to be seen, or rather, heard.
>
> I've had 24/96 equipment for almost a decade, and 24/192 (better grade stuff
> like Lynx) for about 5 years. I've also done my homework.


I'm not sure that I understand you here. Are you saying that you took
a 24 bit recording, downsampled it to 16 bit, and then upsampled that
back to 24 bit, and could not tell any difference? So then, should we
all throw our 24 bit stuff away and just be happy with 16? Sorry, but
I won't be doing that.

It reminds me of the best explanation I've heard for or against the
use of better quality audio cables - "If you can't hear the
difference, you don't need them". I would add to that, "If you CAN
tell the difference, trust your own senses, and use them!"

-Neb

[Arny Krueger]

"nebulax" > wrote in message

> On Sep 12, 6:20 pm, "Arny Krueger" >
> wrote:
>> "nebulax" > wrote in message
>>
>>
>>
>>> On Sep 8, 3:29 pm, Laurence Payne
>>> > wrote:
>>>> On Tue, 8 Sep 2009 10:44:18 -0700 (PDT), nebulax
>>
>>>> > wrote:
>>>>> Music recorded in the analog domain can retain most of
>>>>> it's character even after being transferred to a
>>>>> digital format.
>>
>>>> Why only "most"?
>>> Back in the 16 bit era (oh wait... it's still with us),
>>> analog to digital transfers tended to lose a lot of
>>> their ambiance, spatial depth, high frequency detail,
>>> "warmth", soundstage width, etc etc.
>>
>> It's easy to show that such things need not happen.
>>
>> My approach is to find a great-sounding 24 bit
>> recording, and bounce it down to 16 and then upsample it
>> back to 24. Obviously the damage done in the 16 bit
>> domain is not *undone* by the upsampling. Actually, the
>> upsampling even hurts a bit more. The upsampling simply
>> gives us 2 24 bit recordings to compare, one of which
>> has a 16-bit history while the otther does not. Since
>> the recordings are identical in format, we can compare
>> them with the identical same playback chain.
>>
>> Bottom line, we're comparing formats which are with us
>> for a long time, not equipment which comes and goes.
>>
>> In proper listening tests, nobody hears a difference. Of
>> course you can listen to fade-to-zero with 40 dB added
>> gain and hear a difference, but for the duration of a
>> regular live recording with live endings, reverb tails
>> etc., no difference.
>>
>>> This is probably not so true
>>> in the 24 bit (and above) digital domain, but I don't
>>> own a player that can do those bit rates yet, so for me
>>> that remains to be seen, or rather, heard.
>>
>> I've had 24/96 equipment for almost a decade, and 24/192
>> (better grade stuff like Lynx) for about 5 years. I've
>> also done my homework.

> I'm not sure that I understand you here. Are you saying
> that you took a 24 bit recording, downsampled it to 16
> bit, and then upsampled that back to 24 bit,

Yes, You have to do this right with good resampling software, or you will
introduce some spurious differences.

> and could
> not tell any difference?

Not just me.

> So then, should we all throw our
> 24 bit stuff away and just be happy with 16?

It all depends. Those tests involved finished recordings that peaked within
a few dB of FS, were mixed down, and all that.

The believable arguments related to the use of 24 bits relates to things
like live recordings where you feel the need for putting in a lot of
headroom.

> Sorry, but I won't be doing that.

Suit yourself.

> It reminds me of the best explanation I've heard for or
> against the use of better quality audio cables - "If you
> can't hear the difference, you don't need them"

You're comparing apples and oranges, my friend. Cable magic is BS generally
completely devoid of measurable differences. At least when you compare 16
and 24 bits there is in fact a measurable difference.


>. I would add to that, "If you CAN tell the difference, trust your
> own senses, and use them!"

Anybody who thinks that listening tests are just about the senses doesn't
underdand enough about human perception. Perceiving sound is not just about
the ears, but instead involves the most powerful organ in the human body -
the brain. Lots of people go wrong forgetting the potential of the brain to
convince us that we are hearing what we think we want to hear.

