Stereophile -- LP vs CD
John Atkinson
I was intrigued to read the argument pro and con LP versus CD. In the
March issue of Stereophile, you will be able to read a blow-by-blow
account of the mastering of the LP release of our Liszt piano sonata
recording. The CD was released last year at HI-FI '96, and was
noiseshaped from the master's 20-bit resolution to the 16 bits of the
CD standard using the Meridian 518. The LP was cut by veteran
mastering engineer Stan Ricker from the 20-bit master (played back on
a Nagra-D and decoded to analog with a 20-bit Mark Levinson No.30.5
D/A processor) without any compression or equalization (other than
RIAA). (The LP is cut "hot" and has high levels of vertical modulation
due to the master's significant random-phase content at low
frequencies, but my Linn Arkiv will track it at 1.9gm downforce.)
In careful level-matched listening tests, neither the CD nor the LP
were found to sound identical to the mastertape. As Larry Archibald
describes in the February's issue's "Final Word" column: "The master
tape is effortless -- the sound just happens, without trying hard,
ineffably."
However, it was clear that the noise-shaped CD was closer in specific
"hi-fi" terms---frequency range, noise floor, distortion---to the
20-bit master tape. Yet the LP, while sounding less like the master,
better preserved the "wholeness" of the soundstage. More importantly,
the LP better preserved the ability to envelop the listener in the
music. This was illustrated by the fact that listeners tended to talk
more during the CD playback, while with the LP, they got absorbed in
the music and resented being switched back to the CD.
So if all conventional measurements and even individual subjective
observations on individual aspects of the sound -- bass extension,
clarity, HF cleanliness, etc. -- indicate that the noiseshaped CD has
better "fidelity" to the _sound_, how can the less "accurate" LP be
closer to the _music_? Is it just the addition of groove noise, pre-
echo, and distortion, as has been proclaimed by some of the denizens
of rec.audio.high-end? Is the LP doing something, as of yet unknown,
right? Or is the CD doing something, as of yet unknown, wrong?
Don't get me wrong. I am very proud of the sound of our Sonata CD, and
if you haven't heard the 20-bit master, you don't know what you are
missing. But the LP gets the listening experience right in a way that
the CD only hints at. A paradox!
BTW, Stereophile's investment in producing the LP, to be available in
mid-February, was a hefty five-figure sum. It will be interesting to
see whether we get our money back on a medium that I wonder if many
audiophiles might just be paying lip service to.
John Atkinson, Editor, Stereophile
Stewart Pinkerton
John Atkinson <7447...@CompuServe.COM> writes:
>Don't get me wrong. I am very proud of the sound of our Sonata CD, and
>if you haven't heard the 20-bit master, you don't know what you are
>missing. But the LP gets the listening experience right in a way that
>the CD only hints at. A paradox!
>
>BTW, Stereophile's investment in producing the LP, to be available in
>mid-February, was a hefty five-figure sum. It will be interesting to
>see whether we get our money back on a medium that I wonder if many
>audiophiles might just be paying lip service to.
I can't speak for anyone else, but I will certainly buy both LP and CD
just out of curiosity, since I would in general disagree with your
assessment. Perhaps the replay system is relevant.
--
Stewart Pinkerton | If you can't measure what you're making,
A S P Consulting | how do you know when you've got it made?
(44) 1509 880112 |
"I canna change the laws o' physics" - the other Scotty
Kevin Connery
In article <5cg9ug$r...@eyrie.graphics.cornell.edu>, John Atkinson
<7447...@CompuServe.COM> wrote:
> So if all conventional measurements and even individual subjective
> observations on individual aspects of the sound -- bass extension,
> clarity, HF cleanliness, etc. -- indicate that the noiseshaped CD
> has better "fidelity" to the _sound_, how can the less "accurate" LP
> be closer to the _music_? Is it just the addition of groove noise,
> pre- echo, and distortion, as has been proclaimed by some of the
> denizens of rec.audio.high-end? Is the LP doing something, as of yet
> unknown, right? Or is the CD doing something, as of yet unknown,
> wrong?
One answer is obvious.
IF there's no perceptual bias based on outside prejudices, THEN it's
clear that 'all conventional measurements' are incomplete
I'm not even going to start to address the first one; it's flame-bait.
The second deserves to be investigated, but I'm not qualified in that
arena.
--
kdc
("For every complex problem, there is a straightforward, simple, and
obvious solution. Which is wrong.")
Bob Trosper
John Atkinson wrote:
....
> In careful level-matched listening tests, neither the CD nor the LP
> were found to sound identical to the mastertape. As Larry Archibald
> describes in the February's issue's "Final Word" column: "The master
> tape is effortless -- the sound just happens, without trying hard,
> ineffably."
