[Audio Empire]

On Sat, 2 Oct 2010 13:38:16 -0700, Arny Krueger wrote
(in article ):

"Audio Empire" wrote in message


If you are trying to say that shorter word length (say,
20 or 16-bit instead of �24-bit) and lower sampling rates
(say, 96, 88.2, 48, or 44.1 KHz) will yield sound
indistinguishable from 24-bit/192 Khz, you are incorrect.

This is a clear case of someone trying to establish their opinon by fiat as
being the only valid opinion, in the face of a world of evidence that is
seemingly far more compelling than the limited and questionable data which
he himself has presented.

Not an opinion. Absolute fact.

Just to repeat the obvious, there is considerable evidence, gathered under
highly controlled circumstances by a large number of independent qualified
and amateur observers that says that shorter word length (say, 20 or 16-bit
instead of �24-bit) and lower sampling rates (say, 96, 88.2, 48, or 44.1
KHz) will yield sound
indistinguishable from 24-bit/192 Khz.

Uh, no it doesn't. Sorry that you (and perhaps others) can't hear (or perhaps
recognize) the improvement to imaging, ambience retrieval, space around the
instruments, etc. But that's not my problem.

Understanding this puts you are in total agreement with the best information
that is currently available, both theoretical and real-world.

Reliable and up-to-date knowlege of the real world performance of recording
setups and psychoacoutics, predicts this result. �IOW, if you know how
listeners perform and you know what kind of results you obtain when you
actually record acoustic music, the above real-world results are no surprise
to you at all.

Even when the results are different from what your view predicts?


high-res audio sounds much more alive, with much better
localization cues, more "air" around the instruments, and
much better low-level detail and ambience than is
possible with 16/44.1 or 48 Khz.

This is one of those effects, like the benefits of talking to plants, that
disappers under reasonble experimental controls.

Sure, that's why so much of the recording industry masters at the higher
sampling frequencies and 24 or 32-bit.

Granted, normal CD
resolution is very good, but the higher bit-rates and
longer word lengths are much better yet. They gild the
lily in such a way as to make the argument of analog vs
digital sound completely moot and will disarm the digital
skeptics completely.

Intreresting that so many (probably thousands) have done comparisons �like
this and been reduced to random guessing, once the statistical results are
known.

And I have done several such tests and the 24/192 recordings always sound
better and can be easily distinguished from the lower sample rate and shorter
word-length version.

Unfortunately our correspondent's approach to this problem has seemingly
been to simply avoid gathering enough data for a proper statistical
analysis.

Most working, professional recording engineers, especially those recording
classical music tend to agree with me. That's why most modern
digital recording is done at 24/192 or at least 24/96.

[Arny Krueger]

"Audio Empire" wrote in message

On Sat, 2 Oct 2010 13:38:16 -0700, Arny Krueger wrote
(in article ):

"Audio Empire" wrote in
message

If you are trying to say that shorter word length (say,
20 or 16-bit instead of �24-bit) and lower sampling
rates (say, 96, 88.2, 48, or 44.1 KHz) will yield sound
indistinguishable from 24-bit/192 Khz, you are
incorrect.

This is a clear case of someone trying to establish
their opinon by fiat as being the only valid opinion, in
the face of a world of evidence that is seemingly far
more compelling than the limited and questionable data
which he himself has presented.

Not an opinion. Absolute fact.

If this is such an absolute fact, why do such august bodies such as the AES
treat it as a false or at best controversial claim?

Just to repeat the obvious, there is considerable
evidence, gathered under highly controlled circumstances
by a large number of independent qualified and amateur
observers that says that shorter word length (say, 20
or 16-bit instead of �24-bit) and lower sampling rates
(say, 96, 88.2, 48, or 44.1 KHz) will yield sound
indistinguishable from 24-bit/192 Khz.

Uh, no it doesn't. Sorry that you (and perhaps others)
can't hear (or perhaps recognize) the improvement to
imaging, ambience retrieval, space around the
instruments, etc. But that's not my problem.

Your problem has been explained to you - you have yet to do your first
proper listening test.

At the very least, your previous tests lacked proper statistical analysis.
There weren't enough trials to do proper statistics. You didn't even try.

Understanding this puts you are in total agreement with
the best information
that is currently available, both theoretical and
real-world.

Reliable and up-to-date knowlege of the real world
performance of recording setups and psychoacoutics,
predicts this result. IOW, if you know how listeners
perform and you know what kind of results you obtain
when you actually record acoustic music, the above
real-world results are no surprise to you at all.

