Can mp3 quality be improved? [Archive]

Audio Empire

December 12th 11, 04:24 AM

On Sun, 11 Dec 2011 14:08:58 -0800, Chuck Finley wrote
(in article >):

> I bought an Escient music server several years ago when hard drive space was
> still relatively expensive. Most of my music on there is encoded at 320 and
> 192. Reviews of DACs typically discuss how they can improve the sound of CD
> quality or hi-rez music, but I'm wondering what effect they would have on
> compressed music. Would some kind of up-sampling device have to be added to
> the DAC for this? Thanks.

OK. 192 and 320 KBPS are considered essentially "transparent", and , as far
as I can tell 320 actually is transparent to my ears. 192, OTOH, is
essentially transparent ON SPEAKERS, but on headphones, I can hear the
artifacts. They are essentially the same artifacts that I hear at 128 KBPS
and lower (noise bursts accompanying solo percussive sounds such as piano and
acoustic guitar) only much more attenuated and of shorter duration.

I use a Logitech Squeezebox Touch to listen to streaming internet radio (the
only MP3 I listen to. My CDs are ripped using Apple Lossless Compression
(ALCS) on iTunes) and the digital output of that is fed to my 24/192 DAC
through a Sonic Frontiers D2D digital up-converter (to 24/96). Basically, the
difference between the upsampled and non-upsampled MP3 audio is very subtle.
On a direct, blind comparison using the upconverter's bypass switch and a
friend doing the switching while I wear headphones in another room, I can
hear a difference when he switches but I can't honestly say that one sounds
better than the other, just "different", and frankly I can't even tell which
is which. But to the main point, the one I think you are asking, no,
up-sampling does not eliminate any compression artifacts that might be
present. If they're audible before upsampling, they're there after
upsampling. The main thing to remember about upsampling is that it adds NO
new information to a digital bit stream. It's only advantage (if any) is to
move the sampling filter cut-off from 22.05KHz (given 16-bit/44.1 KHz
sampling rate) to 48 KHz (assuming we're upconverting to 96 KHz). There is
still no information above 22.05 KHz in the reconstructed audio signal.

I hope this answers your question.

Audio Empire

December 12th 11, 04:26 AM

On Sun, 11 Dec 2011 15:47:12 -0800, bob wrote
(in article >):

> On Dec 11, 5:08=A0pm, "Chuck Finley" > wrote:
>> I bought an Escient music server several years ago when hard drive space was
>> still relatively expensive. Most of my music on there is encoded at 320 and
>> 192. Reviews of DACs typically discuss how they can improve the sound of CD
>> quality or hi-rez music, but I'm wondering what effect they would have on
>> compressed music. Would some kind of up-sampling device have to be added to
>> the DAC for this? Thanks.
>
> Once an MP3 is made, the loss in sound quality is permanent. No way to
> restore it.
>
> That said, most of the time for most listeners, even a 192 kbps MP3
> will be indistinguishable from a CD. So you haven't lost as much as
> you think you have.
>
> Also, don't read DAC reviews. They're essentially creative writing
> exercises.
>
> bob
>

I disagree with that. DACs differ quite a bit in their sound. If you don't
think they do, then you probably haven't DBT'ed a DAC such as a MSB DAC IV
against a Musical Fidelity V-DAC or a dCS Debussy against a Benchmark DAC1
or a Music Streamer II! They all sound quite different. Especially in the
treble presentation and soundstaging.

Edmund[_2_]

December 12th 11, 11:35 AM

On Sun, 11 Dec 2011 22:08:58 +0000, Chuck Finley wrote:

> I bought an Escient music server several years ago when hard drive space
> was still relatively expensive. Most of my music on there is encoded at
> 320 and 192. Reviews of DACs typically discuss how they can improve the
> sound of CD quality or hi-rez music, but I'm wondering what effect they
> would have on compressed music. Would some kind of up-sampling device
> have to be added to the DAC for this? Thanks.

Clever sneaky marketeers forced the whole word with a
speech defect, MP3 is by no means is compression, it is
reduction. This idiocy let to even further speech defects
because now we have to make distinction between
compression -which in all kinds of different branches per
definition IS lossless and the reduction scheme from MP3.
Since MP3 has thrown away data I would not bother to try
to make it better again, the quality is gone forever.
In MP3 language, puncturing a tire is compressing it.

Edmund

Arny Krueger[_4_]

December 13th 11, 12:07 AM

"Chuck Finley" > wrote in message
...
>I bought an Escient music server several years ago when hard drive space
>was
> still relatively expensive. Most of my music on there is encoded at 320
> and
> 192.

Shouldn't be a problem.

> Reviews of DACs typically discuss how they can improve the sound of CD
> quality or hi-rez music,

Generally, false claims. Audiophile myths. Dreams, not actualities.
Blatantly false sales pitches. The results of sighted evaluations.

> but I'm wondering what effect they would have on compressed music.

No less false.

> Would some kind of up-sampling device have to be added to the DAC for
> this?

Up sampling, other than that which happens implicitly in modern DACs in
order to facilitate digital filtering, is yet another audiophile myth.

Audio Empire

December 13th 11, 12:08 AM

On Mon, 12 Dec 2011 03:35:39 -0800, Edmund wrote
(in article >):

> On Sun, 11 Dec 2011 22:08:58 +0000, Chuck Finley wrote:
>
>> I bought an Escient music server several years ago when hard drive space
>> was still relatively expensive. Most of my music on there is encoded at
>> 320 and 192. Reviews of DACs typically discuss how they can improve the
>> sound of CD quality or hi-rez music, but I'm wondering what effect they
>> would have on compressed music. Would some kind of up-sampling device
>> have to be added to the DAC for this? Thanks.
>
> Clever sneaky marketeers forced the whole word with a
> speech defect, MP3 is by no means is compression, it is
> reduction. This idiocy let to even further speech defects
> because now we have to make distinction between
> compression -which in all kinds of different branches per
> definition IS lossless and the reduction scheme from MP3.
> Since MP3 has thrown away data I would not bother to try
> to make it better again, the quality is gone forever.
> In MP3 language, puncturing a tire is compressing it.
>
> Edmund

While you do have a point, MP3 does meet the digital definition of
compression. I.E., it does allow one to fit a quart of information into the
proverbial pint pot. That it does so by throwing as much as more than 90% of
the signal away is irrelevant to the definition. It is amazing that MP3 music
at 128 or 64 KBPS is even recognizable, though 8^)

Arny Krueger[_4_]

December 13th 11, 12:08 AM

"Audio Empire" > wrote in message
...
> On Sun, 11 Dec 2011 15:47:12 -0800, bob wrote
> (in article >):
>
>> On Dec 11, 5:08=A0pm, "Chuck Finley" > wrote:
>>> I bought an Escient music server several years ago when hard drive space
>>> was
>>> still relatively expensive. Most of my music on there is encoded at 320
>>> and
>>> 192. Reviews of DACs typically discuss how they can improve the sound of
>>> CD
>>> quality or hi-rez music, but I'm wondering what effect they would have
>>> on
>>> compressed music. Would some kind of up-sampling device have to be added
>>> to
>>> the DAC for this? Thanks.
>>
>> Once an MP3 is made, the loss in sound quality is permanent. No way to
>> restore it.
>>
>> That said, most of the time for most listeners, even a 192 kbps MP3
>> will be indistinguishable from a CD. So you haven't lost as much as
>> you think you have.
>>
>> Also, don't read DAC reviews. They're essentially creative writing
>> exercises.
>>
>> bob

> I disagree with that. DACs differ quite a bit in their sound.

They can, as long as you stay clear of well-done bias-controlled tests.

>If you don't think they do, then you probably haven't DBT'ed a DAC such as
>a MSB DAC IV
> against a Musical Fidelity V-DAC or a dCS Debussy against a Benchmark
> DAC1
> or a Music Streamer II!

I was unaware of such serious technical failings in such expensive hardware.

> They all sound quite different.

It is axiomatic that DACs can only sound different if they have serious
technical flaws.

Safiraya

December 13th 11, 02:37 AM

http://zopzop.ru/images_20.gifhttp://zopzop.ru/images_20.gifhttp://zopzop.ru/images_20.gif

Audio Empire

December 13th 11, 04:07 AM

On Mon, 12 Dec 2011 16:08:37 -0800, Arny Krueger wrote
(in article >):

> "Audio Empire" > wrote in message
> ...
>> On Sun, 11 Dec 2011 15:47:12 -0800, bob wrote
>> (in article >):
>>
>>> On Dec 11, 5:08=A0pm, "Chuck Finley" > wrote:
>>>> I bought an Escient music server several years ago when hard drive space
>>>> was
>>>> still relatively expensive. Most of my music on there is encoded at 320
>>>> and
>>>> 192. Reviews of DACs typically discuss how they can improve the sound of
>>>> CD
>>>> quality or hi-rez music, but I'm wondering what effect they would have
>>>> on
>>>> compressed music. Would some kind of up-sampling device have to be added
>>>> to
>>>> the DAC for this? Thanks.
>>>
>>> Once an MP3 is made, the loss in sound quality is permanent. No way to
>>> restore it.
>>>
>>> That said, most of the time for most listeners, even a 192 kbps MP3
>>> will be indistinguishable from a CD. So you haven't lost as much as
>>> you think you have.
>>>
>>> Also, don't read DAC reviews. They're essentially creative writing
>>> exercises.
>>>
>>> bob
>
>> I disagree with that. DACs differ quite a bit in their sound.
>
> They can, as long as you stay clear of well-done bias-controlled tests.

They differ quite a bit when you DO conduct proper bias-controlled tests. I
would know, having done so.

>> If you don't think they do, then you probably haven't DBT'ed a DAC such as
>> a MSB DAC IV
>> against a Musical Fidelity V-DAC or a dCS Debussy against a Benchmark
>> DAC1
>> or a Music Streamer II!
>
> I was unaware of such serious technical failings in such expensive hardware.

Not all of the DACs I mentioned, above, are expensive. The Musical Fidelity
V-DAC, for instance, is only about $300, the Music Streamer II DAC is only
about $350. Certainly the MSB and the dCS are quite costly by comparison. The
fact that inexpensive DACs are, well, let's be kind and just say
"compromised" in their performance, is, basically, my point.

>> They all sound quite different.
>
> It is axiomatic that DACs can only sound different if they have serious
> technical flaws.

And surprise, surprise, the $14,000 MSB DAC IV and and the $11,000 dCS
Debussy DACs DO sound MUCH better than any of the cheap IC-based DAC boxes
(including the $1000 Benchmark DAC1). All of which just reinforces your
comments, above. Cheap DACs DO have, compared to the expensive spread,
"serious technical flaws". In fact, I have found that the only IC DAC chip
that performs substantially better than the "usual suspects" from TI/Burr
Brown and Analog Devices and comes anywhere within a country mile of the
discrete "Ladder DAC" used by MSB or the discrete "Ring DAC" used by dCS is
the ESS 32-bit "SabreDAC". And it still has a long way to go to equal
either, sonically.