....
> However, it was clear that the noise-shaped CD was closer in
> specific "hi-fi" terms---frequency range, noise floor,
> distortion---to the 20-bit master tape. Yet the LP, while sounding
> less like the master, better preserved the "wholeness" of the
> soundstage. More importantly, the LP better preserved the ability to
> envelop the listener in the music. This was illustrated by the fact
> that listeners tended to talk more during the CD playback, while
> with the LP, they got absorbed in the music and resented being
> switched back to the CD.
Mr. Atkinson -
Did you and the other listeners KNOW during the tests which source was
playing? If so, I find it hard to dismiss the effects of listener bias
on ANY of the conclusions you reached. And then there is the effect of
the additional gain stage required for the LP, the RIAA equalization
and the preference and training of each listener for each medium
(including listening to master tapes) which have no relation to
whether or not you know which source is playing.
And then there's the rather dodgy issue of whether the "20 bitness" of
the LP mastering process makes any difference. Was it too expensive to
cut any LP's using a "16-bit" master?
While your comparison is not apples and oranges, it may very well be
Navels and Valencias.
For the record, I do use CD and LP in my system and don't waste a lot
of time worrying about which one is better. Quite often the choice is
determined by the availability of the music I want on a particular
medium and, of course, the price.
I grew up listening to LP, but it was CD that got me into high-end and
then I discovered what really good LP playback could be. I now have
many more LP's than CDs because used vinyl is a much better deal and
has more music that I want (how old and stodgy I must be getting!).
-- Bob Trosper
Frank Gales
John Atkinson wrote:
>
> I was intrigued to read the argument pro and con LP versus CD.
> However, it was clear that the noise-shaped CD was closer in specific
> "hi-fi" terms---frequency range, noise floor, distortion---to the
> 20-bit master tape. Yet the LP, while sounding less like the master,
> better preserved the "wholeness" of the soundstage. More importantly,
> the LP better preserved the ability to envelop the listener in the
> music. This was illustrated by the fact that listeners tended to talk
> more during the CD playback, while with the LP, they got absorbed in
> the music and resented being switched back to the CD.
Wow, that's a very interesting posting for the beginning of another
war. But I find the point with the talking of the listeners very
interesting. I do have the feeling too, to get more involved in the
musical performance when listening to LPs. I must admit that my
turntable is of better quality than my CD-Player. It's for me like
the difference of life music and reproduction, or like a film in
cinema or on TV. When seeing a film on TV, I'm not so in the film as
in cinema. (I have not seriously tried out a home cinema installation
as I find the difference in visual and sonic virtuality rather large.)
Maybe it is a very interesting point for a discussion. But this
newsgroup might be too anonymous for such a discussion leading to
something else than flame wars.
[Moderator's note - Since we do our best not to allow anonymous
postings I'm a bit puzzled by the last sentence. -- bt]
Frank
--
---------------------------------------------------------------------
Frank Gales Erwin-Schroedinger-Strasse
Digital Systems and Data Processing D-67663 Kaiserlautern / Germany
Department of Electrical Engineering Phone: +49 631 205-2608
University of Kaiserslautern E-mail: ga...@rhrk.uni-kl.de
sdura...@aol.com
>The LP was cut by veteran
>mastering engineer Stan Ricker from the 20-bit master (played back on
>a Nagra-D and decoded to analog with a 20-bit Mark Levinson No.30.5
>D/A processor) without any compression or equalization (other than
>RIAA)
Good Lord, man, what's the point? Why would I, a die-hard analog fan,
have any interest in an LP derived from a digital master? Is it too
much to ask these days to record the original event simultaneously on
digital and analogue media and provide those of us with a preference
an entirely digit-free experience?
>Yet the LP, while sounding less like the master,
>better preserved the "wholeness" of the soundstage. More importantly,
>the LP better preserved the ability to envelop the listener in the
>music. This was illustrated by the fact that listeners tended to talk
>more during the CD playback, while with the LP, they got absorbed in
>the music and resented being switched back to the CD.
While I understand exactly what you're talking about, you are probably
aware that such behavior must be more vigorously tested before it's
accepted by many engineers. Which makes...
>So if all conventional measurements and even individual subjective
>observations on individual aspects of the sound -- bass extension,
>clarity, HF cleanliness, etc. -- indicate that the noiseshaped CD has
>better "fidelity" to the _sound_, how can the less "accurate" LP be
>closer to the _music_? Is it just the addition of groove noise, pre-
>echo, and distortion, as has been proclaimed by some of the denizens
>of rec.audio.high-end? Is the LP doing something, as of yet unknown,
>right? Or is the CD doing something, as of yet unknown, wrong?
...a moot question.