[Laurence Payne[_2_]]

On Sat, 12 Sep 2009 15:55:11 -0700 (PDT), nebulax
> wrote:

>I'm not sure that I understand you here. Are you saying that you took
>a 24 bit recording, downsampled it to 16 bit, and then upsampled that
>back to 24 bit, and could not tell any difference? So then, should we
>all throw our 24 bit stuff away and just be happy with 16?

If his test was valid, and 16-bit reproduction proved to be
sufficient, yes, I suppose it does. Would that be a problem for you?


>Sorry, but
>I won't be doing that.
>


>It reminds me of the best explanation I've heard for or against the
>use of better quality audio cables - "If you can't hear the
>difference, you don't need them". I would add to that, "If you CAN
>tell the difference, trust your own senses, and use them!"

Has anyone won the money yet?

[nebulax]

On Sep 13, 4:35*am, Laurence Payne > wrote:
> On Sat, 12 Sep 2009 15:55:11 -0700 (PDT), nebulax
>
> > wrote:
> >I'm not sure that I understand you here. Are you saying that you took
> >a 24 bit recording, downsampled it to 16 bit, and then upsampled that
> >back to 24 bit, and could not tell any difference? So then, should we
> >all throw our 24 bit stuff away and just be happy with 16?
>
> If his test was valid, and 16-bit reproduction proved to be
> sufficient, yes, I suppose it does. *Would that be a problem for you?

If these tests are all about "measurable differences", and there
turned out to be a measurable difference between 16 and 24 bit (which
in fact, there is), then yes it would be a problem. Do you really want
to continue recording in 16 bit from now on?

> >Sorry, but
> >I won't be doing that.
>
> >It reminds me of the best explanation I've heard for or against the
> >use of better quality audio cables - "If you can't hear the
> >difference, you don't need them". I would add to that, "If you CAN
> >tell the difference, trust your own senses, and use them!"
>
> Has anyone won the money yet?


I guess you're talking about the "Amazing Randi", and his audio cable
challenge. Nope, no one has won the money yet, not that I know of. My
point is: anything involving hearing is subjective. What one person
actually hears won't be the same as what another person does, due to
differences in individual ears, age, hearing damage, internal air
pressure, you name it. If I hear a difference in something, but you
tell me that I can't, because your tests and machines tell you so,
that still doesn't affect what I've actually heard. You can credit (or
blame) whatever phenomena you want for my supposed *delusion*, but my
ears will continue to send my that information, regardless. I won't be
taking part in any arguments in which someone tells me "you didn't
hear that", unless they happen to be residing inside of my ear canal!

-Neb

[nebulax]

On Sep 12, 9:34*pm, "Arny Krueger" > wrote:
> "nebulax" > wrote in message
>
>
>
>
>
> > On Sep 12, 6:20 pm, "Arny Krueger" >
> > wrote:
> >> "nebulax" > wrote in message
>
>
>
> >>> On Sep 8, 3:29 pm, Laurence Payne
> >>> > wrote:
> >>>> On Tue, 8 Sep 2009 10:44:18 -0700 (PDT), nebulax
>
> >>>> > wrote:
> >>>>> Music recorded in the analog domain can retain most of
> >>>>> it's character even after being transferred to a
> >>>>> digital format.
>
> >>>> Why only "most"?
> >>> Back in the 16 bit era (oh wait... it's still with us),
> >>> analog to digital transfers tended to lose a lot of
> >>> their ambiance, spatial depth, high frequency detail,
> >>> "warmth", soundstage width, etc etc.
>
> >> It's easy to show that such things need not happen.
>
> >> My approach is to find a great-sounding 24 bit
> >> recording, and bounce it down to 16 and then upsample it
> >> back to 24. Obviously the damage done in the 16 bit
> >> domain is not *undone* by the upsampling. Actually, the
> >> upsampling even hurts a bit more. The upsampling simply
> >> gives us 2 24 bit recordings to compare, one of which
> >> has a 16-bit history while the otther does not. Since
> >> the recordings are identical in format, we can compare
> >> them with the identical same playback chain.
>
> >> Bottom line, we're comparing formats which are with us
> >> for a long time, not equipment which comes and goes.
>
> >> In proper listening tests, nobody hears a difference. Of
> >> course you can listen to fade-to-zero with 40 dB added
> >> gain and hear a difference, but for the duration of a
> >> regular live recording with live endings, reverb tails
> >> etc., no difference.
>
> >>> This is probably not so true
> >>> in the 24 bit (and above) digital domain, but I don't
> >>> own a player that can do those bit rates yet, so for me
> >>> that remains to be seen, or rather, heard.
>
> >> I've had 24/96 equipment for almost a decade, and 24/192
> >> (better grade stuff like Lynx) for about 5 years. I've
> >> also done my homework.
> > I'm not sure that I understand you here. Are you saying
> > that you took a 24 bit recording, downsampled it to 16
> > bit, and then upsampled that back to 24 bit,
>
> Yes, You have to do this right with good resampling software, or you will
> introduce some spurious differences.
>
> > and could
> > not tell any difference?
>
> Not just me.
>
> > So then, should we all throw our
> > 24 bit stuff away and just be happy with 16?
>
> It all depends. *Those tests involved finished recordings that peaked within
> a few dB of FS, were mixed down, and all that.
>
> The believable arguments related to the use of 24 bits relates to things
> like live recordings where you feel the need for putting in a lot of
> headroom.