Edmund[_2_]

December 13th 11, 12:12 PM

On Tue, 13 Dec 2011 00:08:09 +0000, Audio Empire wrote:

> On Mon, 12 Dec 2011 03:35:39 -0800, Edmund wrote (in article
> >):
>
>> On Sun, 11 Dec 2011 22:08:58 +0000, Chuck Finley wrote:
>>
>>> I bought an Escient music server several years ago when hard drive
>>> space was still relatively expensive. Most of my music on there is
>>> encoded at 320 and 192. Reviews of DACs typically discuss how they can
>>> improve the sound of CD quality or hi-rez music, but I'm wondering
>>> what effect they would have on compressed music. Would some kind of
>>> up-sampling device have to be added to the DAC for this? Thanks.
>>
>> Clever sneaky marketeers forced the whole word with a speech defect,
>> MP3 is by no means is compression, it is reduction. This idiocy let to
>> even further speech defects because now we have to make distinction
>> between compression -which in all kinds of different branches per
>> definition IS lossless and the reduction scheme from MP3. Since MP3 has
>> thrown away data I would not bother to try to make it better again, the
>> quality is gone forever. In MP3 language, puncturing a tire is
>> compressing it.
>>
>> Edmund
>
> While you do have a point, MP3 does meet the digital definition of
> compression.

That IS the speech defect I am talking about!
If that IS the new definition that definition is plain wrong and
misleading. Never before in no other branch compression ever meant
throwing away data or material.
If you end up with less data ( or gas in another area ) then you did
not "compress" it, then you reduced the data by throwing away data and
information.
I suggested before not to call the MP3 reduction- "compression" because
it isn't. Lets call it what it is and forget this idiotic "Lossy Compression"
which is a contradiction in terminus and "lossless compression", compression
always was lossless per definition and I mean the right definition not the raped one.
Using the proper terms for things makes it easier to understand.

Edmund

> I.E., it does allow one to fit a quart of information into
> the proverbial pint pot. That it does so by throwing as much as more
> than 90% of the signal away is irrelevant to the definition. It is
> amazing that MP3 music at 128 or 64 KBPS is even recognizable, though
> 8^)

bob

December 13th 11, 02:26 PM

On Dec 12, 11:07=A0pm, Audio Empire > wrote:

> They differ quite a bit when you DO conduct proper bias-controlled tests.=
I
> would know, having done so.

Our previous discussions of listening tests have revealed a that you
have at best an incomplete understanding of what "bias-controlled
test" means. Meanwhile, there have been numerous published reports of
well-documented such tests, and all have come to conclusions very
different from yours. They've found a few, easily explained
exceptions, but by and large humans can't distinguish between DACs
without their eyes.

> And surprise, surprise, the $14,000 MSB DAC IV and =A0and the $11,000 dCS
> Debussy DACs DO sound MUCH better than any of the cheap IC-based DAC boxe=
s
> (including the $1000 Benchmark DAC1).

Yeah, surprise. Lots of people listen with their wallets.

bob

Walt

December 13th 11, 03:12 PM

On 12/13/2011 7:12 AM, Edmund wrote:
> On Tue, 13 Dec 2011 00:08:09 +0000, Audio Empire wrote:
>

>> While you do have a point, MP3 does meet the digital definition of
>> compression.
>
> That IS the speech defect I am talking about!
> If that IS the new definition that definition is plain wrong and
> misleading. Never before in no other branch compression ever meant
> throwing away data or material.

Um... it's not exactly "new". The term "compression" for what is
essentially data reduction has been in use since at least the 80s,
perhaps earlier.

So while you have a valid point, it's about three decades too late to
fight this linguistic battle.

//Walt

Dick Pierce[_2_]

December 13th 11, 07:20 PM

Walt wrote:
> On 12/13/2011 7:12 AM, Edmund wrote:
>
>>On Tue, 13 Dec 2011 00:08:09 +0000, Audio Empire wrote:
>>
>
>
>>>While you do have a point, MP3 does meet the digital definition of
>>>compression.
>>
>>That IS the speech defect I am talking about!
>>If that IS the new definition that definition is plain wrong and
>>misleading. Never before in no other branch compression ever meant
>>throwing away data or material.
>
>
> Um... it's not exactly "new". The term "compression" for what is
> essentially data reduction has been in use since at least the 80s,
> perhaps earlier.
>
> So while you have a valid point, it's about three decades too late to
> fight this linguistic battle.

Try a lot later than that.

Data reduction techniques have been used in audio for a
long time. And they have been used in music, even "high-
quality" music, for some time as well.

Strictly speaking, techniques such as dBx and Dolby A, Dolby
B and such, are all data-reduction compression techniques.
Their purpose is to attempt to fit as much of the "important"
data into a naroowed-bandwidth channel, be it a transmission
channel or a cassette tape. They all work on the smae principle:
they (physically) discard information which, in the eyes of the
designer, are deemed "insignificant."

And, if you want to play the linguistics game, while still being
technically accurate, the human peripheral auditory system
imposes HUGE amounts of lossy data compression. While it is
possible for the human ear to discern sounds ranging over
a power range in excess of 12 orders of magnitude, it CANNOT
hear, at the same time, two sounds whose level differs by that:
in fact, the instantaneous dynamic range of the peripheral
auditory system is FAR less than that, by many orders of
magnitude. And it does it, through among other things, masking.

--
+--------------------------------+
+ Dick Pierce |
+ Professional Audio Development |
+--------------------------------+

dave a

December 13th 11, 07:20 PM

On 12/13/2011 4:12 AM, Edmund wrote:

>
> That IS the speech defect I am talking about!
> If that IS the new definition that definition is plain wrong and
> misleading. Never before in no other branch compression ever meant
> throwing away data or material.
> If you end up with less data ( or gas in another area ) then you did
> not "compress" it, then you reduced the data by throwing away data and
> information.
> I suggested before not to call the MP3 reduction- "compression" because
> it isn't. Lets call it what it is and forget this idiotic "Lossy Compression"
> which is a contradiction in terminus and "lossless compression", compression
> always was lossless per definition and I mean the right definition not the raped one.
> Using the proper terms for things makes it easier to understand.
>
> Edmund
>

So what would you call jpeg? Or mpeg?

Edmund[_2_]

December 13th 11, 11:56 PM

On Tue, 13 Dec 2011 15:12:09 +0000, Walt wrote:

> On 12/13/2011 7:12 AM, Edmund wrote:
>> On Tue, 13 Dec 2011 00:08:09 +0000, Audio Empire wrote:
>>
>>
>>> While you do have a point, MP3 does meet the digital definition of
>>> compression.
>>
>> That IS the speech defect I am talking about! If that IS the new
>> definition that definition is plain wrong and misleading. Never before
>> in no other branch compression ever meant throwing away data or
>> material.
>
> Um... it's not exactly "new". The term "compression" for what is
> essentially data reduction has been in use since at least the 80s,
> perhaps earlier.
>
> So while you have a valid point, it's about three decades too late to
> fight this linguistic battle.

I mentioned it at the time too.

Edmund
>
> //Walt

Arny Krueger[_4_]

December 14th 11, 12:38 AM

"Audio Empire" > wrote in message
...
> On Mon, 12 Dec 2011 16:08:37 -0800, Arny Krueger wrote
> (in article >):

>> It is axiomatic that DACs can only sound different if they have serious
>> technical flaws.

> And surprise, surprise, the $14,000 MSB DAC IV and and the $11,000 dCS
> Debussy DACs DO sound MUCH better than any of the cheap IC-based DAC boxes
> (including the $1000 Benchmark DAC1).

People say things like that, but of course the scientific evidence is AFAIK
not present and IME probably not forthcoming.

This is a shame, since doing listening tests for the audible flaws of DACs
is a very easy thing to do. They have no appreciable time delays and they
have electrical inputs and outputs.

> All of which just reinforces your comments, above. Cheap DACs DO have,
> compared to the expensive spread,
> "serious technical flaws".

Actually they don't. Not only do many far less expensive DACs lack audible
flaws, they even lack audible flaws when cascaded many times. There are many
scientifically -done listening tests that show this to be true.

DACs are now among the most perfected of all audio components, even when not
costly. They are among the easiest to test scientifically, and also among
the most often tested by scientific means.

Here's just one of many real-world examples:

http://www.ethanwiner.com/aes/

Arny Krueger[_4_]

December 14th 11, 12:40 AM

"bob" > wrote in message
...
On Dec 12, 11:07 pm, Audio Empire > wrote:

>> They differ quite a bit when you DO conduct proper bias-controlled tests.
>> I
>> would know, having done so.

> Our previous discussions of listening tests have revealed a that you
> have at best an incomplete understanding of what "bias-controlled
> test" means. Meanwhile, there have been numerous published reports of
> well-documented such tests, and all have come to conclusions very
> different from yours. They've found a few, easily explained
> exceptions, but by and large humans can't distinguish between DACs
> without their eyes.

This point needs to be underscored as many have been mislead to think that
there is some technical purpose to ultra-high-priced DACs.

If one compares the well-known thresholds of hearing for various technical
flaws to the measured performance of even inexpensive modern DACs, the
probability of audible flaws is put into a proper perspective.

At one time sonically-transparent DACs were *not* the rule. For example, the
DACs in the first generation CDP 101 could be detected by ear if one used
certain non-musical program material during the audition. With 90%+ of
commercial CDs, detection was very difficult or impossible. Even though
there were subtle differences, the lsight audible changes were not a
detraction from listening enjoyment. Thse are the result of
scientifically-conducted tests.

Audiophile lore is that the CDP 101 sounded *horrible*. I actually don't
necesarily doubt this characterization in every case as there is more to CD
player sound quality than just DAC quality. With the CDP 101, performance
with imperfect CDs (e.g. scratched) could include clearly audible flaws that
later technology overcame.

In 1983-85 the far more significant issue was serious mastering problems
that remain irritating characteristics of those specific discs to this day.
No DAC can *fix* a badly-made recording.

Fast forward to today, and one can find up to 6 sonically-transparent DACs
in SOC chips that sell for less than $15 along with a full-function computer
system.

Audio Empire

December 14th 11, 12:45 AM

On Tue, 13 Dec 2011 04:12:44 -0800, Edmund wrote
(in article >):

> On Tue, 13 Dec 2011 00:08:09 +0000, Audio Empire wrote:
>
>> On Mon, 12 Dec 2011 03:35:39 -0800, Edmund wrote (in article
>> >):
>>
>>> On Sun, 11 Dec 2011 22:08:58 +0000, Chuck Finley wrote:
>>>
>>>> I bought an Escient music server several years ago when hard drive
>>>> space was still relatively expensive. Most of my music on there is
>>>> encoded at 320 and 192. Reviews of DACs typically discuss how they can
>>>> improve the sound of CD quality or hi-rez music, but I'm wondering
>>>> what effect they would have on compressed music. Would some kind of
>>>> up-sampling device have to be added to the DAC for this? Thanks.
>>>
>>> Clever sneaky marketeers forced the whole word with a speech defect,
>>> MP3 is by no means is compression, it is reduction. This idiocy let to
>>> even further speech defects because now we have to make distinction
>>> between compression -which in all kinds of different branches per
>>> definition IS lossless and the reduction scheme from MP3. Since MP3 has
>>> thrown away data I would not bother to try to make it better again, the
>>> quality is gone forever. In MP3 language, puncturing a tire is
>>> compressing it.
>>>
>>> Edmund
>>
>> While you do have a point, MP3 does meet the digital definition of
>> compression.
>
> That IS the speech defect I am talking about!
> If that IS the new definition that definition is plain wrong and
> misleading. Never before in no other branch compression ever meant
> throwing away data or material.