So the better question is: how do we measure, test and verify the fact
that certain listeners became disengaged from listening to the CD as
opposed to the greater apparent interest in the sound of the LP.
This, to me, is the greatest challenge facing audio.
Anthony Clarke
In article <5cj83m$t...@eyrie.graphics.cornell.edu>, Bob Trosper
<URL:mailto:rtro...@hpsrjtc.sr.hp.com> wrote:
Mr Bob Trosper wrote in an article regards Stereophile -- LP vs CD
[snip]
> Did you and the other listeners KNOW during the tests which source
was
> playing? If so, I find it hard to dismiss the effects of listener
bias
> on ANY of the conclusions you reached. And then there is the effect
of
> the additional gain stage required for the LP, the RIAA equalization
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
What about the gain stage in the rear end of a CD player :-)) As for
RIAA equalisation....lets not mention DACs', quantitisation, filtering
eg "brick wall filtering" and error correction which is all part of
the CD play back chain.
> And then there's the rather dodgy issue of whether the "20 bitness"
of
> the LP mastering process makes any difference. Was it too expensive
to
> cut any LP's using a "16-bit" master?
Maybe the article and test shoud be reappraised.... for example.. why
not also include an analogue tape master and mixing desk etc. for the
Mastering and subsequent comparisons !!
> While your comparison is not apples and oranges, it may very well be
> Navels and Valencias.
Yeah I fully agree with this, I recall a rather frightening experience
some years back... I had just upgraded to a Metaxas preamp Australian
made High End, with Conrad Johnson MV75 power amp, Goldmund Turntable
Souther SLA3 Tonearm and Gamma ClearAudio cartridge... first listening
set my mind racing....what the hecks going on.... the system sounded
very CD-ish!!!... obviously greater dynamic range was being heard off
the vinyl for the first time. But the air, spaciousness and
tangability of the performance itself was breath-taking.. I still
find this to be the forte of the analogue side of things...it FEELS
far more emotive than anything I have heard to date on offerings from
the digital medium.
> I grew up listening to LP, but it was CD that got me into high-end
and
> then I discovered what really good LP playback could be.
As you may deduce from the above the opposite holds true at this end.
--
Anthony Clarke | Oak Seed Computers Pty. Ltd. |
ant...@om.com.au | Phone / Fax (066) 246 200 |
Bob Myers
sdura...@aol.com wrote:
>> The LP was cut by veteran mastering engineer Stan Ricker from the
>> 20-bit master (played back on a Nagra-D and decoded to analog with
>> a 20-bit Mark Levinson No.30.5 D/A processor) without any
>> compression or equalization (other than RIAA)
> Good Lord, man, what's the point? Why would I, a die-hard analog
> fan, have any interest in an LP derived from a digital master? Is
> it too much to ask these days to record the original event
> simultaneously on digital and analogue media and provide those of us
> with a preference an entirely digit-free experience?
Well, why don't you wait and see what this LP sounds like?
Unfortunately, given that you now know that it's from a digital
master, you'll more than likely have some trouble with it. It would
certainly have been interesting to see how you liked the LP WITHOUT
KNOWING its source. Could you, in fact, have told whether or not you
were having a "digit-free" experience? The information John presented
here strongly suggests that you would not have objected, and that in
fact the behavior of the LP medium itself would once again deliver the
sound that you expect from it. In short, the fact that the original
source was digital appears to have had NO impact here - the listeners
still found differences between the CD and LP versions, which we can
only presume come from something inherent in the LP itself.
--
Bob Myers KC0EW Hewlett-Packard Co. |Opinions expressed here are not
O- Workstations Systems Div.|those of my employer or any other
my...@fc.hp.com Fort Collins, Colorado |sentient life-form on this planet.
Curtis Leeds
Bob Myers wrote:
>...the listeners
> still found differences between the CD and LP versions, which we can
> only presume come from something inherent in the LP itself.
Bob, just exactly who do you claim to be speaking for when you
say "WE can only presume...."? Certainly, it's not a presumption that
I'd be likely to make in the absence of some evidence.
Of course, you're free to presume anything that you like, Bob.
Just don't assert it as fact, and please don't claim to speak for
others.
--
********************************************************
Curtis Leeds cle...@mail.idt.net
"A man hears what he wants to hear, and disregards
the rest."
Arnold B. Krueger
John Atkinson <7447...@CompuServe.COM> wrote
> In careful level-matched listening tests, neither the CD nor the LP
> were found to sound identical to the mastertape. As Larry Archibald
> describes in the February's issue's "Final Word" column: "The master
> tape is effortless -- the sound just happens, without trying hard,
> ineffably."