There's more than headroom being lost there. Maybe you don't have a
test instrument that can measure the 'detail' in a recording, but
simple logic would would determine that more information can be stored
in a 24 bit format over a 16 one. If you're buying into the "if it
can't be measured, it doesn't exist" policy, then the differences
between those formats can be easily measured.


> > *Sorry, but I won't be doing that.
>
> Suit yourself.


I will. And if you're using resigned to 16 bit converters, good luck
drumming up business in the future.


> > It reminds me of the best explanation I've heard for or
> > against the use of better quality audio cables - "If you
> > can't hear the difference, you don't need them"
>
> You're comparing apples and oranges, my friend. Cable magic is BS generally
> completely devoid of measurable differences. At least when you compare 16
> and 24 bits there is in fact a measurable difference.


You're right, there is a measurable difference. So, why use 16 bit
instead of 24?


> >. I would add to that, "If you CAN tell the difference, trust your
> > own senses, and use them!"
>
> Anybody who thinks that listening tests are just about the senses doesn't
> underdand enough about human perception. Perceiving sound is not just about
> the ears, but instead involves the most powerful organ in the human body -
> the brain. Lots of people go wrong forgetting the potential of the brain to
> convince us that we are hearing what we think we want to hear.


The last time I checked, my senses and brain were all connected
together and working fine. If you want to determine that I'm having
'auditory delusions', then I can't stop you, but you don't really
don't know what my ears are (or are not) hearing unless you live in my
ear canal. And as I said earlier, I won't be taking part in any
discussions about "you're not hearing that, because it can't be
measured". If you want to continue with that train of thought, they're
all waiting for you over in 'rec.audio.opinion'. I think the creator
of our ears (whomever you want to give credit to) knows more about how
they work than we do, and our puny test instruments may never be able
to measure everything our senses can feel.

-Neb

[hank alrich]

Les Cargill > wrote:

> nebulax wrote:
> > On Sep 11, 11:17 pm, Les Cargill > wrote:
> >> nebulax wrote:
> >>> On Sep 8, 3:29 pm, Laurence Payne > wrote:
> >>>> On Tue, 8 Sep 2009 10:44:18 -0700 (PDT), nebulax
> >>>> > wrote:
> >>>>> Music recorded in the analog domain can retain most of it's character
> >>>>> even after being transferred to a digital format.
> >>>> Why only "most"?
> >>> Back in the 16 bit era (oh wait... it's still with us), analog to
> >>> digital transfers tended to lose a lot of their ambiance, spatial
> >>> depth, high frequency detail, "warmth", soundstage width, etc etc.
> >>> This is probably not so true in the 24 bit (and above) digital domain,
> >>> but I don't own a player that can do those bit rates yet, so for me
> >>> that remains to be seen, or rather, heard.
> >>> -Neb
> >> http://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem
> >>
> >> "If a function x(t) contains no frequencies higher than B hertz, it is
> >> completely determined by giving its ordinates at a series of points
> >> spaced 1/(2B) seconds apart."
> >>
> >> --
> >> Les Cargill
> >
> >
> > Yes, that's the Nyquist theorem. And your point is...?
> >
> > -Neb
>
> "Completely determined". This does not mean there were not bad
> implementations.