Even when the results are different from what your view
predicts?

Show me results that don't require me to totally suspend disbelief.

high-res audio sounds much more alive, with much better
localization cues, more "air" around the instruments,
and much better low-level detail and ambience than is
possible with 16/44.1 or 48 Khz.

This is one of those effects, like the benefits of
talking to plants, that disappers under reasonble
experimental controls.

Sure, that's why so much of the recording industry
masters at the higher sampling frequencies and 24 or
32-bit.

For a person who claims vast production experience, you seem to be unaware
of the fact that good practice can involve using far higher quality during
processing steps than is required for the final result. Therefore, the fact
that higher sampling rates and/or word lengths are used during production
does not support the contention that higher quality is needed in media for
distribution.

Furthermore, the real-world costs of using higher sampling rates and/or
longer words during produciton is now about nil. This still isn't the case
for media that is used for mass distribution. Therefore, cautious persons
might use higher sampling rates and longer data words during production,
even though the final benefits may be non-existent. The incremental cost can
be that low.

Finally, there are some kinds of audio processing such as adding large
amounts of nonlinear distortion, where far higher sampling rates can be
audibly beneficial due to problems with imaging. Good practice is to avoid
adding large amounts of nonlinear distoriton in the distribution phase, so
the extra bandwidth has no such benefit in media that is widely distributed.

Granted, normal CD
resolution is very good, but the higher bit-rates and
longer word lengths are much better yet. They gild the
lily in such a way as to make the argument of analog vs
digital sound completely moot and will disarm the
digital skeptics completely.

Intreresting that so many (probably thousands) have done
comparisons �like this and been reduced to random
guessing, once the statistical results are known.

And I have done several such tests and the 24/192
recordings always sound better and can be easily
distinguished from the lower sample rate and shorter
word-length version.

I have already commented on these evaluations, and it wasn't pretty. If you
want to keep inciting people to make negative comments by repeating the
identical same old claims over and over again, then the ugliness falls on
you!


Unfortunately our correspondent's approach to this
problem has seemingly been to simply avoid gathering
enough data for a proper statistical analysis.

Most working, professional recording engineers,
especially those recording classical music tend to agree
with me.

Yet another unsupported assertion. This claim is probably impossible to
accurately ascertain, and is just more agumentation for the sake of
argumentation.

That's why most modern
digital recording is done at 24/192 or at least 24/96.

Please prove this with audited figures completely documented, or please
cease and desist with vain repetitions of the same old, same old.

[Audio Empire]

On Mon, 4 Oct 2010 06:49:36 -0700, Arny Krueger wrote
(in article ):

"Audio Empire" wrote in message

On Sat, 2 Oct 2010 13:38:16 -0700, Arny Krueger wrote
(in article ):

"Audio Empire" wrote in
message

If you are trying to say that shorter word length (say,
20 or 16-bit instead of �24-bit) and lower sampling
rates (say, 96, 88.2, 48, or 44.1 KHz) will yield sound
indistinguishable from 24-bit/192 Khz, you are
incorrect.

This is a clear case of someone trying to establish
their opinon by fiat as being the only valid opinion, in
the face of a world of evidence that is seemingly far
more compelling than the limited and questionable data
which he himself has presented.

Not an opinion. Absolute fact.

If this is such an absolute fact, why do such august bodies such as the AES
treat it as a false or at best controversial claim?

I don't know. Perhaps it's someone's best financial interest to promote lossy
compression (in this greedy world, I wouldn't put anything past money
interests). Perhaps some people really can't hear the artifacts while others
can. I know that I CAN hear them and so can a lot of people in the recording
business who I know.

Just to repeat the obvious, there is considerable
evidence, gathered under highly controlled circumstances
by a large number of independent qualified and amateur
observers that says that shorter word length (say, 20
or 16-bit instead of �24-bit) and lower sampling rates
(say, 96, 88.2, 48, or 44.1 KHz) will yield sound
indistinguishable from 24-bit/192 Khz.

Uh, no it doesn't. Sorry that you (and perhaps others)
can't hear (or perhaps recognize) the improvement to
imaging, ambience retrieval, space around the
instruments, etc. But that's not my problem.

Your problem has been explained to you - you have yet to do your first
proper listening test.