There are a lot of very carefully controlled tests that are relevant
to this. One is of the basic process of taking a stereo master tape
and converting it to 44 kHZ 16 bit stereo sampled digital information
and converting it back again. This test showed that for a master tape
that its owner considered "technically challenging", the conversion
had no reliably detectable effect. Do you think that there was
processing other than the simple digital conversion, in the production
of the CD you evaluated, that explains what you heard?
> However, it was clear that the noise-shaped CD was closer in specific
> "hi-fi" terms---frequency range, noise floor, distortion---to the
> 20-bit master tape. Yet the LP, while sounding less like the master,
> better preserved the "wholeness" of the soundstage.
Its no secret among LP cutting engineers that the analog disk cutting
and playback process indroduces a large number of technical
inaccuracies. Many of these are inherently present to a degree that is
so gross as to be audible under any critical listening situation. Do
you think that these grossly audible distortions and degradations
increase your perceptions of the "wholeness"of the soundstage?
> More importantly,
> the LP better preserved the ability to envelop the listener in the
> music. This was illustrated by the fact that listeners tended to talk
> more during the CD playback, while with the LP, they got absorbed in
> the music and resented being switched back to the CD.
Was there any means by which the listeners could, by non audble means,
know the kind of source material they were listening to?
Its been my experience that due to the well known technical
limitations of vinyl disc recording and playback, that it makes
enjoying the music is a more tenious process. In other words, I have
to listen harder to get the same level enjoyment because the program
material has been through additional steps that arguably garble it. Do
you think that this may explain what you observed?
> So if all conventional measurements and even individual subjective
> observations on individual aspects of the sound -- bass extension,
> clarity, HF cleanliness, etc. -- indicate that the noiseshaped CD has
> better "fidelity" to the _sound_, how can the less "accurate" LP be
> closer to the _music_?
Obviously, it can't. And, there is nothing I see in your report that
proves that it is not.
> Is it just the addition of groove noise, pre- echo, and distortion, as
has been proclaimed by some of the denizens > of rec.audio.high-end? Is the
LP doing something, as of yet unknown, right? Or is the CD doing something,
as of yet unknown, wrong?
When you have two sources of the same information, one that is very
hard to distinguish from the original (digital audio done right), and
one that is easy to distinguish from the orginal (vinyl recording and
playback done as good as possible) then the answer to this question
seems evident.
> Don't get me wrong. I am very proud of the sound of our Sonata CD, and
> if you haven't heard the 20-bit master, you don't know what you are
> missing. But the LP gets the listening experience right in a way that
> the CD only hints at. A paradox!
Just a matter of euphonic distortion, and possibly operation of
listener expectation (since you have not disclosed whether the
evaluations were truely blind).
Were your evaluations truely blind?
Bob Myers
Anthony Clarke (ant...@om.com.au) wrote:
>> bias on ANY of the conclusions you reached. And then there is the
>> effect of the additional gain stage required for the LP, the RIAA
>> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>> equalization
>> ^^^^^^^^^^^^
> What about the gain stage in the rear end of a CD player :-)) As for
> RIAA equalisation....lets not mention DACs', quantitisation,
> filtering eg "brick wall filtering" and error correction which is
> all part of the CD play back chain.
I think that Bob brought up the "additional gain stage" of the LP
system to point out that NOT everything would be identical in the
analog pieces of the system - they CAN'T be, because of the need for
the extra stage to bring the LP's systems cartidge output up to "line"
level (whereas the CD system generally does this directly as a part of
the ADC operation). Bob's not trying to say (at least as I read this)
that the CD system doesn't have it's own unique aspects - it's just
that it's a bit more difficult to say that "outside of these few
things HERE, these two systems are identical".
Steven R. Rochlin
sdura...@aol.com wrote:
> Good Lord, man, what's the point? Why would I, a die-hard analog fan,
> have any interest in an LP derived from a digital master? Is it too
> much to ask these days to record the original event simultaneously on
> digital and analogue media and provide those of us with a preference
> an entirely digit-free experience?
Agreed. M A Recordings goes so far as to use totally different setups
for the analogue master and the digital master INCLUDING the
microphones used. It deeply saddens me to see a proclaimed
"audiophile" magazine use a digital master for their vinyl when they
have Mr Tracking Angle on board. FREMSTER!!!! WHERE ARE YOU?!?!?!
Please help us Obi-Wan Michael, you're our only hope!
Just for the record, i had the luxury to hear a TRUE engineer's master
tape of one of the passages on Stereophiles test CD. The difference
was more then subtle. Not just the difference between 20 bit and 16
bit. i mean, like a world of differences. He, too, is gravely
concerned. Please, please, pretty please JA, seek a professional
help.