It's neither new or misleading. the discrimination between "lossy" and
"lossless" compression, means exactly what it says. Lossy compression makes
files smaller by discarding what someone (or something - such as an
algorithm) has decided to be non-essential information. Lossless compression,
OTOH means that the file has been made smaller by using a less verbose coding
scheme of some type. An example of lossless compression would be the ZIP
format on one's PC. If anything were missing on an expanded copy of a ZIP
file of say, Photoshop, would mean that Photoshop would not and could not
run.

> If you end up with less data ( or gas in another area ) then you did
> not "compress" it, then you reduced the data by throwing away data and
> information.

That's why it's called "lossy compression". In gasses, compression is
compression in it's purest form. Nothing is discarded, but the gas has been
processed to take up less volume by eliminating the empty space between gas
molecules. Often this results in the gas becoming a liquid (like with propane
or LNG), but sometimes not. If you merely vent-off a volume of gas to make
the remaining take up less space, that's not compression, that's merely
reducing the volume (like filling a water bottle and letting the excess run
down the drain). Now if you could throw away a certain volume of, say,
propane, and still have the same amount energy in what's left as you did
before the gas's volume was reduced, then that would be an analogy of digital
compression. Remember in audio, we're don't measure the final sound in those
terms. We measure the perception of that sound at our ear/brain interface.
Remember, the file itself (whether it be a digital audio file or a record
groove) is NOT the sound, it is merely a representation of that sound. If
much of the original waveform has been discarded to make the digital file
representing the audio take up less media storage space, and most of the
listening audience doesn't perceive that anything is missing, then whatever
compression scheme was used was successful.

> I suggested before not to call the MP3 reduction- "compression" because
> it isn't. Lets call it what it is and forget this idiotic "Lossy Compression"
> which is a contradiction in terminus and "lossless compression", compression
> always was lossless per definition and I mean the right definition not the
> raped one.

You are talking apples and oranges. Language is a living thing. The day that
we can't accommodate new meanings for existing words, is the day that the
language starts to die. Might as well go back to Roman-era Latin...

> Using the proper terms for things makes it easier to understand.

Understanding context can also help make words easier to understand.

Audio Empire

December 14th 11, 12:47 AM

On Mon, 12 Dec 2011 16:07:13 -0800, Arny Krueger wrote
(in article >):

> "Chuck Finley" > wrote in message
> ...
>> I bought an Escient music server several years ago when hard drive space
>> was
>> still relatively expensive. Most of my music on there is encoded at 320
>> and
>> 192.
>
> Shouldn't be a problem.
>
>> Reviews of DACs typically discuss how they can improve the sound of CD
>> quality or hi-rez music,
>
> Generally, false claims. Audiophile myths. Dreams, not actualities.
> Blatantly false sales pitches. The results of sighted evaluations.

Not exactly true. While I agree that most DACs that are constructed using IC
converters sound so much alike that the differences (if any) are trivial,
high-end DACs using discrete, proprietary circuitry not only can sound better
than the mass-produced IC chip-based DACs, but they sound significantly
different from one another. These differences manifest themselves mostly as
differences in top-end musicality and sound-staging. The better the DAC, the
more real the top-end sounds, strings, even percussion such as high-hats,
take on a sheen and a realism that one generally only hears live. This is
hardly subtle, and in a DBT is jaw-droppingly and statistically apparent.

>> but I'm wondering what effect they would have on compressed music.
>
> No less false.

What is no less false? He's asking if a stand-alone DAC will help MP3 files
sound better. The answer, of course, is no. Any compression artifacts audible
before up-converting or playing through a mega-buck DAC, will be there after,
as well. The damage is done, there is no "fixing" it after the fact.

>> Would some kind of up-sampling device have to be added to the DAC for
>> this?
>
> Up sampling, other than that which happens implicitly in modern DACs in
> order to facilitate digital filtering, is yet another audiophile myth.

This seems to be a common reaction of people who have limited experience in
this area.

Doug McDonald[_6_]

December 14th 11, 12:47 AM

On 12/13/2011 1:20 PM, Dick Pierce wrote:

>
> Strictly speaking, techniques such as dBx and Dolby A, Dolby
> B and such, are all data-reduction compression techniques.
> Their purpose is to attempt to fit as much of the "important"
> data into a naroowed-bandwidth channel, be it a transmission
> channel or a cassette tape. They all work on the smae principle:
> they (physically) discard information which, in the eyes of the
> designer, are deemed "insignificant."
>
>
I disagree with that. If the ideas behind any of those three systems
work perfectly, as they are supposed to, then they do not discard
information.

The recording media they are used with do of course cover up
information with noise, but those three systems do in fact reduce
the amount lost.

By working perfectly I mean that the encode and decode cycles
are the exact inverse of each other.

Consider the simplest, dBx. It (the wideband one) is a simple
volume changing scheme, with the encode and decode systems being
feedback systems that are, if no sounds are out of frequency range
of the recording medium, exact inverses.

Working properly, so that so that noting between the encoder and decoder
overloads, the only thing that happens is that the noise floor at the
output goes up and down. If the noise floor of the transmission medium
is much larger than half the dynamic range of the
input signal, the output should be essentially an exact copy of the
input. If the transmission system has a much larger noise than that,
then the the output will, at low levels, be a much better
copy of the input than if the encode-decode cycle were not used.

In practice, of course, none of these are every truly perfectly
tuned and effects other than noise level changes will be there to some
degree. If badly mistuned, they can be huge.

But the idea of these does NOT include "lossy" compression ideas.

Doug McDonald

Safiraya

December 14th 11, 03:30 AM

http://zopzop.ru/images_20.gifhttp://zopzop.ru/images_20.gifhttp://zopzop.ru/images_20.gif

Dick Pierce[_2_]

December 14th 11, 05:25 AM

Doug McDonald wrote:
> On 12/13/2011 1:20 PM, Dick Pierce wrote:
>
>>
>> Strictly speaking, techniques such as dBx and Dolby A, Dolby
>> B and such, are all data-reduction compression techniques.
>> Their purpose is to attempt to fit as much of the "important"
>> data into a naroowed-bandwidth channel, be it a transmission
>> channel or a cassette tape. They all work on the smae principle:
>> they (physically) discard information which, in the eyes of the
>> designer, are deemed "insignificant."
>>
>>
> I disagree with that. If the ideas behind any of those three systems
> work perfectly, as they are supposed to, then they do not discard
> information.
>
> The recording media they are used with do of course cover up
> information with noise, but those three systems do in fact reduce
> the amount lost.
>
> By working perfectly I mean that the encode and decode cycles
> are the exact inverse of each other.

Then if that is true, the input should match exactly the output.

The simple fact is that some information in the original is
masked by the noise of the process and therefore lost.

> Consider the simplest, dBx. It (the wideband one) is a simple
> volume changing scheme, with the encode and decode systems being
> feedback systems that are, if no sounds are out of frequency range
> of the recording medium, exact inverses.

Yes, and in the process, information is lost.

> Working properly, so that so that noting between the encoder and decoder
> overloads, the only thing that happens is that the noise floor at the
> output goes up and down. If the noise floor of the transmission medium
> is much larger than half the dynamic range of the input signal, the
> output should be essentially an exact copy of the input. If the
> transmission system has a much larger noise than that,
> then the the output will, at low levels, be a much better
> copy of the input than if the encode-decode cycle were not used.

The broadband dynamic range of the coompact cassette is pretty
seriously limited, open reel tape less so, but in either case,
they can be significantly less, by not a small margin, then
the input signal. SImply consider the noise floor of a pair of
good studio microphones, equivalent to maybe on the order of
10-20 dB SPL, with signals in a concert hall situation easily
exceeding 90 dB SPL: that's a dynamic range that exceeds magnetic
media and FM broadcast: the result is loss of information.

--
+--------------------------------+
+ Dick Pierce |
+ Professional Audio Development |
+--------------------------------+

Doug McDonald[_6_]

December 14th 11, 05:25 AM

On 12/13/2011 6:47 PM, Doug McDonald wrote:
Woe is me!!! TYPOS! ERRORS!, so sorry! Fixed below, corrections ALL CAPS

>
>
> Working properly, so that so that NOTHING between the encoder and decoder
> overloads, the only thing that happens is that the noise floor at the
> output goes up and down. If the DYNAMIC RANGE of the transmission medium
> is much larger than half the dynamic range of the
> input signal (DBX IS 2:1 COMPRESSION) , the output should be essentially an exact copy of the
> input. If the transmission system has a much larger noise than that,
> then the the output will, at low levels, be a much better
> copy of the input than if the encode-decode cycle were not used.
>
Doug

Sebastian Kaliszewski

December 14th 11, 01:57 PM

Audio Empire wrote:
> On Mon, 12 Dec 2011 16:07:13 -0800, Arny Krueger wrote
> (in article >):
>> "Chuck Finley" > wrote in message
>> ...
>>> I bought an Escient music server several years ago when hard drive space
>>> was
>>> still relatively expensive. Most of my music on there is encoded at 320
>>> and
>>> 192.
>> Shouldn't be a problem.
>>
>>> Reviews of DACs typically discuss how they can improve the sound of CD
>>> quality or hi-rez music,
>> Generally, false claims. Audiophile myths. Dreams, not actualities.
>> Blatantly false sales pitches. The results of sighted evaluations.
>
> Not exactly true. While I agree that most DACs that are constructed using IC
> converters sound so much alike that the differences (if any) are trivial,
> high-end DACs using discrete, proprietary circuitry not only can sound better
> than the mass-produced IC chip-based DACs, but they sound significantly
> different from one another. These differences manifest themselves mostly as
> differences in top-end musicality and sound-staging. The better the DAC, the
> more real the top-end sounds, strings, even percussion such as high-hats,
> take on a sheen and a realism that one generally only hears live.

Sorry, but this simply defies common sense. All those completely different made
integrated DACs (they differ in their high level designs, algorithms, filters,
etc) sound the same and measurement show they should. Everyone should expect
that they're simply transparent -- if very different designs sounds converge to
just the one sound, one would expect the convergence is to some predesigned
sound -- and all manufactureres claim that "predesigned sound" is simply being
neutral. Measurements of those devices show the same -- that they are just neutral.

And then some butique discrete component devices (with all the problems of
discrete components like uneven heating) are claimed to all sound different from
both all those integrated things as well as from one another. And sound better.
Better than completely neutral?

> This is
> hardly subtle, and in a DBT is jaw-droppingly and statistically apparent.

Sorry, but all published DBTs show otherwise. And in the case of that one test
you failed to show that there was anything statistically aparent. We only can go
with what you have disclosed about that test, and from what you have disclosed
there is no statistical siginificance (as tehre were serious flaws wrt statistics).