In theory we can have perfect implementations. In practice, not so much.

That said, the current state of decent kit is wonderful, and even the
good stuff is amazing

--
ha
shut up and play your guitar

[Laurence Payne[_2_]]

>> If his test was valid, and 16-bit reproduction proved to be
>> sufficient, yes, I suppose it does. *Would that be a problem for you?
>
>If these tests are all about "measurable differences", and there
>turned out to be a measurable difference between 16 and 24 bit (which
>in fact, there is), then yes it would be a problem. Do you really want
>to continue recording in 16 bit from now on?

You're changing the question. I'll record into 24 bits, distribute
16. Until an advantage in distributing 24 can be shown.



>> Has anyone won the money yet?
>
>
>I guess you're talking about the "Amazing Randi", and his audio cable
>challenge. Nope, no one has won the money yet, not that I know of. My
>point is: anything involving hearing is subjective. What one person
>actually hears won't be the same as what another person does, due to
>differences in individual ears, age, hearing damage, internal air
>pressure, you name it.

Fine. And the challenge is to prove you can hear it consistently. If
the difference is real, Randi's test will find it.

[Arny Krueger]

"nebulax" > wrote in message


>> It all depends. Those tests involved finished recordings
>> that peaked within a few dB of FS, were mixed down, and
>> all that.
>>
>> The believable arguments related to the use of 24 bits
>> relates to things like live recordings where you feel
>> the need for putting in a lot of headroom.

> There's more than headroom being lost there.

Not really, since the dynamic range limiations based on things like mics,
mic preamps and rooms does a serious number available dynamic range on the
lower end of the equation.

> Maybe you don't have a test instrument that can measure the
> 'detail' in a recording,

Why wouldn't I? Detail has a formal definition - bandwidth and dynamic
range. Both are eminently measurable.

> but simple logic would would

You mean audiophile-grade publication-based intuition?

> determine that more information can be stored in a 24 bit
> format over a 16 one.

Here's a news flash that every live recordist should already have heard -
the limit to the amount of information that we can pack into a live
recording context is not in the digital audio interface these days, if you
can afford even just a good mid-priced one.

> If you're buying into the "if it
> can't be measured, it doesn't exist" policy,

Straw man argument. It can be measured, it correlates with what we hear, and
for a few years we've had far more of it in the realm of digital interfaces
than in the real world.

> then the differences between those formats can be easily measured.

Been there, done that.


>>> Sorry, but I won't be doing that.
>>
>> Suit yourself.

> I will. And if you're using resigned to 16 bit converters,

Resigned? Not I. I have literally dozens of channels of so-called 24 bit
converters, most of which even have more than more than 16 bits worth of
dynamic range. I even have 4 channels of LynxTwo with about 20 dB more
dynamic range than 16 bits theoretical. But, what good does it do me when
good mics amd preamps in a good real-world performance space can only
generate signals with maybe 70 or a shade more worth of dynamic range?

> good luck drumming up business in the future.

Strange thing almost everybody judges my recordings based on how they sound
compared to the source, not any irrelevant numbers that I might throw
around.

I did have one and only one client ask for 24/96 on one gig more than a year
back, but that was for a technical paper. While the client (someone whose
name is well-known in technical audio circles) agrees with my math, he
didn't want to waste time fielding that kind of stupid question from
smart-alecs when he gave the paper at the AES.

>>> It reminds me of the best explanation I've heard for or
>>> against the use of better quality audio cables - "If you
>>> can't hear the difference, you don't need them"
>>
>> You're comparing apples and oranges, my friend. Cable
>> magic is BS generally completely devoid of measurable
>> differences. At least when you compare 16 and 24 bits
>> there is in fact a measurable difference.

> You're right, there is a measurable difference. So, why
> use 16 bit instead of 24?