Enjoy the music (original Everest promo pressing of "Scheherazade" on
vinyl right now),
--
Steven
Music lover, Head toilet bowl scrubber, workaholic, drivin' maniac,
Baron of Grey Matter, and just one kooky dude who's havin' a
goooood time as he travels through this time/space continuum.
http://ourworld.compuserve.com/homepages/enjoy_the_music/
John Ongtooguk
John Atkinson (7447...@CompuServe.COM) wrote:
: However, it was clear that the noise-shaped CD was closer in
: the master, better preserved the "wholeness" of the soundstage.
: the listener in the music....
Are you saying that the imaging was better, or at least more
pleasing on the LP ? I've noticed more than a few comments to that
effect when people are describing differences between CDs and LPs.
Some will say that if the imaging was more pleasing on the LP it's
just due to the spurious phase distortion inherent in the medium,
but others will suggest that for whatever reason LPs seem to
present a better, perhaps more accurate soundstage. It's a tough
question to resolve as unlike other measures of a sound system
there seems to be little agreement on whether imaging is even an
important attribute, much less one that should or could be done
well. Unlike most of us you're probably in a better position to
try to sort this one out, having access to a wide variety of
equipment, being involved in the recording process, and being
able to measure some characteristics of equipment. The subject
also touches upon the threads on speaker accuracy as it seems
that sound systems, especially speakers, that offer minimum
phase and/or a good step/impulse response seem to consistently
offer better imaging.
Assuming good recording technique it might be possible to
address the questions using combinations of LP vs CD and good
vs poor step response of the sound system, listening for the
combination that seems to offer the most accurate reproduction
of the recorded performance. It might also be easier to start
with some basic, simple rcordings instead of a musical
performance, perhaps someone playing different instruments at
different locations in an attempt to obtain a recording having
easy to distinguish cues of left/right, height, depth, as
well as the usual ones of how an instrument sounds.
I'm of the opinion that imaging is essential part of
reproduced music, regardless of how often poor recording
technique butchers it, and that for a sound system to be
considered as a candidate as an 'accurate' system it should
have a good step/impulse response in addition to the
generally agreed upon requirments. If the differences
between the LP and CD aren't really differences in imaging,
then excuse me for carrying on as so :^)
John Ongtooguk (jo...@vcd.hp.com)
Hamish Hubbard
John Atkinson (7447...@CompuServe.COM) wrote:
: Don't get me wrong. I am very proud of the sound of our Sonata CD, and
: if you haven't heard the 20-bit master, you don't know what you are
: missing. But the LP gets the listening experience right in a way that
: the CD only hints at. A paradox!
A paradox? That is a most interesting use of logical deduction :-)
Since you put so much time and expense into this test, why not round
it out by digitising the LP and putting it on CD? Then test the CD to
see how much of the character of the LP is retained.
FWIW, I find the encoding of a musical event as a stereo pair of
electrical signals fairly amusing, since these signals are so totally
incapable of containing enough information to reproduce physically the
musical event. Compared to that, arguing over whether to encode in
analog or digital seems insignificant. Spending thousands on hi fi
gear is highly amusing, to me, from this perspective. Why not buy a
nice instrument and learn to play it, for the same amount of money? A
'sound system' in its current form is never going to approach reality,
merely provide ear-candy. Nothing wrong with indulging, though :-)
Hamish
Curtis Leeds
Arnold B. Krueger wrote:
> Its no secret among LP cutting engineers that the analog disk cutting
> and playback process indroduces a large number of technical
> inaccuracies. Many of these are inherently present to a degree that is
> so gross as to be audible under any critical listening situation.
This is patently false. While the LP mastering and manufacturing
process is a veritable minefield, and while it is much easier to make
a bad LP than a good one, and while the process does potentially
introduce distortion, the claim that these flaws are "so gross as to
be audible under any critical listening situation" must be accompanied
by some documentation to be accepted. Clearly, there are a great many
listeners whose own listening experience is completely in conflict
with that statement. So, Mr. Krueger, perhaps you will identify the LP
cutting engineers who will support your statement.
> Its been my experience that due to the well known technical
> limitations of vinyl disc recording and playback, that it makes
> enjoying the music is a more tenious process. In other words, I have
> to listen harder to get the same level enjoyment because the program
> material has been through additional steps that arguably garble it.
Would you please identify what equipment you are using for LP
playback? The comment that you make here is the complete opposite of
what many - including myself - have discovered. Perhaps there is some
explanation for your dissatisfaction other than "inherent, well known
technical problems".
> When you have two sources of the same information, one that is very
> hard to distinguish from the original (digital audio done right), and
> one that is easy to distinguish from the orginal (vinyl recording and
> playback done as good as possible) then...