At the very least, your previous tests lacked proper statistical analysis.
There weren't enough trials to do proper statistics. You didn't even try.

Like I said, I'm just reporting my (and others) findings in this matter. I
know what I can hear and can't hear. I don't really care that you don't agree
with me.

Look, this isn't like cable sound, or even amplifier sound whereby one can
point to measurements and say that there can be no differences from a
scientific point of view, and then prove it with DBTs where no one can hear
the difference between two similar amps or a $4000 length of specialized
speaker "hose" against $10 worth of lamp cord, this is a perceptual thing
that cannot be directly measured and is more like switching between a vinyl
record of a performance and a CD of the same performance and then asking
someone if they can tell the difference. Well, it doesn't take a DBT for
anyone to make that determination. The ticks and pops are THERE on one
source, and NOT THERE on the other. Easy Peasy. No DBT or statistical
analysis required. The same is true with lossy compression. Play the CD and
play a 320 kbps MP3 rip of that CD. The noise modulation is THERE on one
source and NOT THERE on the other. It's really that simple.

Understanding this puts you are in total agreement with
the best information
that is currently available, both theoretical and
real-world.

Reliable and up-to-date knowlege of the real world
performance of recording setups and psychoacoutics,
predicts this result. IOW, if you know how listeners
perform and you know what kind of results you obtain
when you actually record acoustic music, the above
real-world results are no surprise to you at all.

Even when the results are different from what your view
predicts?

Show me results that don't require me to totally suspend disbelief.

Sorry, I can't do that. 1) I don't care that you can't hear compression
artifacts. 2) I'm not trying to convince anyone of anything. I posted
segments of a presentation by a well known and well respected recording
engineer/producer that mirrors my own experiences and conclusions simply to
show that not everyone looks at the emperor and sees his new clothes. Many
people do notice that he is naked. I also must admit that part of my reason
for posting was to wake-up this forum. It looked like the last scene from "On
the Beach" for more than a week.

high-res audio sounds much more alive, with much better
localization cues, more "air" around the instruments,
and much better low-level detail and ambience than is
possible with 16/44.1 or 48 Khz.

This is one of those effects, like the benefits of
talking to plants, that disappers under reasonble
experimental controls.

Sure, that's why so much of the recording industry
masters at the higher sampling frequencies and 24 or
32-bit.

For a person who claims vast production experience, you seem to be unaware
of the fact that good practice can involve using far higher quality during
processing steps than is required for the final result. Therefore, the fact
that higher sampling rates and/or word lengths are used during production
does not support the contention that higher quality is needed in media for
distribution.

I understand that. But you seem unwilling to admit that it MIGHT be done
because many recording engineers and producers can hear the benefits.

Furthermore, the real-world costs of using higher sampling rates and/or
longer words during produciton is now about nil.

Costs notwithstanding, no recording studio would use 24 (or 32-bit) and 96 or
192 KHz sampling rate were there no advantage to both.

This still isn't the case
for media that is used for mass distribution.

Yes it is. I think you'll find that many projects from studios all over the
world are recorded only in high-resolution formats, even if mass distribution
is ONLY going to be at 16/44.1 on CD.


Therefore, cautious persons
might use higher sampling rates and longer data words during production,
even though the final benefits may be non-existent. The incremental cost can
be that low.

While I have no doubt that this is true, I number among my acquaintances (and
have read enough interviews and articles by others whom I do NOT know) enough
recording engineers and producers who say that they HEAR real advantages to
high-resolution recordings to know that most sat that they use these higher
bit-rates and longer word lengths because they result in better sound.

Finally, there are some kinds of audio processing such as adding large
amounts of nonlinear distortion, where far higher sampling rates can be
audibly beneficial due to problems with imaging. Good practice is to avoid
adding large amounts of nonlinear distoriton in the distribution phase, so
the extra bandwidth has no such benefit in media that is widely distributed.

I'm sure that too is true. But nothing you've said discounts the simple
notion that many engineers use high-res recording practices because they
think it sounds superior to 16/44.1

Granted, normal CD
resolution is very good, but the higher bit-rates and
longer word lengths are much better yet. They gild the
lily in such a way as to make the argument of analog vs
digital sound completely moot and will disarm the
digital skeptics completely.

Intreresting that so many (probably thousands) have done
comparisons �like this and been reduced to random
guessing, once the statistical results are known.

And I have done several such tests and the 24/192
recordings always sound better and can be easily
distinguished from the lower sample rate and shorter
word-length version.