[...]
rgds
\SK
--
"Never underestimate the power of human stupidity" -- L. Lang
--
http://www.tajga.org -- (some photos from my travels)

Edmund[_2_]

December 14th 11, 03:25 PM

On Wed, 14 Dec 2011 00:45:05 +0000, Audio Empire wrote:

> On Tue, 13 Dec 2011 04:12:44 -0800, Edmund wrote (in article
> >):
>
>> On Tue, 13 Dec 2011 00:08:09 +0000, Audio Empire wrote:
>>
>>> On Mon, 12 Dec 2011 03:35:39 -0800, Edmund wrote (in article
>>> >):
>>>
>>>> On Sun, 11 Dec 2011 22:08:58 +0000, Chuck Finley wrote:
>>>>
>>>>> I bought an Escient music server several years ago when hard drive
>>>>> space was still relatively expensive. Most of my music on there is
>>>>> encoded at 320 and 192. Reviews of DACs typically discuss how they
>>>>> can improve the sound of CD quality or hi-rez music, but I'm
>>>>> wondering what effect they would have on compressed music. Would
>>>>> some kind of up-sampling device have to be added to the DAC for
>>>>> this? Thanks.
>>>>
>>>> Clever sneaky marketeers forced the whole word with a speech defect,
>>>> MP3 is by no means is compression, it is reduction. This idiocy let
>>>> to even further speech defects because now we have to make
>>>> distinction between compression -which in all kinds of different
>>>> branches per definition IS lossless and the reduction scheme from
>>>> MP3. Since MP3 has thrown away data I would not bother to try to make
>>>> it better again, the quality is gone forever. In MP3 language,
>>>> puncturing a tire is compressing it.
>>>>
>>>> Edmund
>>>
>>> While you do have a point, MP3 does meet the digital definition of
>>> compression.
>>
>> That IS the speech defect I am talking about! If that IS the new
>> definition that definition is plain wrong and misleading. Never before
>> in no other branch compression ever meant throwing away data or
>> material.
>
> It's neither new or misleading. the discrimination between "lossy" and
> "lossless" compression, means exactly what it says.

I am sure you understand that this addition -lossy vs lossless- is a
result of the misleading term "compression" for something that isn't
compression but reduction.

> Lossy compression
> makes files smaller by discarding what someone (or something - such as
> an algorithm) has decided to be non-essential information. Lossless
> compression, OTOH means that the file has been made smaller by using a
> less verbose coding scheme of some type. An example of lossless
> compression would be the ZIP format on one's PC. If anything were
> missing on an expanded copy of a ZIP file of say, Photoshop, would mean
> that Photoshop would not and could not run.

You got it!
>
>> If you end up with less data ( or gas in another area ) then you did
>> not "compress" it, then you reduced the data by throwing away data and
>> information.
>
> That's why it's called "lossy compression". In gasses, compression is
> compression in it's purest form. Nothing is discarded, but the gas has
> been processed to take up less volume by eliminating the empty space
> between gas molecules. Often this results in the gas becoming a liquid
> (like with propane or LNG), but sometimes not. If you merely vent-off a
> volume of gas to make the remaining take up less space, that's not
> compression, that's merely reducing the volume (like filling a water
> bottle and letting the excess run down the drain). Now if you could
> throw away a certain volume of, say, propane, and still have the same
> amount energy in what's left as you did before the gas's volume was
> reduced, then that would be an analogy of digital compression.

Indeed and physics say that is simply not possible, neither in gas nor
data files.

> Remember
> in audio, we're don't measure the final sound in those terms. We measure
> the perception of that sound at our ear/brain interface. Remember, the
> file itself (whether it be a digital audio file or a record groove) is
> NOT the sound, it is merely a representation of that sound. If much of
> the original waveform has been discarded to make the digital file
> representing the audio take up less media storage space, and most of the
> listening audience doesn't perceive that anything is missing, then
> whatever compression scheme was used was successful.

No if the input file is different that the output file, it is not
compression.

>
>> I suggested before not to call the MP3 reduction- "compression" because
>> it isn't. Lets call it what it is and forget this idiotic "Lossy
>> Compression" which is a contradiction in terminus and "lossless
>> compression", compression always was lossless per definition and I mean
>> the right definition not the raped one.
>
> You are talking apples and oranges. Language is a living thing. The day
> that we can't accommodate new meanings for existing words, is the day
> that the language starts to die. Might as well go back to Roman-era
> Latin...

I do know that language is a living thing, and by all means change the
meaning of words to make it more clear for everyone.
Do NOT change the meaning of words to make it more complicated or misleading.

So let us audio lovers all use the proper terms from now on and hopefully
in a few years from now everyone forgot the stupid term Lossy compression
which is a contradiction in terminus, and use "reduction" or "lousy compression"
for reduction schemes.

>> Using the proper terms for things makes it easier to understand.
>
> Understanding context can also help make words easier to understand.

You know very well I do understand the context perfectly.

Edmund

jwvm

December 14th 11, 07:53 PM

On Dec 13, 7:47=A0pm, Audio Empire > wrote:

<snip>

> Not exactly true. While I agree that most DACs that are constructed using=
IC
> converters sound so much alike that the differences (if any) are trivial,
> high-end DACs using discrete, proprietary circuitry not only can sound be=
tter
> than the mass-produced IC chip-based DACs, but they sound significantly
> different from one another.

You will find that it is much harder to get uniform performance from
electronics using discrete parts compared to using integrated
circuits. The tolerance requirements are much harder to meet and
maintain with discrete components. Indeed, there is good reason to
believe that discrete converters will sound different from each other
since component tolerances will be much poorer.

> These differences manifest themselves mostly as
> differences in top-end musicality and sound-staging. The better the DAC, =
the
> more real the top-end sounds, strings, even percussion such as high-hats,
> take on a sheen and a realism that one generally only hears live. This is
> hardly subtle, and in a DBT is jaw-droppingly and statistically apparent.

These are audiophile terms that are impossible to measure and have not
been demonstrated to actually be reliably perceptible in credible
double-blind tests.

Rockinghorse Winner[_8_]

December 14th 11, 11:48 PM

"Chuck Finley" > writes:

>I bought an Escient music server several years ago when hard drive space was
>still relatively expensive. Most of my music on there is encoded at 320 and
>192. Reviews of DACs typically discuss how they can improve the sound of CD
>quality or hi-rez music, but I'm wondering what effect they would have on
>compressed music. Would some kind of up-sampling device have to be added to
>the DAC for this? Thanks.

My outboard DAC improves all my digital music. It's effects on low
resolution MP3's and streaming audio is variable. It definitely adds some
'thickness' to the Internet radio I listen to, making it less objectionable.
However on particularly bad MP3 transfers it can magnify it's defects.

In sum, I would say, a decent DAC, preferably with variable filters that you
can switch in and out would be a pretty good investment if you listen to a
lot of MP3's.

Terry

Rockinghorse Winner[_8_]

December 14th 11, 11:51 PM

bob > writes:

>On Dec 11, 5:08=A0pm, "Chuck Finley" > wrote:
>> I bought an Escient music server several years ago when hard drive space =
>was
>> still relatively expensive. Most of my music on there is encoded at 320 a=
>nd
>> 192. Reviews of DACs typically discuss how they can improve the sound of =
>CD
>> quality or hi-rez music, but I'm wondering what effect they would have on
>> compressed music. Would some kind of up-sampling device have to be added =
>to
>> the DAC for this? Thanks.

>Once an MP3 is made, the loss in sound quality is permanent. No way to
>restore it.

>That said, most of the time for most listeners, even a 192 kbps MP3
>will be indistinguishable from a CD. So you haven't lost as much as
>you think you have.

192? Really? The difference to me is huge. I mean, it's not objectionable
like more compressed music, but it is deficient. Even 320, while better yet,
is still not up to CD quality.

However, if you are listening on an IPod or through a mass market stereo,
you prolly wouldn't notice much difference, it's true.

Terry

>Also, don't read DAC reviews. They're essentially creative writing
>exercises.

>bob

Arny Krueger[_4_]

December 15th 11, 12:36 AM

"Doug McDonald" > wrote in message
...
> On 12/13/2011 1:20 PM, Dick Pierce wrote:
>
>>
>> Strictly speaking, techniques such as dBx and Dolby A, Dolby
>> B and such, are all data-reduction compression techniques.
>> Their purpose is to attempt to fit as much of the "important"
>> data into a naroowed-bandwidth channel, be it a transmission
>> channel or a cassette tape. They all work on the smae principle:
>> they (physically) discard information which, in the eyes of the
>> designer, are deemed "insignificant."

> I disagree with that. If the ideas behind any of those three systems work
> perfectly, as they are supposed to, then they do not discard information.

I agree with the idea that these methodologies discard information that is
less audible in order to better preserve information that is more audible.

Their basic principle is as simple as the old adage that "There is no such
thing as a free lunch." In fact Dolby A, B, and C do sacrifice elements of
technical accuracy that are not so audible such as gain tracking in order to
obtain improved noise performance where it is audible.

> The recording media they are used with do of course cover up
> information with noise, but those three systems do in fact reduce
> the amount lost.

They reduce the loss of audible accuracy by sacrificing forms of accuracy
that are less audible.

> By working perfectly I mean that the encode and decode cycles
> are the exact inverse of each other.

That is just it. None of the Dolby systems offer perfectly accurate encode
and decode cycles. Pick the right technical and audible test material and
they fall full flat on their pretty little faces.

> Consider the simplest, dBx. It (the wideband one) is a simple
> volume changing scheme, with the encode and decode systems being
> feedback systems that are, if no sounds are out of frequency range
> of the recording medium, exact inverses.

It doesn't happen that way on the test bench.

> Working properly, so that so that noting between the encoder and decoder
> overloads, the only thing that happens is that the noise floor at the
> output goes up and down. If the noise floor of the transmission medium is
> much larger than half the dynamic range of the
> input signal, the output should be essentially an exact copy of the input.
> If the transmission system has a much larger noise than that, then the the
> output will, at low levels, be a much better
> copy of the input than if the encode-decode cycle were not used.

There are time constants in the encode and decode process that cause perfect
transient reproduction to suffer. Furthermore, the recording medium is
itself highly nonlinear, and in general distortion rises rapidly as levels
increase. Therefore, any effort to avoid noise by raising recording levels
adds more nonlinear distoriton. The thing is that the increase in nonlinear
distortion is usually less objectionable to the ear than the noise.

> In practice, of course, none of these are every truly perfectly
> tuned and effects other than noise level changes will be there to some
> degree. If badly mistuned, they can be huge.

Even if "perfectly tuned" there are a number of inherent sources of
inaccuracy.

Look at it this way, there is Dolby A and there is Dolby B. Does Dolby A
have any audible advantages over Dobly B? Of course it does, admittedly at a
cost in terms of added equipment complexity and more demanding setup.

Therefore, Dolby B fails to provide as accurate encode/decode cycle
performance as Dolby A, and we have falsified any claim that the encode and
decode cycles of Dolby B are just as exact inverses of each other as are the
encode and decode cycles of Dolby A. Your argument is falsified by the fact
that both Dolby A and B exist!

Arny Krueger[_4_]

December 15th 11, 12:36 AM

"dave a" > wrote in message
...
> On 12/13/2011 4:12 AM, Edmund wrote:
>
>>
>> That IS the speech defect I am talking about!
>> If that IS the new definition that definition is plain wrong and
>> misleading. Never before in no other branch compression ever meant
>> throwing away data or material.
>> If you end up with less data ( or gas in another area ) then you did
>> not "compress" it, then you reduced the data by throwing away data and
>> information.
>> I suggested before not to call the MP3 reduction- "compression" because
>> it isn't. Lets call it what it is and forget this idiotic "Lossy
>> Compression"
>> which is a contradiction in terminus and "lossless compression",
>> compression
>> always was lossless per definition and I mean the right definition not
>> the raped one.
>> Using the proper terms for things makes it easier to understand.
>>
>> Edmund
>>
>
> So what would you call jpeg? Or mpeg?

Your point is well taken. JPEG and MPEG are examples of lossy compression.
L-Z compressed TIFF is an example of lossless compression of images. I don't
believe there are any common examples of lossless compression of video, but
DV/AVI is far less lossy than MPEG.