You've confused the use of an adverb with an adjective here, and
you've also confused what makes vinyl "as good as possible." So,
again, perhaps you'll explain what sort of LP equipment has led you to
these very dubious conclusions about the inherent weaknesses - and
their audibility - of LP. I'd also be interested to know about the
measurements that you've made on this equipment and how they support
your statements.
--
********************************************************
Curtis Leeds cle...@mail.idt.net
"A man hears what he wants to hear, and disregards
the rest."
********************************************************
Bob Myers
Curtis Leeds (cle...@mail.idt.net) wrote:
> Bob Myers wrote:
> >...the listeners
> > still found differences between the CD and LP versions, which we can
> > only presume come from something inherent in the LP itself.
> Bob, just exactly who do you claim to be speaking for when you
> say "WE can only presume...."? Certainly, it's not a presumption that
> I'd be likely to make in the absence of some evidence.
> Of course, you're free to presume anything that you like, Bob.
> Just don't assert it as fact, and please don't claim to speak for
> others.
Well, let's take a look at what we DO know for sure, and what might
reasonably be inferred from that.
1. The master in this case was a digital recording.
2. This same master was used to produce both LP and CD recordings.
3. The LP is reported to sound different than the CD, and in a manner
consistent with the typical report of "LP sound". Further, the
listeners expressed a preference for the LP version which was also
consistent with such reports.
4. We were also told that, in terms of all objective tests performed,
the signal produced by the CD was the more accurate in comparison
with that from the master. This can be stated in other terms which
some will no doubt object to, but I see no way around it: the LP added
more distortion to the signal than did the CD.
From this, I don't any way around at least the following conclusions:
- The preference for the LP cannot in this case be due to something
inherently "wrong" about the basic notion of digital recording. The
signal on the LP came from a "digital" source in the first place, and
if the LP IS the accurate medium that many claim, any faults which have
to do with simply "being digital" would have to be preserved.
- The only real difference between the digital master and the final CD
is the number of bits per sample. Therefore, any problems with the sound
of the CD have nothing to do with the sample rate, which is the ONLY other
possible variable in dealing with digital data. This assumes that the
CD produced WAS an otherwise faithful representation of the data, but if
that were NOT the case, then something grossly wrong happened in the
process of making the CD, and the test is invalid anyway. The only other
possible channel for difference is in the manner in which 20 bits/sample
became 16, but we have been told - and I must accept, in the absence of
evidence to the contrary and the fact that it's something so damned simple
to do - that this was done properly. This conclusion is supported by the
evidence given, in that the CD output was the more objectively accurate
of the two. Therefore, I seriously doubt that the preference for the LP
was due to an inadequate number of bits/sample in the CD format.
- Given the above, and from #3 and #4, the most logical hypothesis
appears
to be that the preference for the LP sound comes from qualities inherent
in the LP medium, which could best be characterized as "euphonic"
distortions. A second possibility exists - that the preference was due to
the effect of some other component in the signal chain which was NOT
identical between the two - in which case the test says nothing about the
inherent superiority of either medium, and most certainly does NOT imply
that the LP is "more accurate".
If you care to offer an alternative hypothesis, Curtis, I'm all ears.
Scott Drysdale
In article <5cjb9k$2...@canyon.sr.hp.com>, sdura...@aol.com wrote:
>In article <5cg9ug$r...@eyrie.graphics.cornell.edu>, John Atkinson
><7447...@CompuServe.COM> writes:
>
>>mastering engineer Stan Ricker from the 20-bit master
>
>have any interest in an LP derived from a digital master? Is it too
>much to ask these days to record the original event simultaneously on
>digital and analogue media and provide those of us with a preference
>an entirely digit-free experience?
good point. it would be most enlightening to have four playback media:
MASTER PLAYBACK
1) 20-bit 16-bit
2) 20-bit LP
3) analog 16-bit
4) analog LP
then you could compare any two of the above to see how the master or
the playback medium contributed to the sound.
>>Yet the LP, while sounding less like the master,
>>better preserved the "wholeness" of the soundstage. More importantly,
>>the LP better preserved the ability to envelop the listener in the
>>music. This was illustrated by the fact that listeners tended to talk
>>more during the CD playback, while with the LP, they got absorbed in
>>the music and resented being switched back to the CD.
>
>While I understand exactly what you're talking about, you are probably
>aware that such behavior must be more vigorously tested before it's
>accepted by many engineers.
yep.
1) there's no mention of what level of involvement the listeners
exhibited when listening to the 20-bit master (the "control").
2) we don't know if they did the LP first, and listeners were simply
bored/hungry/etc. by the time the CD was being played.
3) we don't know if the listeners were aware of which was being played.
4) since they were apparently able to talk to each other, it's hard to
imagine the listeners independently reached their level of involvement.
once one person starts talking, everyone's pretty much distracted.
hardly what i'd call a good environment for a listening test.