I'm interested in sound quality, not irrelevant numbers.
>
>>> . I would add to that, "If you CAN tell the difference,
>>> trust your own senses, and use them!"
>>
>> Anybody who thinks that listening tests are just about
>> the senses doesn't underdand enough about human
>> perception. Perceiving sound is not just about the ears,
>> but instead involves the most powerful organ in the
>> human body - the brain. Lots of people go wrong
>> forgetting the potential of the brain to convince us
>> that we are hearing what we think we want to hear.

> The last time I checked, my senses and brain were all
> connected together and working fine.

Are you suggesting that this elminates the possibility of bias?

> If you want to
> determine that I'm having 'auditory delusions',

Let the record show that it was not I that mentioned anything or even
suggested anything about delusions. Presumably, someone is free-associating
based on their own anxieties.

If I mention a word that sounds like the d-word in these contexts, the word
would be illusions. Illusions are normal human behavior, nothing
pathological about them at all.

> then I
> can't stop you, but you don't really don't know what my
> ears are (or are not) hearing unless you live in my ear
> canal.

There you go, you've already disconnected your brain from your ears and have
fastened your attention on a purely innocent segement of the human body's
audio chain.

> And as I said earlier, I won't be taking part in
> any discussions about "you're not hearing that, because
> it can't be measured".

Another straw man that you seem to have found floating out there in space or
perhaps the reaches of your mind.

> If you want to continue with that
> train of thought, they're all waiting for you over in
> 'rec.audio.opinion'.

You're sadly out of date in that sort of thinking as well - rec.audio.opinon
is not an audio discussion group and hasn't been one for years. But if you
want to have intelligent discusionss about human perception and audio, might
I recommend

www.hydrogenaudio.org/forums/index.php

????

> I think the creator of our ears
> (whomever you want to give credit to) knows more about
> how they work than we do,

Of course, but why focus on what we don't know, when we now (since the early
1990s) know so much more than the outdated thinking that one often finds.
You do know that Fletcher and Munson were flaming optimists when it comes to
characterizing the performance of the human ear?

> and our puny test instruments may never be able to measure everything our
> senses can
> feel.

Obviously ignoring the fact that many of our feelings only exist in our
minds.

[Arny Krueger]

"nebulax" > wrote in message

> On Sep 13, 4:35 am, Laurence Payne
> > wrote:
>> On Sat, 12 Sep 2009 15:55:11 -0700 (PDT), nebulax
>>
>> > wrote:
>>> I'm not sure that I understand you here. Are you saying
>>> that you took a 24 bit recording, downsampled it to 16
>>> bit, and then upsampled that back to 24 bit, and could
>>> not tell any difference? So then, should we all throw
>>> our 24 bit stuff away and just be happy with 16?
>>
>> If his test was valid, and 16-bit reproduction proved to
>> be sufficient, yes, I suppose it does. Would that be a
>> problem for you?
>
> If these tests are all about "measurable differences",
> and there turned out to be a measurable difference
> between 16 and 24 bit (which in fact, there is),

Very old news, but news that needs to be interpreted in context.

You're aware of the extant JAES paper about the non-discernable impact of
the inclusion of 16 bit, 44 KHz coding on so-called high-resolution audio
channels for listening to finished productions?

> then yes it would be a problem. Do you really want to continue
> recording in 16 bit from now on?

Straw man argument that I deep-sixed for you in another thread.

>>> Sorry, but
>>> I won't be doing that.
>>
>>> It reminds me of the best explanation I've heard for or
>>> against the use of better quality audio cables - "If
>>> you can't hear the difference, you don't need them". I
>>> would add to that, "If you CAN tell the difference,
>>> trust your own senses, and use them!"

>> Has anyone won the money yet?

No, but lots of money has been wasted by charlatans telling people to
"follow their ears" instead of using their brains.

> I guess you're talking about the "Amazing Randi", and his
> audio cable challenge.

Well, that's the carnival side-show version. Note that Randi's board of
directors is still waiting for someone to demonstrate what the cable mavens
still continuously say long and long is obvious.

> Nope, no one has won the money
> yet, not that I know of. My point is: anything involving
> hearing is subjective.

And that's why I invented ABX subjective testing as applied to audio
evaluations back in the mid-1970s.