I have already commented on these evaluations, and it wasn't pretty. If you
want to keep inciting people to make negative comments by repeating the
identical same old claims over and over again, then the ugliness falls on
you!

This again is your opinion. I don't know why one with as much experience as
you certainly have would be so myopic and close-minded on this subject, but
it is your business, not mine.

Unfortunately our correspondent's approach to this
problem has seemingly been to simply avoid gathering
enough data for a proper statistical analysis.

Most working, professional recording engineers,
especially those recording classical music tend to agree
with me.

Yet another unsupported assertion. This claim is probably impossible to
accurately ascertain, and is just more agumentation for the sake of
argumentation.

I simply would not mention these people's names without their permission.
That just isn't done.

That's why most modern
digital recording is done at 24/192 or at least 24/96.

Please prove this with audited figures completely documented, or please
cease and desist with vain repetitions of the same old, same old.

If you don't want to discuss this any further, then I suggest that you stop
responding to this particular thread.

[Arny Krueger]

"Audio Empire" wrote in message

On Mon, 4 Oct 2010 06:49:36 -0700, Arny Krueger wrote
(in article ):

"Audio Empire" wrote in
message
On Sat, 2 Oct 2010 13:38:16 -0700, Arny Krueger wrote
(in article ):

"Audio Empire" wrote in
message

If you are trying to say that shorter word length
(say, 20 or 16-bit instead of �24-bit) and lower
sampling rates (say, 96, 88.2, 48, or 44.1 KHz) will
yield sound indistinguishable from 24-bit/192 Khz,
you are incorrect.

This is a clear case of someone trying to establish
their opinon by fiat as being the only valid opinion,
in the face of a world of evidence that is seemingly
far more compelling than the limited and questionable
data which he himself has presented.

Not an opinion. Absolute fact.

If this is such an absolute fact, why do such august
bodies such as the AES treat it as a false or at best
controversial claim?

I don't know. Perhaps it's someone's best financial
interest to promote lossy compression (in this greedy
world, I wouldn't put anything past money interests).

No fair. You keep changing the topic. The topic is sample rates, not lossy
compression.

Perhaps some people really can't hear the artifacts while
others can. I know that I CAN hear them and so can a lot
of people in the recording business who I know.

I know that there are lots of people who refuse to proper listening tests,
and almost to a person they report exceptional results.

Just to repeat the obvious, there is considerable
evidence, gathered under highly controlled
circumstances by a large number of independent
qualified and amateur observers that says that shorter
word length (say, 20 or 16-bit instead of �24-bit)
and lower sampling rates (say, 96, 88.2, 48, or 44.1
KHz) will yield sound indistinguishable from
24-bit/192 Khz.

Uh, no it doesn't. Sorry that you (and perhaps others)
can't hear (or perhaps recognize) the improvement to
imaging, ambience retrieval, space around the
instruments, etc. But that's not my problem.

Your problem has been explained to you - you have yet to
do your first proper listening test.

At the very least, your previous tests lacked proper
statistical analysis. There weren't enough trials to do
proper statistics. You didn't even try.

Like I said, I'm just reporting my (and others) findings
in this matter.

Which turn out to be meaningless according to accepted standards for doing
listening tests of this kind.

I know what I can hear and can't hear. I
don't really care that you don't agree with me.

Of course you care that I agree with you - which is why you keep asserting
your opinion over and over again. If you didn't care, you'd stop posting
this stuff over and over again.

Look, this isn't like cable sound, or even amplifier
sound whereby one can point to measurements and say that
there can be no differences from a scientific point of
view,

In fact the identical same techniques, whether practical or theorectical can
and have been used to evaluate the sample rate problem as were used on the
cable problem and the amplifier problem.

snip vain repetitions of the same-old, same-old

[Jenn[_2_]]

In article ,
Audio Empire wrote:

Perhaps some people really can't hear the artifacts while others
can. I know that I CAN hear them and so can a lot of people in the recording
business who I know.

In casual conversation with our assistant rec art faculty today, I asked
him about the audibility of different sampling rates. He told me about
a challenge that our lead rec arts guy (who is on sabbatical this
semester) accepted last Spring. A student challenged him to a test
between two sampling rates, the music being a recording of one of our
jazz ensembles. He sat out in the studio/rehearsal room listening to
the big JBL monitors and the student was in the booth switching between
the two recordings. There was no way that the faculty guy could see
what recording was being played. According to the account that I heard
today, the music was played in 15 second lengths. The faculty's job was
to indicate via pen and paper whether the example was the higher or
lower rate. He evidently was correct 19 out of 20 times.