>So the better question is: how do we measure, test and verify the fact
>that certain listeners became disengaged from listening to the CD as
>opposed to the greater apparent interest in the sound of the LP.
>This, to me, is the greatest challenge facing audio.
stuff like the above would help.
// Scott Drysdale // Visual Networks Inc // Software Engineer
// sdry...@visual.mctec.com // 60 HD FLH & 57..69 HD FLH // AMIGA!
John Busenitz
On 28 Jan 1997, Bob Myers wrote:
> Well, why don't you wait and see what this LP sounds like?
> Unfortunately, given that you now know that it's from a digital
> master, you'll more than likely have some trouble with it. It would
> certainly have been interesting to see how you liked the LP WITHOUT
> KNOWING its source. Could you, in fact, have told whether or not you
> were having a "digit-free" experience? The information John presented
> here strongly suggests that you would not have objected, and that in
> fact the behavior of the LP medium itself would once again deliver the
> sound that you expect from it. In short, the fact that the original
> source was digital appears to have had NO impact here - the listeners
> still found differences between the CD and LP versions, which we can
> only presume come from something inherent in the LP itself.
Never thought I'd say this, but actually, the presumed "fact" (it is
not really a fact, since the differences were perceived during sighted
and thus biased listening) that there are differences between the CD
and LP versions means that either or both the CD and the LP are
different from the master. The difference *could* be that from going
from the 20-bit master to the 16-bit CD.
However, since it is very unlikely that the recording used all of the
120 dB dynamic range of the master and the listening room in which the
CD and LP versions were compared was sufficiently quiet to make this
dynamic range audible, the slight reduction in dynamic range (the only
change from the 20-bit master to the 16-bit CD, unless the noise
shaping audibly modifies the noise floor, as Apogee suggests) isn't a
factor.
The "better imaging, etc" of the LP can most likely be attributed to
its "phasiness", and possibly even the surface noise, which
subjectively adds a HF "ambience".
_____________________________________________________________
John Busenitz buse...@ecn.purdue.edu
P.U. ECE http://cernan.ecn.purdue.edu/~busenitz
Disclaimer: My statements do not represent Purdue University.
Jim Andrews
Steven R. Rochlin wrote:
[snipping sduraybito's comment about not wanting Stereophile's
LP that was cut from a 20-bit digital master]
> Agreed. M A Recordings goes so far as to use totally different setups
> for the analogue master and the digital master INCLUDING the
> microphones used.
An interesting bit (pun intended) of anecdotal evidence that will
probably only fan the flames of the "_digital_ is missing something --
no, _analog_ is missing something" arguments.
> It deeply saddens me to see a proclaimed
> "audiophile" magazine use a digital master for their vinyl when they
> have Mr Tracking Angle on board. FREMSTER!!!! WHERE ARE YOU?!?!?!
> Please help us Obi-Wan Michael, you're our only hope!
I have read one of Mr. Fremer's columns, and all I can say is that
analog lovers have already lost the battle if he's their "leader". In
my opinion, the man is delusional. But I should add that I, an
engineer who honestly prefers most digital to most analog, would have
been interested to see what would have happened had they cut the CD
and LP from an ANALOG master. That way, the analog lovers would have
had no argument if/when the CD came across as indistinguishable from
the master, but the LP had more "pace and microdynamics", or whatever
they call it all.
> Just for the record, i had the luxury to hear a TRUE engineer's master
> tape of one of the passages on Stereophiles test CD. The difference
> was more then subtle. Not just the difference between 20 bit and 16
> bit. i mean, like a world of differences. He, too, is gravely
> concerned. Please, please, pretty please JA, seek a professional
> help.
Well, if you bring this up, you had better be prepared to explain
yourself in more detail. Did you compare the digital master to the CD
1. in the same system (presumably a recording studio)?
2. with high-quality digital tape and CD players?
2. at the same volume?
3. (you guessed it) blind?
I'd say it's completely irrelevant that you heard a difference if any
of these statements are false.
Going back up the thread, John Atkinson's original post states:
> In careful level-matched listening tests, neither the CD nor the LP
> were found to sound identical to the mastertape . . .
> However, it was clear that the noise-shaped CD was closer in specific
> "hi-fi" terms---frequency range, noise floor, distortion---to the
> 20-bit master tape. Yet the LP, while sounding less like the master,
> better preserved the "wholeness" of the soundstage. More importantly,
> the LP better preserved the ability to envelop the listener in the
> music. This was illustrated by the fact that listeners tended to talk
> more during the CD playback, while with the LP, they got absorbed in
> the music and resented being switched back to the CD.