> What one person actually hears
> won't be the same as what another person does,

But, if we change the question to something that is relevant like "does it
make a reliably audible difference", then we can easily quantify an answer.

> due to
> differences in individual ears, age, hearing damage,
> internal air pressure, you name it.

Ever try to best the data in the established curves for the threshold of
hearing for pure tones?

There's a dirty little secret that many don't seem to appreciate the
practical relevance of. The human threshold of hearing at 4 KHz is about the
same as the noise created by the Brownian Movement of air particles against
a diaphragm about the size of a human ear drum at STP. The fact that our
ears are less sensitive than that at other frequencies is partially
explained by the fact that our ear's sensitivity at 4 KHz is accentuated by
an internal resonance that obviously can only be effective over a limited
range because that's how resonances work.

Bottom line, your ears can only be so sensitive unless you want to do your
listening with your head immersed in air that is so cold that it is nearly
liquid.

> If I hear a
> difference in something, but you tell me that I can't,
> because your tests and machines tell you so, that still
> doesn't affect what I've actually heard.

No, but I have taken many people who are very unrealistically optimistic
about what they hear and actually put them to the test. Eliminate their
ability to take the test with the answers showing, and they generally come
out being merely human.

> You can credit
> (or blame) whatever phenomena you want for my supposed
> *delusion*,

There's that nasty d-word again. Why do you keep bringing it up? Guilty
conscience? ;-)

> but my ears will continue to send my that
> information, regardless.

Nahh, its your brain that sends you that information. Unfortunately you've
been spending too much time reading the likes of Harry Pearson and John
Atkinson.

>I won't be taking part in any
> arguments in which someone tells me "you didn't hear
> that", unless they happen to be residing inside of my ear
> canal!

Why argue? Why not rely on the abundant empirical evidence that is well
known and generally agreed upon by knowledgeable folks. Or, do you believe
that the AES are just tools of the establishment audio industry that
apparently wants us all to stop improving our technical tools and be happy
with some of the cheaper stuff they make? ;-)

[nebulax]