Arny Krueger[_4_]

December 15th 11, 12:37 AM

Audio Empire

December 15th 11, 12:40 AM

On Tue, 13 Dec 2011 16:40:51 -0800, Arny Krueger wrote
(in article >):

> "bob" > wrote in message
> ...
> On Dec 12, 11:07 pm, Audio Empire > wrote:
>
>>> They differ quite a bit when you DO conduct proper bias-controlled tests.
>>> I
>>> would know, having done so.
>
>> Our previous discussions of listening tests have revealed a that you
>> have at best an incomplete understanding of what "bias-controlled
>> test" means. Meanwhile, there have been numerous published reports of
>> well-documented such tests, and all have come to conclusions very
>> different from yours. They've found a few, easily explained
>> exceptions, but by and large humans can't distinguish between DACs
>> without their eyes.
>
> This point needs to be underscored as many have been mislead to think that
> there is some technical purpose to ultra-high-priced DACs.
>
> If one compares the well-known thresholds of hearing for various technical
> flaws to the measured performance of even inexpensive modern DACs, the
> probability of audible flaws is put into a proper perspective.
>
> At one time sonically-transparent DACs were *not* the rule. For example, the
> DACs in the first generation CDP 101 could be detected by ear if one used
> certain non-musical program material during the audition. With 90%+ of
> commercial CDs, detection was very difficult or impossible. Even though
> there were subtle differences, the lsight audible changes were not a
> detraction from listening enjoyment. Thse are the result of
> scientifically-conducted tests.
>
> Audiophile lore is that the CDP 101 sounded *horrible*. I actually don't
> necesarily doubt this characterization in every case as there is more to CD
> player sound quality than just DAC quality. With the CDP 101, performance
> with imperfect CDs (e.g. scratched) could include clearly audible flaws that
> later technology overcame.

It did sound lousy. I auditioned one, at the time, next to the pretty little
Philips/Magnavox CD-100, which was only 14-bit. It sounded so much more
musical than the Sony, that I bought it instead. I believe that much of CD's
early bad reputation came from players like the Sony and the Kyocera which
both sounded harsh and distorted, especially in the top end.
>
> In 1983-85 the far more significant issue was serious mastering problems
> that remain irritating characteristics of those specific discs to this day.
> No DAC can *fix* a badly-made recording.

Well, that's certainly true.
>
> Fast forward to today, and one can find up to 6 sonically-transparent DACs
> in SOC chips that sell for less than $15 along with a full-function computer
> system.

Those cheap IC-based DAC chips MIGHT be sonically transparent to the
Hoi-Polloi, but they sure don't sound as good, or image as well as a good,
discrete component DAC such as those used in the MSB DACIV or the dCS
Debussy. This has been noted in DBTs against a number of other DACs using
IC-based Burr-Brown (TI), AMD and ESS DAC chips from manufacturers such as
Benchmark, Antelope, Musical Fidelity, Music Streamer, Cambridge and Weiss,
to name a few and the differences can be measured and easily seen. Look at
the 1/3-octave spectrum-with-noise data, or the Intermodulation spectrum
plots or the high-resolution jitter spectrum data and contrast the results of
the IC-based DACs with those from some of these discrete component units such
as the MSB and dCS units mentioned above. Their superiority is as easy to see
as it is to hear.

I suspect that the DBTs that show all modern DACs to be more-or-less equally
transparent were comparing DACs using the more popular mass-produced
integrated circuit DAC chips. Like I said in another post, only the ESS
32-bit "SabreDAC" has any real edge here, either sonically or by measurement,
and it still doesn't sound or measure as good as the MSB proprietary "Ladder
DAC" or the dCS "Ring DAC".

This type of "everything sounds the same" argument certainly makes audio
cheaper. If everything sounds the same, then there's no reason to buy
anything expensive. A $50 CD player sounds exactly like a multiple thousand
dollar unit so all one needs to buy is the $50 player. All amplifiers sound
exactly alike, so why buy a Krell integrated for $3000 when a $150 TEAC
receiver from Costco performs exactly like it? It's tempting to believe this.
Too bad that neither of these money-saving assumptions is true.....

Take heart, though. A $5 Radio-Shack interconnect DOES sound exactly like a
$4000 pair of Nordost Valhallahs and a hank of 14-gauge lamp cord does
perform identically to a speaker cable from Oracle costing many hundreds of
dollars per foot. Also, the IC processes keep improving and chips like the
SabreDAC are closing-in on the cost-is-no-object designs, so there is hope
for us financial mortals after all!

Audio Empire

December 15th 11, 12:42 AM

On Tue, 13 Dec 2011 16:38:30 -0800, Arny Krueger wrote
(in article >):

> "Audio Empire" > wrote in message
> ...
>> On Mon, 12 Dec 2011 16:08:37 -0800, Arny Krueger wrote
>> (in article >):
>
>>> It is axiomatic that DACs can only sound different if they have serious
>>> technical flaws.
>
>> And surprise, surprise, the $14,000 MSB DAC IV and and the $11,000 dCS
>> Debussy DACs DO sound MUCH better than any of the cheap IC-based DAC boxes
>> (including the $1000 Benchmark DAC1).
>
> People say things like that, but of course the scientific evidence is AFAIK
> not present and IME probably not forthcoming.

Perform your own DBTs and the measurement data for all these DACs "is out
there" Just look at the data. You can see that these designs (and a few more)
are superior.
>
> This is a shame, since doing listening tests for the audible flaws of DACs
> is a very easy thing to do. They have no appreciable time delays and they
> have electrical inputs and outputs.

You obviously haven't been paying attention. These DBTs have been performed
and I have been on the listening panel as I am about once a month. Friday,
I'm going to attend a DBT between a DaVinci 384K DAC against both an MSB DAC
IV (with enhanced outboard power supply) and, time permitting, a dCS Debussy
>
>> All of which just reinforces your comments, above. Cheap DACs DO have,
>> compared to the expensive spread,
>> "serious technical flaws".
>
> Actually they don't. Not only do many far less expensive DACs lack audible
> flaws, they even lack audible flaws when cascaded many times. There are many
> scientifically -done listening tests that show this to be true.

I'm well aware of this. What you seem to be not aware of is the incredible
gap that exists between cost-is-no-object designs, and the predictable
"sameness" of DACs which use off-the-shelf DAC chips from the likes of Analog
Devices, Crystal and TI (Burr-Brown). These devices, such as the AD 1955, the
B-B PCM1737 and the Crystal CS4398 are ubiquitous, and are used in almost
all stand-alone DACs as well as as in disc players. No wonder there is little
or nothing to choose between them.

> DACs are now among the most perfected of all audio components, even when not
> costly. They are among the easiest to test scientifically, and also among
> the most often tested by scientific means.

Yes, and such tests show that the cost-is-no-object designs measure
significantly better and I can tell you that in DBTs, they sound
significantly better too. Especially in the presentation of high-frequencies
and sound-staging. These characteristics are easily heard in a DBT.
>
> Here's just one of many real-world examples:
>
> http://www.ethanwiner.com/aes/

I've seen this before. Let's just say, that the conclusions arrived at are
incomplete.

You want me to say that all modern DACs are essentially transparent? OK, I
agree, they are, given the somewhat narrow and very specific definition of
the term. But when you listen to a really GOOD DAC next to one based on IC
chip technology (which is most of them), you see the error of this
definition. You can daisy-chain a bunch of these DACs and ADCs together, and
yes, they alter the sound not at all, but when you finally use a really good
DAC in a really good, high-resolution sound system. you hear things revealed
in the music that just don't come through with these lesser converters. A
sense of space, a life-like smoothness to strings and a splatter-free
reproduction of cymbals and snare drums that one simply does not hear through
these lesser DACs and one really only encounters in live, unamplified
performances. It's like these lesser DACs perform a homogenization on the
initial conversion, and then preserve that level of homogenization through
all subsequent serial A/D and D/A conversions. That's the best I can do to
describe this.

To bad I can't take you with me to one of these DAC DBTs, so that you can
hear this phenomenon for yourself. You (any of you) would likely be an
instant convert.

But maybe you're better off believing what you now believe. Because once
converted, you would want this level of performance in your own listening
environment, and at the price of a new compact car, these things are simply
out of reach for the vast majority of us.

Audio Empire

December 15th 11, 12:43 AM

On Wed, 14 Dec 2011 07:25:06 -0800, Edmund wrote
(in article >):

> On Wed, 14 Dec 2011 00:45:05 +0000, Audio Empire wrote:
>
>> On Tue, 13 Dec 2011 04:12:44 -0800, Edmund wrote (in article
>> >):
>>
>>> On Tue, 13 Dec 2011 00:08:09 +0000, Audio Empire wrote:
>>>
>>>> On Mon, 12 Dec 2011 03:35:39 -0800, Edmund wrote (in article
>>>> >):
>>>>
>>>>> On Sun, 11 Dec 2011 22:08:58 +0000, Chuck Finley wrote:
<snip>
>>
>> It's neither new or misleading. the discrimination between "lossy" and
>> "lossless" compression, means exactly what it says.
>
> I am sure you understand that this addition -lossy vs lossless- is a
> result of the misleading term "compression" for something that isn't
> compression but reduction.

Lossy compression can be said to be "reduction" (the "standard' music
compression data rate in MP3 is 128 KBPS is 11:1. That means that more than
90% of the original waveform is actually discarded). But Lossless schemes
such as ALCS (Apple Lossless Compression Scheme) and FLAC (Free Lossless
Audio Compression) are not "reductions" as the original waveform is preserved
in the round-trip process.

>> Lossy compression
>> makes files smaller by discarding what someone (or something - such as
>> an algorithm) has decided to be non-essential information. Lossless
>> compression, OTOH means that the file has been made smaller by using a
>> less verbose coding scheme of some type. An example of lossless
>> compression would be the ZIP format on one's PC. If anything were
>> missing on an expanded copy of a ZIP file of say, Photoshop, would mean
>> that Photoshop would not and could not run.
>
> You got it!

Yes, I do "got it." and I see no reason to play at semantic games. "A rose by
any other name", and all that.

>>> If you end up with less data ( or gas in another area ) then you did
>>> not "compress" it, then you reduced the data by throwing away data and
>>> information.
>>
>> That's why it's called "lossy compression". In gasses, compression is
>> compression in it's purest form. Nothing is discarded, but the gas has
>> been processed to take up less volume by eliminating the empty space
>> between gas molecules. Often this results in the gas becoming a liquid
>> (like with propane or LNG), but sometimes not. If you merely vent-off a
>> volume of gas to make the remaining take up less space, that's not
>> compression, that's merely reducing the volume (like filling a water
>> bottle and letting the excess run down the drain). Now if you could
>> throw away a certain volume of, say, propane, and still have the same
>> amount energy in what's left as you did before the gas's volume was
>> reduced, then that would be an analogy of digital compression.
>
> Indeed and physics say that is simply not possible, neither in gas nor
> data files.

Well it is possible in data files because we're not dealing with actual
matter, we are dealing with information. It's nothing new. Let's say we have
two secretaries, both taking dictation of a letter from a single speaker. One
takes the dictation long-hand, and the other uses standard stenographer's
shorthand. When the final letter is typed-up, both should be identical, but
the long-hand dictation might be 5 or 6 pages long while the shorthand
version might be two pages. Both say exactly the same thing and nothing has
been lost but one form of coding the dictation is far less verbose than the
other even though it is represented by far less data.