Now, before someone says something: I know that this is not a proper
test. I don't even know the two sampling rates, though I will try to
find out tomorrow. But it might lend credence to your point.

--
www.jennifermartinmusic.com

[Audio Empire]

On Mon, 4 Oct 2010 19:13:55 -0700, Jenn wrote
(in article ):

In article ,
Audio Empire wrote:
=20
Perhaps some people really can't hear the artifacts while others=20
can. I know that I CAN hear them and so can a lot of people in the=20
recording=20
business who I know.=20
=20
In casual conversation with our assistant rec art faculty today, I aske=
d=20
him about the audibility of different sampling rates. He told me about=
=20
a challenge that our lead rec arts guy (who is on sabbatical this=20
semester) accepted last Spring. A student challenged him to a test=20
between two sampling rates, the music being a recording of one of our=20
jazz ensembles. He sat out in the studio/rehearsal room listening to=20
the big JBL monitors and the student was in the booth switching between=
=20
the two recordings. There was no way that the faculty guy could see=20
what recording was being played. According to the account that I heard=
=20
today, the music was played in 15 second lengths. The faculty's job wa=
s=20
to indicate via pen and paper whether the example was the higher or=20
lower rate. He evidently was correct 19 out of 20 times.
=20
Now, before someone says something: I know that this is not a proper=20
test. I don't even know the two sampling rates, though I will try to=20
find out tomorrow. But it might lend credence to your point.
=20
=20

Interesting. This result tallies with several similar ad-hoc tests that I=
=20
have performed with similar results. One thing you need to find out thoug=
h =AD=20
were the two samples perfectly level matched? If they weren't, then even =
a=20
small mismatch will make the two samples sound different enough to be=20
distinguished one from the other, even if the two samples were otherwise=20
identical.=20

[Arny Krueger]

"Jenn" wrote in message

In article ,
Audio Empire wrote:

Perhaps some people really can't hear the artifacts
while others
can. I know that I CAN hear them and so can a lot of
people in the recording business who I know.

In casual conversation with our assistant rec art faculty
today, I asked him about the audibility of different
sampling rates. He told me about a challenge that our
lead rec arts guy (who is on sabbatical this semester)
accepted last Spring. A student challenged him to a test
between two sampling rates, the music being a recording
of one of our jazz ensembles. He sat out in the
studio/rehearsal room listening to the big JBL monitors
and the student was in the booth switching between the
two recordings.

At this point many critical points remain unanswered about the recordings.

There was no way that the faculty guy
could see what recording was being played. According to
the account that I heard today, the music was played in
15 second lengths. The faculty's job was to indicate via
pen and paper whether the example was the higher or lower
rate. He evidently was correct 19 out of 20 times.

Three words - single blind test. The ghost of Clever Hans the talking horse
was running all through this. Other potentially serious flaws exist as
well.

Now, before someone says something: I know that this is
not a proper test. I don't even know the two sampling
rates, though I will try to find out tomorrow. But it
might lend credence to your point.

The flaw here is the idea that a test that "is not a proper test" is
evidence of *anything* or could lend one iota of credence to any point.

It is a shame that there's so much fuzzy thinking in the world of audio that
anecdotes like this get repeated.

[Jenn[_2_]]

In article ,
Audio Empire wrote:

On Mon, 4 Oct 2010 19:13:55 -0700, Jenn wrote
(in article ):

In article ,
Audio Empire wrote:
=20
Perhaps some people really can't hear the artifacts while others=20
can. I know that I CAN hear them and so can a lot of people in the=20
recording=20
business who I know.=20
=20
In casual conversation with our assistant rec art faculty today, I aske=
d=20
him about the audibility of different sampling rates. He told me about=
=20
a challenge that our lead rec arts guy (who is on sabbatical this=20
semester) accepted last Spring. A student challenged him to a test=20
between two sampling rates, the music being a recording of one of our=20
jazz ensembles. He sat out in the studio/rehearsal room listening to=20
the big JBL monitors and the student was in the booth switching between=
=20
the two recordings. There was no way that the faculty guy could see=20
what recording was being played. According to the account that I heard=
=20
today, the music was played in 15 second lengths. The faculty's job wa=
s=20
to indicate via pen and paper whether the example was the higher or=20
lower rate. He evidently was correct 19 out of 20 times.
=20
Now, before someone says something: I know that this is not a proper=20
test. I don't even know the two sampling rates, though I will try to=20
find out tomorrow. But it might lend credence to your point.
=20
=20