On Sep 13, 5:51*pm, "Arny Krueger" > wrote:
> "nebulax" > wrote in message
>
>
>
> >> It all depends. Those tests involved finished recordings
> >> that peaked within a few dB of FS, were mixed down, and
> >> all that.
>
> >> The believable arguments related to the use of 24 bits
> >> relates to things like live recordings where you feel
> >> the need for putting in a lot of headroom.
> > There's more than headroom being lost there.
>
> Not really, since the dynamic range limiations based on things like mics,
> mic preamps and rooms does a serious number available dynamic range on the
> lower end of the equation.
>
> > Maybe you *don't have a test instrument that can measure the
> > 'detail' in a recording,
>
> Why wouldn't I? Detail has a formal definition - bandwidth and dynamic
> range. Both are eminently measurable.
>
> > but simple logic would would
>
> You mean audiophile-grade publication-based intuition?
>
> > determine that more information can be stored in a 24 bit
> > format over a 16 one.
>
> Here's a news flash that every live recordist should already have heard -
> the limit to the amount of information that we can pack into a live
> recording context is not in the digital audio interface these days, if you
> can afford even just a good mid-priced one.
>
> > If you're buying into the "if it
> > can't be measured, it doesn't exist" policy,
>
> Straw man argument. It can be measured, it correlates with what we hear, and
> for a few years we've had far more of it in the realm of digital interfaces
> than in the real world.
>
> > then the differences between those formats can be easily measured.
>
> Been there, done that.
>
> >>> Sorry, but I won't be doing that.
>
> >> Suit yourself.
> > I will. And if you're using resigned to 16 bit converters,
>
> Resigned? Not I. I have literally dozens of channels of so-called 24 bit
> converters, most of which even have more than more than 16 bits worth of
> dynamic range. I even have 4 channels of LynxTwo with about 20 dB more
> dynamic range than 16 bits theoretical. But, what good does it do me when
> good mics amd preamps in a good real-world performance space can only
> generate signals with maybe 70 or a shade more worth of dynamic range?
>
> > good luck drumming up business in the future.
>
> Strange thing almost everybody judges my recordings based on how they sound
> compared to the source, not any irrelevant numbers that I might throw
> around.
>
> I did have one and only one client ask for 24/96 on one gig more than a year
> back, but that was for a technical paper. While the client (someone whose
> name is well-known in technical audio circles) agrees with my math, he
> didn't want to waste time fielding that kind of stupid question from
> smart-alecs when he gave the paper at the AES.
>
> >>> It reminds me of the best explanation I've heard for or
> >>> against the use of better quality audio cables - "If you
> >>> can't hear the difference, you don't need them"
>
> >> You're comparing apples and oranges, my friend. Cable
> >> magic is BS generally completely devoid of measurable
> >> differences. At least when you compare 16 and 24 bits
> >> there is in fact a measurable difference.
> > You're right, there is a measurable difference. So, why
> > use 16 bit instead of 24?
>
> I'm interested in sound quality, not irrelevant numbers.
>
>
>
> >>> . I would add to that, "If you CAN tell the difference,
> >>> trust your own senses, and use them!"
>
> >> Anybody who thinks that listening tests are just about
> >> the senses doesn't underdand enough about human
> >> perception. Perceiving sound is not just about the ears,
> >> but instead involves the most powerful organ in the
> >> human body - the brain. Lots of people go wrong
> >> forgetting the potential of the brain to convince us
> >> that we are hearing what we think we want to hear.
> > The last time I checked, my senses and brain were all
> > connected together and working fine.
>
> Are you suggesting that this elminates the possibility of bias?
>
> > If you want to
> > determine that I'm having 'auditory delusions',
>
> Let the record show that it was not I that mentioned anything or even
> suggested anything about delusions. Presumably, someone is free-associating
> based on their own anxieties.
>
> If I mention a word that sounds like the d-word *in these contexts, the word
> would be illusions. Illusions are normal human behavior, nothing
> pathological about them at all.
>
> > then I
> > can't stop you, but you don't really don't know what my
> > ears are (or are not) hearing unless you live in my ear
> > canal.
>
> There you go, you've already disconnected your brain from your ears and have
> fastened your attention on a purely innocent segement of the human body's
> audio chain.
>
> > And as I said earlier, I won't be taking part in
> > any discussions about "you're not hearing that, because
> > it can't be measured".
>
> Another straw man that you seem to have found floating out there in space or
> perhaps the reaches of your mind.
>
> > If you want to continue with that
> > train of thought, they're all waiting for you over in
> > 'rec.audio.opinion'.
>
> You're sadly out of date in that sort of thinking as well - rec.audio.opinon
> is not an audio discussion group and hasn't been one for years. But if you
> want to have intelligent discusionss about human perception and audio, might
> I recommend
>
> www.hydrogenaudio.org/forums/index.php
>
> ????
>
> > I think the creator of our ears
> > (whomever you want to give credit to) knows more about
> > how they work than we do,
>
> Of course, but why focus on what we don't know, when we now (since the early
> 1990s) know so much more than the outdated thinking that one often finds.
> You do know that Fletcher and Munson were flaming optimists when it comes to
> characterizing the performance of the human ear?
>
> > *and our puny test instruments may never be able to measure everything our
> > senses can
> > feel.
>
> Obviously ignoring the fact that many of our feelings only exist in our
> minds.



I'm not going to keep contributing to this thread after this point.
The reason I mentioned 'rec.audio.opinion' is that these same
discussions have been going on over there for years (decades?), and
the whole group has turned into a flame throwing hate-fest. When it
comes down to what an individual person hears or doesn't hear, it
would be wrong for me to even pretend that I know what they're hearing
better than they do, and I hope they extend the same courtesy to me.

-Neb

[Peter Larsen[_3_]]

muzician21 wrote:

> Someone posted part of a documentary lifted from the "bonus material"
> from The Incredibles on YouTube. They make a big point of talking
> about how the orchestra was recorded analog because they wanted "that
> analog 60's sound" or something to that effect. They show a bank of
> sync'd reel to reel machines.

Didn't see the link posted, here it is:

http://www.youtube.com/watch?v=IrVF3a2Wtuk


Kind regards

Peter Larsen