>> Remember
>> in audio, we're don't measure the final sound in those terms. We measure
>> the perception of that sound at our ear/brain interface. Remember, the
>> file itself (whether it be a digital audio file or a record groove) is
>> NOT the sound, it is merely a representation of that sound. If much of
>> the original waveform has been discarded to make the digital file
>> representing the audio take up less media storage space, and most of the
>> listening audience doesn't perceive that anything is missing, then
>> whatever compression scheme was used was successful.
>
> Not if the input file is different that the output file, it is not
> compression.

But compression is useless without a complementary de-compression. It's a
round-trip process.

>>> I suggested before not to call the MP3 reduction- "compression" because
>>> it isn't. Lets call it what it is and forget this idiotic "Lossy
>>> Compression" which is a contradiction in terminus and "lossless
>>> compression", compression always was lossless per definition and I mean
>>> the right definition not the raped one.
>>
>> You are talking apples and oranges. Language is a living thing. The day
>> that we can't accommodate new meanings for existing words, is the day
>> that the language starts to die. Might as well go back to Roman-era
>> Latin...
>
> I do know that language is a living thing, and by all means change the
> meaning of words to make it more clear for everyone.
> Do NOT change the meaning of words to make it more complicated or misleading.

I don't see that anyone has done that. Most people don't care what audio
compression scheme is used for their music as long as it works and allows
them to fit more music into a finite physical space - such as the fixed
memory size on an iPod. Those of us who do care, seem to understand the
concepts involved at least well enough to know the difference between lossy
and lossless compression schemes.
>
> So let us audio lovers all use the proper terms from now on and hopefully
> in a few years from now everyone forgot the stupid term Lossy compression
> which is a contradiction in terminus, and use "reduction" or "lousy
> compression"
> for reduction schemes.

I'm sorry, I simply don't see a problem here.
>
>
>>> Using the proper terms for things makes it easier to understand.
>>
>> Understanding context can also help make words easier to understand.
>
> You know very well I do understand the context perfectly.

That's not my point. My point is that everyone involved with music
compression algorithms from those constructing them, to the teenager ripping
CDs to iTunes understands from the context what lossy and lossless
compression schemes mean. There is simply no need to change the universally
accepted nomenclature at this stage of the game. At least that's my opinion
on the subject anyway.

Audio Empire

December 15th 11, 12:44 AM

On Wed, 14 Dec 2011 05:57:35 -0800, Sebastian Kaliszewski wrote
(in article >):

> Audio Empire wrote:
>> On Mon, 12 Dec 2011 16:07:13 -0800, Arny Krueger wrote
>> (in article >):
>>> "Chuck Finley" > wrote in message
>>> ...
>>>> I bought an Escient music server several years ago when hard drive space
>>>> was
>>>> still relatively expensive. Most of my music on there is encoded at 320
>>>> and
>>>> 192.
>>> Shouldn't be a problem.
>>>
>>>> Reviews of DACs typically discuss how they can improve the sound of CD
>>>> quality or hi-rez music,
>>> Generally, false claims. Audiophile myths. Dreams, not actualities.
>>> Blatantly false sales pitches. The results of sighted evaluations.
>>
>> Not exactly true. While I agree that most DACs that are constructed using
>> IC
>> converters sound so much alike that the differences (if any) are trivial,
>> high-end DACs using discrete, proprietary circuitry not only can sound
>> better
>> than the mass-produced IC chip-based DACs, but they sound significantly
>> different from one another. These differences manifest themselves mostly as
>> differences in top-end musicality and sound-staging. The better the DAC,
>> the
>> more real the top-end sounds, strings, even percussion such as high-hats,
>> take on a sheen and a realism that one generally only hears live.
>
>
> Sorry, but this simply defies common sense. All those completely different
> made
> integrated DACs (they differ in their high level designs, algorithms,
> filters,
> etc) sound the same and measurement show they should. Everyone should expect
> that they're simply transparent -- if very different designs sounds converge
> to
> just the one sound, one would expect the convergence is to some predesigned
> sound -- and all manufactureres claim that "predesigned sound" is simply
> being
> neutral. Measurements of those devices show the same -- that they are just
> neutral.
>
> And then some butique discrete component devices (with all the problems of
> discrete components like uneven heating) are claimed to all sound different
> from
> both all those integrated things as well as from one another. And sound
> better.
> Better than completely neutral?
>
>> This is
>> hardly subtle, and in a DBT is jaw-droppingly and statistically apparent.
>
> Sorry, but all published DBTs show otherwise. And in the case of that one
> test
> you failed to show that there was anything statistically aparent. We only can

> go
> with what you have disclosed about that test, and from what you have
> disclosed
> there is no statistical siginificance (as tehre were serious flaws wrt
> statistics).
>
> [...]
> rgds
> \SK
>

You realize that this is your opinion and I have mine. Mine is a result of
many DBTs between different DACs to which I've been privy. This stuff is so
easy to hear, that in a recent DBT between a Benchmark DAC1 and a dCS
Debussy, everyone on the listening panel was able to pick out the differences
between the two an average of better than 9 out of 10 tries. Everyone agreed
that the Debussy was far more musical than the Benchmark - and we came to
that conclusion before we even knew what the two DUT even were! The
Benchmark, which is based on an IC DAC (a Burr-Brown, IIRC - but I could be
misremembering), sounded somewhat dull and homogenous by comparison to the
Debussy, lacking in finesse, realism and image specificity. I realize that
conventional wisdom says that an IC DAC from AD, Crystal, ESS or Burr-Brown
should be "transparent", but if transparent vs non-transparent (according to
you that would be the cost-is-no-object discrete designs) yields poorer sound
on the former, and richer, fuller and more lifelike sound with the latter,
then I'll take the latter every time.

Arny Krueger[_4_]

December 15th 11, 12:45 AM

"Audio Empire" > wrote in message
...
> On Mon, 12 Dec 2011 16:07:13 -0800, Arny Krueger wrote
> (in article >):
>
>> "Chuck Finley" > wrote in message
>> ...
>>> I bought an Escient music server several years ago when hard drive space
>>> was
>>> still relatively expensive. Most of my music on there is encoded at 320
>>> and
>>> 192.
>>
>> Shouldn't be a problem.
>>
>>> Reviews of DACs typically discuss how they can improve the sound of CD
>>> quality or hi-rez music,
>>
>> Generally, false claims. Audiophile myths. Dreams, not actualities.
>> Blatantly false sales pitches. The results of sighted evaluations.

> Not exactly true. While I agree that most DACs that are constructed using
> IC
> converters sound so much alike that the differences (if any) are trivial,
> high-end DACs using discrete, proprietary circuitry not only can sound
> better
> than the mass-produced IC chip-based DACs, but they sound significantly
> different from one another.

How can these significant differences arise? We already know that the better
chip DACs are highly sonically transparent, which is to say that you can
pass audio through them many times without causing any reliably-detectible
audible difference. If we hypothesize the existance of some more sonically
accurate component, then the audible differences that it creates must be
even smaller and even less audible.

> These differences manifest themselves mostly as
> differences in top-end musicality and sound-staging. The better the DAC,
> the
> more real the top-end sounds, strings, even percussion such as high-hats,
> take on a sheen and a realism that one generally only hears live. This is
> hardly subtle, and in a DBT is jaw-droppingly and statistically apparent.

How can there be more realism than there is in a signal that
indistinguishable from the origional signal?

>>> but I'm wondering what effect they would have on compressed music.

>> No less false.

> What is no less false?

I'm talking about the audiophile myth that ADC quality can be improved
beyond that which we already achieve with reasonably-priced chips that are
demonstrably sonically transparent, even when cascaded over more than a
dozen repeated conversions.

> He's asking if a stand-alone DAC will help MP3 files
> sound better. The answer, of course, is no. Any compression artifacts
> audible
> before up-converting or playing through a mega-buck DAC, will be there
> after,
> as well. The damage is done, there is no "fixing" it after the fact.

On that we can agree.

>>> Would some kind of up-sampling device have to be added to the DAC for
>>> this?

>> Up sampling, other than that which happens implicitly in modern DACs in
>> order to facilitate digital filtering, is yet another audiophile myth.

> This seems to be a common reaction of people who have limited experience
> in
> this area.

The idea that upsampling could possibly make a improvement is a mystical
belief that sufficient education in how digital audio works can easily
dispel.

That's what information theory says, and that is what suitable experiments
and listening tests can demonstrate.

Once you sample a signal, upsampling can add no more information or relevant
detail. It only spreads the same information across more samples.

Edmund[_2_]

December 15th 11, 12:02 PM

On Thu, 15 Dec 2011 00:43:40 +0000, Audio Empire wrote:

> On Wed, 14 Dec 2011 07:25:06 -0800, Edmund wrote (in article
> >):
>
>> On Wed, 14 Dec 2011 00:45:05 +0000, Audio Empire wrote:
>>
>>> On Tue, 13 Dec 2011 04:12:44 -0800, Edmund wrote (in article
>>> >):
>>>
>>>> On Tue, 13 Dec 2011 00:08:09 +0000, Audio Empire wrote:
>>>>
>>>>> On Mon, 12 Dec 2011 03:35:39 -0800, Edmund wrote (in article
>>>>> >):
>>>>>
>>>>>> On Sun, 11 Dec 2011 22:08:58 +0000, Chuck Finley wrote:
> <snip>
>>>
>>> It's neither new or misleading. the discrimination between "lossy" and
>>> "lossless" compression, means exactly what it says.
>>
>> I am sure you understand that this addition -lossy vs lossless- is a
>> result of the misleading term "compression" for something that isn't
>> compression but reduction.
>
> Lossy compression can be said to be "reduction" (the "standard' music
> compression data rate in MP3 is 128 KBPS is 11:1. That means that more
> than 90% of the original waveform is actually discarded). But Lossless
> schemes such as ALCS (Apple Lossless Compression Scheme) and FLAC (Free
> Lossless Audio Compression) are not "reductions" as the original
> waveform is preserved in the round-trip process.
>
>>> Lossy compression
>>> makes files smaller by discarding what someone (or something - such as
>>> an algorithm) has decided to be non-essential information. Lossless
>>> compression, OTOH means that the file has been made smaller by using a
>>> less verbose coding scheme of some type. An example of lossless
>>> compression would be the ZIP format on one's PC. If anything were
>>> missing on an expanded copy of a ZIP file of say, Photoshop, would
>>> mean that Photoshop would not and could not run.
>>
>> You got it!
>
> Yes, I do "got it." and I see no reason to play at semantic games. "A
> rose by any other name", and all that.
>
>>>> If you end up with less data ( or gas in another area ) then you did
>>>> not "compress" it, then you reduced the data by throwing away data
>>>> and information.
>>>
>>> That's why it's called "lossy compression". In gasses, compression is
>>> compression in it's purest form. Nothing is discarded, but the gas has
>>> been processed to take up less volume by eliminating the empty space
>>> between gas molecules. Often this results in the gas becoming a liquid
>>> (like with propane or LNG), but sometimes not. If you merely vent-off
>>> a volume of gas to make the remaining take up less space, that's not
>>> compression, that's merely reducing the volume (like filling a water
>>> bottle and letting the excess run down the drain). Now if you could
>>> throw away a certain volume of, say, propane, and still have the same
>>> amount energy in what's left as you did before the gas's volume was
>>> reduced, then that would be an analogy of digital compression.
>>
>> Indeed and physics say that is simply not possible, neither in gas nor
>> data files.
>
> Well it is possible in data files because we're not dealing with actual
> matter, we are dealing with information. It's nothing new. Let's say we
> have two secretaries, both taking dictation of a letter from a single
> speaker. One takes the dictation long-hand, and the other uses standard
> stenographer's shorthand. When the final letter is typed-up, both should
> be identical, but the long-hand dictation might be 5 or 6 pages long
> while the shorthand version might be two pages. Both say exactly the
> same thing and nothing has been lost but one form of coding the
> dictation is far less verbose than the other even though it is
> represented by far less data.