Interesting. This result tallies with several similar ad-hoc tests that I=
=20
have performed with similar results. One thing you need to find out thoug=
h =AD=20
were the two samples perfectly level matched? If they weren't, then even =
a=20
small mismatch will make the two samples sound different enough to be=20
distinguished one from the other, even if the two samples were otherwise=20
identical.=20

Yes, I understand that they were level matched.

By the way, the faculty in question is really an amazing guy. SUPERB
musician (played both jazz piano and jazz and legit trumpet in the LA
studios for a few years, wrote charts for Maynard Ferguson, the Tonight
Show Band, et al), perfect pitch, and knows his way around the Digi
Command D like no one I've seen. He's a great teacher too.

--
www.jennifermartinmusic.com

[Jenn[_2_]]

In article ,
"Arny Krueger" wrote:

"Jenn" wrote in message

In article ,
Audio Empire wrote:

Perhaps some people really can't hear the artifacts
while others
can. I know that I CAN hear them and so can a lot of
people in the recording business who I know.

In casual conversation with our assistant rec art faculty
today, I asked him about the audibility of different
sampling rates. He told me about a challenge that our
lead rec arts guy (who is on sabbatical this semester)
accepted last Spring. A student challenged him to a test
between two sampling rates, the music being a recording
of one of our jazz ensembles. He sat out in the
studio/rehearsal room listening to the big JBL monitors
and the student was in the booth switching between the
two recordings.

At this point many critical points remain unanswered about the recordings.

There was no way that the faculty guy
could see what recording was being played. According to
the account that I heard today, the music was played in
15 second lengths. The faculty's job was to indicate via
pen and paper whether the example was the higher or lower
rate. He evidently was correct 19 out of 20 times.

Three words - single blind test. The ghost of Clever Hans the talking horse
was running all through this. Other potentially serious flaws exist as
well.

Now, before someone says something: I know that this is
not a proper test. I don't even know the two sampling
rates, though I will try to find out tomorrow. But it
might lend credence to your point.

The flaw here is the idea that a test that "is not a proper test" is
evidence of *anything* or could lend one iota of credence to any point.

It is a shame that there's so much fuzzy thinking in the world of audio that
anecdotes like this get repeated.

Yes, an incomplete account, granted. That's why I wrote, "I know that
this is not a proper test." I should just shut up...

--
www.jennifermartinmusic.com

[Audio Empire]

On Tue, 5 Oct 2010 09:26:10 -0700, Jenn wrote
(in article ):

In article ,
Audio Empire wrote:

On Mon, 4 Oct 2010 19:13:55 -0700, Jenn wrote
(in article ):

In article ,
Audio Empire wrote:
=20
Perhaps some people really can't hear the artifacts while others=20
can. I know that I CAN hear them and so can a lot of people in the=20
recording=20
business who I know.=20
=20
In casual conversation with our assistant rec art faculty today, I aske=
d=20
him about the audibility of different sampling rates. He told me about=
=20
a challenge that our lead rec arts guy (who is on sabbatical this=20
semester) accepted last Spring. A student challenged him to a test=20
between two sampling rates, the music being a recording of one of our=20
jazz ensembles. He sat out in the studio/rehearsal room listening to=20
the big JBL monitors and the student was in the booth switching between=
=20
the two recordings. There was no way that the faculty guy could see=20
what recording was being played. According to the account that I heard=
=20
today, the music was played in 15 second lengths. The faculty's job wa=
s=20
to indicate via pen and paper whether the example was the higher or=20
lower rate. He evidently was correct 19 out of 20 times.
=20
Now, before someone says something: I know that this is not a proper=20
test. I don't even know the two sampling rates, though I will try to=20
find out tomorrow. But it might lend credence to your point.
=20
=20

Interesting. This result tallies with several similar ad-hoc tests that I=
=20
have performed with similar results. One thing you need to find out thoug=
h =AD=20
were the two samples perfectly level matched? If they weren't, then even =
a=20
small mismatch will make the two samples sound different enough to be=20
distinguished one from the other, even if the two samples were otherwise=20
identical.=20

Yes, I understand that they were level matched.

By the way, the faculty in question is really an amazing guy. SUPERB
musician (played both jazz piano and jazz and legit trumpet in the LA
studios for a few years, wrote charts for Maynard Ferguson, the Tonight
Show Band, et al), perfect pitch, and knows his way around the Digi
Command D like no one I've seen. He's a great teacher too.



Sounds like a great guy who seems to be able to hear. I find that so many
people in the audio world actually have very poor listening skills and even
though they pontificate endlessly about listening tests and AES white papers,
they have not actually trained their ears to really hear what's going on in
recorded and reproduced music.

[Orlando Enrique Fiol]

Audio Empire wrote:
I find that so many people in the audio world actually have very poor
listening skills and even though they pontificate endlessly about listening
tests and AES white papers, they have not actually trained their ears to really
hear what's going on in
recorded and reproduced music.

Being totally blind and having absolute pitch, I find I can hear both sonic
details and actual musical intricacy.

What's being lost in this discussion is the portability, convenience and cross
platform intelligibility of compressed file formats such as mp3. Most media
storage devices, such as Ipods, mp3 players and laptops, are consciously
marketed as being able to store average numbers of songs. Consumers, intrigued
and enthused by the theoretical ability to store as much of their collection as
possible, opt for lower quality files if it means they can carry more of their
collection with them. Although external terrabite drives can be purchased for
under $100, they are by no means as portable or directly playable as other
media storage devices, which ultimately means that consumers are forced to pay
much more for less storage space.

Another perhaps uncomfortable overlooked point is that certain types of audio
simply do not require pristine representation. Intentionally distorted vocals
and guitars, compressed drums, drum machine beats and electronically
manipulated samples simply do not improve with higher bit and sampling rates.
Only acoustic instruments slightly improve their spacial characteristics when
recorded this way. It is therefore plausible that consumers remain largely
unable to hear differences between compressed and uncompressed audio because
the recorded material minimizes those differences.

Orlando

[Audio Empire]

On Wed, 6 Oct 2010 10:21:12 -0700, Orlando Enrique Fiol wrote
(in article ):

Audio Empire wrote:
I find that so many people in the audio world actually have very poor
listening skills and even though they pontificate endlessly about listening
tests and AES white papers, they have not actually trained their ears to
really
hear what's going on in
recorded and reproduced music.

Being totally blind and having absolute pitch, I find I can hear both sonic
details and actual musical intricacy.

What's being lost in this discussion is the portability, convenience and
cross
platform intelligibility of compressed file formats such as mp3.

I have no problem with those who use or listen to compressed formats such as
MP3 for convenience and/or portability. Basically, it's down to whether or
not you can hear and are bothered enough by the artifacts of these formats to
not be able to listen "around" them. I suspect that varies from person to
person. I have no problem, for instance, listening around ticks and pops in
vinyl records, but cannot abide MP3 artifacts for serious listening. OTOH, I
can and do listen to MP3 et al via both internet radio (as background) and
satellite radio (in the car where these artifacts are masked by tire and road
noise). I do not use MP3 files on my iPod for headphone listening, preferring
instead to use Apple Lossless Compression for that task.



Most media
storage devices, such as Ipods, mp3 players and laptops, are consciously
marketed as being able to store average numbers of songs. Consumers,
intrigued
and enthused by the theoretical ability to store as much of their collection
as
possible, opt for lower quality files if it means they can carry more of
their
collection with them.

Of course.

Although external terrabite drives can be purchased for
under $100, they are by no means as portable or directly playable as other
media storage devices, which ultimately means that consumers are forced to
pay
much more for less storage space.

Yes.

Another perhaps uncomfortable overlooked point is that certain types of audio

simply do not require pristine representation. Intentionally distorted vocals

and guitars, compressed drums, drum machine beats and electronically
manipulated samples simply do not improve with higher bit and sampling rates.


That may well be true, and I SUSPECT that it is, However, I do not like or
listen to those types of music where this might be the case. I listen mostly
to Classical music, Symphonic film scores, and traditional jazz (Bill Evans,
Stan Getz, Miles Davis, et al).


Only acoustic instruments slightly improve their spacial characteristics when

recorded this way. It is therefore plausible that consumers remain largely
unable to hear differences between compressed and uncompressed audio because
the recorded material minimizes those differences.

I do not disagree and in fact have offered that explanation as a possibility
here on this NG before. Thank you for your comments.

Orlando