Don't stretch it so much, if one can restore the original message
it is compression, if something is lost or changed it is not
compressed.
>
>>> Remember
>>> in audio, we're don't measure the final sound in those terms. We
>>> measure the perception of that sound at our ear/brain interface.
>>> Remember, the file itself (whether it be a digital audio file or a
>>> record groove) is NOT the sound, it is merely a representation of that
>>> sound. If much of the original waveform has been discarded to make
>>> the digital file representing the audio take up less media storage
>>> space, and most of the listening audience doesn't perceive that
>>> anything is missing, then whatever compression scheme was used was
>>> successful.
>>
>> Not if the input file is different that the output file, it is not
>> compression.
>
> But compression is useless without a complementary de-compression. It's
> a round-trip process.

That is -of course- what I meant with the "output" file.
>
>>>> I suggested before not to call the MP3 reduction- "compression"
>>>> because it isn't. Lets call it what it is and forget this idiotic
>>>> "Lossy Compression" which is a contradiction in terminus and
>>>> "lossless compression", compression always was lossless per
>>>> definition and I mean the right definition not the raped one.
>>>
>>> You are talking apples and oranges. Language is a living thing. The
>>> day that we can't accommodate new meanings for existing words, is the
>>> day that the language starts to die. Might as well go back to
>>> Roman-era Latin...
>>
>> I do know that language is a living thing, and by all means change the
>> meaning of words to make it more clear for everyone. Do NOT change the
>> meaning of words to make it more complicated or misleading.
>
> I don't see that anyone has done that. Most people don't care what audio
> compression scheme is used for their music as long as it works and
> allows them to fit more music into a finite physical space - such as the
> fixed memory size on an iPod. Those of us who do care, seem to
> understand the concepts involved at least well enough to know the
> difference between lossy and lossless compression schemes.

At the time when Compression meant exactly that,: "compressing" and some
sneaky developers "sold" there reduction scheme as "compression" and with
that, they introduced a misleading and plain wrong term.
At the time the claim was made it was inaudible, inaudible by who and
played on what? later they improved there reduction scheme, why?
Unfortunately, most people didn't realize they where cheated this way
and that forced us to introduce another stupid term: "Lossless Compression "
which is pleonasm this vs "Lossy ( lousy ) compression" which is a contradiction
in terminus.

In the mean time more reduction schemes are used and it is getting unclearer
every time if a reduction scheme is used or compression.
DTS seems to use both versions in which - AFAIK- DTS HD is compression
and DTS is reduction, but I am not even sure about that.

>>
>> So let us audio lovers all use the proper terms from now on and
>> hopefully in a few years from now everyone forgot the stupid term Lossy
>> compression which is a contradiction in terminus, and use "reduction"
>> or "lousy compression" for reduction schemes.
>
> I'm sorry, I simply don't see a problem here.
>>
>>
>>>> Using the proper terms for things makes it easier to understand.
>>>
>>> Understanding context can also help make words easier to understand.
>>
>> You know very well I do understand the context perfectly.
>
> That's not my point. My point is that everyone involved with music
> compression algorithms from those constructing them, to the teenager
> ripping CDs to iTunes understands from the context what lossy and
> lossless compression schemes mean. There is simply no need to change the
> universally accepted nomenclature at this stage of the game. At least
> that's my opinion on the subject anyway.

My opinion is that the proper term "compression" shouldn't have been
compromised in the first place. Now there seems to be reduction schemes used
with a certain name ( DTS and others ) and sometimes they use Compression,
sometime reduction, that isn't making things easier and in this "High-End"
group people using MP3 tells me it is not so clear to each and everyone.
Spending multi thousands on your HiFi set and throwing away a part of the
recording right at the source doesn't seem so smart to me.
I have nothing more to add about this.

Edmund

Edmund[_2_]

December 15th 11, 12:02 PM

On Tue, 13 Dec 2011 19:20:52 +0000, dave a wrote:

> On 12/13/2011 4:12 AM, Edmund wrote:
>
>
>> That IS the speech defect I am talking about! If that IS the new
>> definition that definition is plain wrong and misleading. Never before
>> in no other branch compression ever meant throwing away data or
>> material.
>> If you end up with less data ( or gas in another area ) then you did
>> not "compress" it, then you reduced the data by throwing away data and
>> information.
>> I suggested before not to call the MP3 reduction- "compression" because
>> it isn't. Lets call it what it is and forget this idiotic "Lossy
>> Compression" which is a contradiction in terminus and "lossless
>> compression", compression always was lossless per definition and I mean
>> the right definition not the raped one. Using the proper terms for
>> things makes it easier to understand.
>>
>> Edmund
>>
>>
> So what would you call jpeg? Or mpeg?

I am sure you know they answer.
So you don't have to ask.
I call it just jpeg and Mpech which means ( pech ) bad luck.

Edmund

Mr. Finsky[_2_]

December 15th 11, 12:03 PM

On Dec 14, 6:37=A0pm, "Arny Krueger" > wrote:
> "Audio Empire" > wrote in message
>
> ...
>
>
>
> Reliable proof?
>
> The skeptical members of this forum should form a consortium to offer a
> signficiant cash reward for reliable proof.
>
> After all, it worked for the Great James Randi! =A0AFAIK, no cable snake =
oil
> artist or reviewer has ever even tried to collect the million =A0dollars.
>

I do recall reading that various audiophile writers have attempted to
test the Great James Randi and collect the million smackers. The
problem is that Randi avoided all contact with anyone trying to do a
test. Randi is wonderful about publicity but appears to avoid a
confrontation that might cost him money or demonstrate that he may be
as big a phony as his "magician" targets.

Personally, I know I can hear the difference between mass market junk
and "audiophile" gear. I do think that the law of diminishing returns
exists in the audiophile world. The improvements obtained by spending
more money become small, once you have reached a certain price point.

I examined and tested the difference between 192 .wma files and a aiff
files ripped through iTunes. The difference was quite minor. However,
the cost of hard drives is so low that saving space is not worth the
potential problems. Besides, ripping CD's through dbpoweramp to FLAC
is loads of fun and flexibility.

Edmund[_2_]

December 15th 11, 12:03 PM

On Thu, 15 Dec 2011 00:37:27 +0000, Arny Krueger wrote:

> "Audio Empire" > wrote in message
> ...
>> On Mon, 12 Dec 2011 16:07:13 -0800, Arny Krueger wrote (in article
>> >):
>>
>>> "Chuck Finley" > wrote in message
>>> ...
>>>> I bought an Escient music server several years ago when hard drive
>>>> space was
>>>> still relatively expensive. Most of my music on there is encoded at
>>>> 320 and
>>>> 192.
>>>
>>> Shouldn't be a problem.
>>>
>>>> Reviews of DACs typically discuss how they can improve the sound of
>>>> CD quality or hi-rez music,
>>>
>>> Generally, false claims. Audiophile myths. Dreams, not actualities.
>>> Blatantly false sales pitches. The results of sighted evaluations.
>
>> Not exactly true. While I agree that most DACs that are constructed
>> using IC
>> converters sound so much alike that the differences (if any) are
>> trivial, high-end DACs using discrete, proprietary circuitry not only
>> can sound better
>> than the mass-produced IC chip-based DACs, but they sound significantly
>> different from one another.
>
> Reliable proof?
>
> The skeptical members of this forum should form a consortium to offer a
> signficiant cash reward for reliable proof.
>
> After all, it worked for the Great James Randi!

You did? if you see him again say Hi to him, I love that man :-)

Edmund

Andrew Haley

December 15th 11, 04:25 PM

Edmund > wrote:

> At the time when Compression meant exactly that,: "compressing" and
> some sneaky developers "sold" there reduction scheme as
> "compression" and with that, they introduced a misleading and plain
> wrong term.

When was that time, though? In 1984 Terry Welch wrote in his classic
paper: "The data compression described by this model is a reversible
process that is different from other forms of 'data compression'
..... in which data is deleted according to some relevance criterion."
So it's clear that "lossy compression" was familiar at that time.

> At the time the claim was made it was inaudible, inaudible by who
> and played on what?

What claim was that? The people who designed perceptual coders did a
lot of blind testing, and in _Subjective Evaluation of
State-of-the-Art 2-Channel Audio Codecs_ (1999) its audibility was
studied in considerable detail.

Andrew.

Audio Empire

December 16th 11, 03:58 AM

On Thu, 15 Dec 2011 04:03:00 -0800, Mr. Finsky wrote
(in article >):

> On Dec 14, 6:37=A0pm, "Arny Krueger" > wrote:
>> "Audio Empire" > wrote in message
>>
>> ...
>>
>>
>>
>> Reliable proof?
>>
>> The skeptical members of this forum should form a consortium to offer a
>> signficiant cash reward for reliable proof.
>>
>> After all, it worked for the Great James Randi! AFAIK, no cable snake oil
>> artist or reviewer has ever even tried to collect the million dollars.
>>
>
> I do recall reading that various audiophile writers have attempted to
> test the Great James Randi and collect the million smackers. The
> problem is that Randi avoided all contact with anyone trying to do a
> test. Randi is wonderful about publicity but appears to avoid a
> confrontation that might cost him money or demonstrate that he may be
> as big a phony as his "magician" targets.
>
> Personally, I know I can hear the difference between mass market junk
> and "audiophile" gear. I do think that the law of diminishing returns
> exists in the audiophile world. The improvements obtained by spending
> more money become small, once you have reached a certain price point.

Unfortunately, when much of this stuff is subjected to a so-called
"bias-controlled" test (like a DBT or ABX test) in a lot of cases, these
differences disappear. In other words, the guy who, an hour ago, swore that
he could hear the difference between mass-market junk and "audiophile" gear,
couldn't tell which was which once he could no longer see what it was that he
was listening to in a DBT or ABX test. I know that this is fact in cables.
Cheap, throw-away RCA interconnects, for instance, sound no different from an
expensive multi-hundred (or even multi-thousand) dollar interconnect in a
double-blind test AND, they even measure EXACTLY the same. But people still
insist that they can hear the difference and the expensive cables always
sound better than the cheap ones! However once subjected to a DBT, suddenly
these cables sound exactly the same.

Recently I've been involved with group of other recording engineers and
recording enthusiasts in DBTs of DACs, What we have found is that essentially
all DACs built using IC D-to-A converters sound so much alike that in a blind
test, nobody could tell (most of the time) that the people conducting the
tests had switched from one converter to another! What we found was that a
$300 Musical Fidelity V-DAC sounded identical to a $1000 Benchmark DAC1, and
that the Benchmark sounded identical to a $1400 Antelope Zodiac +! Later we
contrasted the Antelope Zodiac + with a $6000 Weiss DAC202, and could not
tell the difference between them in any statistically significant way,
either. When we contrasted these IC-based DACs against cost-is-no-object
discrete component designs from MSB and dCS, for instance, we found it quite
trivial to tell the difference between the two expensive DACs and their
cheaper IC-based competition. Those who post here and believe that everything
sounds the same because electronics for audio is a "mature technology"
discredit this notion by stating that IC-based DACs HAVE to better than
discrete component units because it's easier to control the variables in an
integrated circuit design than it is to control them in a design built-up of
discrete components. They also say that the IC DACs are transparent, and that
this has been proven because some testers have daisy-chained DACs and ADCs
together and compared the results with only one conversion, this proving that
modern DACs add nothing to the sound.

What I have found is that IC-based DACs do is to homogenize the audio and
then preserve that homogenization through repeated conversions. I say that
because, compared to the dCS Debussy and the MSB DAC-IV, that's what IC-based
DACs all sound like - homogenized!

Now with amps, some sound better than others. I have a pair of Behringer
A-500 power amps. They're cheap ($200 for about 160 Watts/channel) and they
sound fine....... until you A/B them against a Krell S-300i! (150
Watts/Channel) then, even in a DBT, the Krell shows what it is made of. It
sounds much cleaner, much more musical with better dynamic contrasts, and
better sound-staging. Many of these improvements aren't immediately apparent,
and some only show-up with certain kinds of program material and certain
signal conditions, but they ARE there and do show-up in DBTs. In fact there
are certain circumstances where ALL amplifiers sound different from one
another, even similar amps. This comes down to things like power supply
design, but these differences do exist, and usually the audiophile gear
outperforms the cheap, mass market gear, even if they have similar specs (the
everything-sounds-the-same crowd is going howl at this statement!).

> I examined and tested the difference between 192 .wma files and a aiff
> files ripped through iTunes. The difference was quite minor. However,
> the cost of hard drives is so low that saving space is not worth the
> potential problems. Besides, ripping CD's through dbpoweramp to FLAC
> is loads of fun and flexibility.

FLAC and Apple Lossless both seem pretty flawless.

Arny Krueger[_4_]

December 16th 11, 11:24 PM

"Audio Empire" > wrote in message
...

> Unfortunately, when much of this stuff is subjected to a so-called
> "bias-controlled" test (like a DBT or ABX test) in a lot of cases, these
> differences disappear.

Why would that be unfortunate?

>In other words, the guy who, an hour ago, swore that
> he could hear the difference between mass-market junk and "audiophile"
> gear,
> couldn't tell which was which once he could no longer see what it was that
> he
> was listening to in a DBT or ABX test.

I've certainly seen that happen, a great many times. It first happened to me
after I built the first ABX box and did the first ABX test. I then saw it
happen to an ever-widening group of people. First close friends, and then
casual acquaintances, and ultimately people I never knew.

> I know that this is fact in cables.

It is a fact for an ever-increasingly large circle of equipment. However, it
will *never* be true for LP playback and analog tape.

> Cheap, throw-away RCA interconnects, for instance, sound no different from
> an
> expensive multi-hundred (or even multi-thousand) dollar interconnect in a
> double-blind test AND, they even measure EXACTLY the same.

Pretty much. I have measured subtle differences among interconnects.
Different interconnects of reasonable lengths also have small but clearly
measurable effects on some of the equipment that they are used with. I can
reliably measure these differences using test equipment based on the
single-chip mainstream ADCs and DACs that are commonly used these days.

>But people still
> insist that they can hear the difference and the expensive cables always
> sound better than the cheap ones! However once subjected to a DBT,
> suddenly
> these cables sound exactly the same.

No surprise. And, now that we have ADCs, DACs and other components that have
fewer measurable differences among them and as compared to the technical
ideal, than exist among some interconnects as they are commonly used...

Surprised?

BTW, these ADCs and DACs that perform technically better than some
interconnects as the interconnects are used, are all mainstream single-chip
devices, or even single chips with a goodly number of both ADCs and DACs on
the same chip.

I see no reliable evidence that *anybody* can reliably discern audible
differences among ADCs and DACs with better than +/- 0.1 dB response in the
usual audio band, that also have all spurious responses 100 dB or more down.
I hear occasional anecdotes, but they are so incomplete and vague as to be
IMO completely irrelevant to any reasonable discussion. Please prove me
wrong!

Sebastian Kaliszewski

December 16th 11, 11:27 PM

Audio Empire wrote:

> Those cheap IC-based DAC chips MIGHT be sonically transparent to the
> Hoi-Polloi, but they sure don't sound as good, or image as well as a good,
> discrete component DAC such as those used in the MSB DACIV or the dCS
> Debussy. This has been noted in DBTs against a number of other DACs using
> IC-based Burr-Brown (TI), AMD and ESS DAC chips from manufacturers such as
> Benchmark, Antelope, Musical Fidelity, Music Streamer, Cambridge and Weiss,
> to name a few and the differences can be measured and easily seen. Look at
> the 1/3-octave spectrum-with-noise data, or the Intermodulation spectrum
> plots or the high-resolution jitter spectrum data and contrast the results of
> the IC-based DACs with those from some of these discrete component units such
> as the MSB and dCS units mentioned above. Their superiority is as easy to see
> as it is to hear.
>
> I suspect that the DBTs that show all modern DACs to be more-or-less equally
> transparent were comparing DACs using the more popular mass-produced
> integrated circuit DAC chips. Like I said in another post, only the ESS
> 32-bit "SabreDAC" has any real edge here, either sonically or by measurement,
> and it still doesn't sound or measure as good as the MSB proprietary "Ladder
> DAC" or the dCS "Ring DAC".

OK, where are those measurements of dCD or MSB devices? And how about
repeatability of those measurements?

>
> This type of "everything sounds the same" argument certainly makes audio
> cheaper. If everything sounds the same, then there's no reason to buy
> anything expensive. A $50 CD player sounds exactly like a multiple thousand
> dollar unit so all one needs to buy is the $50 player. All amplifiers sound
> exactly alike, so why buy a Krell integrated for $3000 when a $150 TEAC
> receiver from Costco performs exactly like it? It's tempting to believe this.
> Too bad that neither of these money-saving assumptions is true.....
>
> Take heart, though. A $5 Radio-Shack interconnect DOES sound exactly like a
> $4000 pair of Nordost Valhallahs and a hank of 14-gauge lamp cord does
> perform identically to a speaker cable from Oracle costing many hundreds of
> dollars per foot. Also, the IC processes keep improving and chips like the
> SabreDAC are closing-in on the cost-is-no-object designs, so there is hope
> for us financial mortals after all!

Sorry, but measurements of speaker wires differ much more than
measurements of decent DACs (if those DACs are considered neutral and
have measurements considered good).

As I wrote before -- those things do not add up.

Decent DACs are esiencially neutral down to -100dB or better (noise,
thd, nonlinearities, intermodulation, channel separation, etc). that
is down to 0.001% -- all that in range from 1 to 40000Hz.

Virtually no speaker wire is as flat (due to low - just few Ohms,
impdenace of load, even 0.1 Ohm wire impedance translates to much
greater distortions).

Thus, my questions: where are the measurements of this discrete
component DACs? How they were made and by whom...

rgds
\SK
--
"Never underestimate the power of human stupidity" -- L. Lang
--
http://www.tajga.org -- (some photos from my travels)

Audio Empire

December 19th 11, 03:42 PM

On Sun, 18 Dec 2011 06:37:06 -0800, ScottW wrote
(in article >):

> On Dec 15, 7:58=A0pm, Audio Empire > wrote:
>> On Thu, 15 Dec 2011 04:03:00 -0800, Mr. Finsky wrote
>> (in article >):
>>> On Dec 14, 6:37=3DA0pm, "Arny Krueger" > wrote:
>>>>
>>>> Reliable proof?
>>>>
>>>> The skeptical members of this forum should form a consortium to offer a
>>>> signficiant cash reward for reliable proof.
>>>>
>>>> After all, it worked for the Great James Randi! AFAIK, no cable snake oil
>>>> artist or reviewer has ever even tried to collect the million dollars.
>>>
>>> I do recall reading that various audiophile writers have attempted to
>>> test the Great James Randi and collect the million smackers. The
>>> problem is that Randi avoided all contact with anyone trying to do a
>>> test. Randi is wonderful about publicity but appears to avoid a
>>> confrontation that might cost him money or demonstrate that he may be
>>> as big a phony as his "magician" targets.
>>>
>>> Personally, I know I can hear the difference between mass market junk
>>> and "audiophile" gear. I do think that the law of diminishing returns
>>> exists in the audiophile world. The improvements obtained by spending
>>> more money become small, once you have reached a certain price point.
>>
>> Unfortunately, when much of this stuff is subjected to a so-called
>> "bias-controlled" test (like a DBT or ABX test) in a lot of cases, these
>> differences disappear. In other words, the guy who, an hour ago, swore that
>> he could hear the difference between mass-market junk and "audiophile" gear,
>> couldn't tell which was which once he could no longer see what it was that
>> he
>> was listening to in a DBT or ABX test. I know that this is fact in cables.
>> Cheap, throw-away RCA interconnects, for instance, sound no different from
>> an
>> expensive multi-hundred (or even multi-thousand) dollar interconnect in a
>> double-blind test AND, they even measure EXACTLY the same. But people still
>> insist that they can hear the difference and the expensive cables always
>> sound better than the cheap ones! However once subjected to a DBT, suddenly
>> these cables sound exactly the same.
>
> Except in single ended systems. Some cheap interconnects have such a
> high resistance on the ground/return that system noise is elevated.
> No need for an interconnect to be expensive to deal with this but some
> of the stock cheap stuff is simply too poorly made.
>
> Jensen has a couple of excellent articles that touch on the subject.
> This is just one.
>
> http://www.jensen-transformers.com/an/generic%20seminar.pdf
>
> ScottW

I've never run across any interconnects like the ones you mention and the
ones Mr. Whitlock mentions in his paper, but they could exist, I guess. He
is, however, very correct when he says that the advertising hype about
"transmission-line theory" often used by cable manufacturers to justify their
high-prices, is a lot of hooey and that at audio frequencies,
transmission-line effects don't come into play until the average audio
interconnect is more than FOUR THOUSAND feet long!

There are a number of things that the average audio enthusiast and measure
quite inexpensively if he so wishes. One can go on E-bay and buy a used
function generator for just a few bucks, likewise a used oscilloscope with a
6 MHz bandwidth can be had very cheaply as well. With these two instruments
(and some cable adaptors) it is possible to test some of these cable claims
quite thoroughly. With the o'scope and the signal generator set to sine wave,
it is possible to test the cable's bandwidth and its frequency response. This
will tell you if the manufacturer is "cheating" by putting outside components
on his cables to alter their frequency response and therefore cause them to
sound "different" from their competition. You can also check their bandwidth.
I've never measured any audio interconnect of a meter or less that wasn't
dead flat from DC to at least 1 MHz! Next, using the function generator
switched to "square wave" output and the oscilloscope again, it is possible
to look at a cables' square-wave response at 10 Hz or 1 MHz, and anything in
between. No interconnect that I've tested has shown anything but a perfect
square wave at any frequency up to one MHz, which is plenty for any audio
component. Square wave shape shows bandwidth limitation, excess capacitance
and inductance, as well as distortion and induced noise. It doesn't take many
such tests of high-end vs cheap cables to see that there is no measureable
difference between any of them.