Alright, we have an admission that the CD gets at least some aspects
of the recording better than the LP. To mitigate this horrible turn
of events, JA then refers to what is possibly a completely
unscientific test (other than level-matched) with unspecified
listeners and their associated biases, and claims a victory for LP
because they TALKED MORE during the CD? Crimony, guys . . . are you
that stupid, or do you think WE'RE that stupid? I'm scared to learn
the answer to that one.
I'd like to also add that, in terms of scientific validity, this
"test" no more proves that CD has better frequency response, etc.,
than it proves the subjective statements that follow. It seems to be
beyond some audiophile's comprehension that you can actually perform
tests that PROPERLY account for and/or determine listener preference,
and it appears that time and again, these tests are eschewed in favor
of anecdotal evidence that supports one particular camp over the
other. I'll tell you what, I am not so insecure as to be afraid of my
own preference. Why do digiphobes insist on chasing their tails to
avoid accounting for their own preferences?
jim andrews
John Busenitz
On Tue, 28 Jan 1997, Steve Zipser (Sunshine Stereo) wrote:
> No, the Velodyne is not serious because the amp/xover isn't seperate.
Well, and to think I thought what made a subwoofer serious is their
low-frequency cutoff, flat response, low distortion, and output
capability. So it's the old "separates" thing that makes a difference.
What makes a subwoofer "not serious" when the amp/filter is designed
into the subwoofer itself?
Kevin Connery
In article <5cmeg2$3...@canyon.sr.hp.com>,
Bob Myers <my...@hpfcla.fc.hp.com> wrote:
>4. We were also told that, in terms of all objective tests performed,
> the signal produced by the CD was the more accurate in comparison
> with that from the master. This can be stated in other terms which
> some will no doubt object to, but I see no way around it: the LP added
> more distortion to the signal than did the CD.
One minor, but critical point: added more *measurable* distortion.
It's not as though we do measure everything important, so the
distinction is important. It's possible, for example, that the
'distortion' which occured during the transfer from 20-bit digital is
less than that going from the 20-bit master to analog. I'm not saying
it WAS, I'm noting the possibility you preclude. (I'm inclined to
consider downsampling to be distortion, though it's often a
handwavable type.)
From this, I don't any way around at least the following conclusions:
>- The preference for the LP cannot in this case be due to something
> inherently "wrong" about the basic notion of digital recording. The
> signal on the LP came from a "digital" source in the first place, and
> if the LP IS the accurate medium that many claim, any faults which have
> to do with simply "being digital" would have to be preserved.
Probably true. I do note that most analog fans aren't anti-digital so
much as anti-the-current-digital-implementation -- there's a big
difference.
--kdc
Arnold B. Krueger
John Ongtooguk <jo...@vprme3.vcd.hp.com> wrote in article
> Are you saying that the imaging was better, or at least more
> pleasing on the LP ? I've noticed more than a few comments to that
> effect when people are describing differences between CDs and LPs.
> I'm of the opinion that imaging is essential part of reproduced
> music, regardless of how often poor recording technique butchers it,
> and that for a sound system to be considered as a candidate as an
> 'accurate' system it should have a good step/impulse response in
> addition to the generally agreed upon requirments.
The word "imgaging" is hard to pin down. Claims of improved imaging
have circulated for years and are almost as old as high end itself.
The word imaging suggests the painting of a sonic picture, a sort of 2
or 3 dimensional effect. In that context, imaging would be most
dependent on the techniques used to record original program material.
During reproduction, a sonic image could become smeared if anything
changed the phase or amplitude relationships in the signal for each
channel. Additional smearing can be expected if the phase or
amplitude relationships were changed between the channels. The major
source of sonic smearing in most systems are reflections and
inconsistencies in the acoustics of the listening room.
From a psychoacoustic standpoint, the ear is fairly insensitive to
changes in phase between various components of a monophonic
signal. Introducing, say, 1,000 degrees of phase shift to all signals
above 400 or 1000 Hz is hard to detect, even when listening through
headphones.
Relative phase changes between the channels are well known to be far
more obvious. Anybody who has had the misfortune to listen to a stereo
with one channel 180 degrees out of phase with the other (reversed
polarity) knows about that.
I don't know how well it is known, but the process of tracing grooves
and problems within phono cartridges can create measurably large phase
and amplitude variations between the channels. One amazing demo
involves playing a monophonic signal (both channels recorded as equal
as possible) and viewing the cartridge output in X-Y mode on an
oscilliscope. Idealy, a line of constant hight and angle would
result. In the real world, neither the angle of the line, nor even the
display of a line is guraranteed, or perfectly stable, even at mid
band. Variious loops and curliques are not unusual, and the whole
display may shift as the record turns, even when low frequencies are
cut off sharply.
In its early issues, The Absolute Sound gave a glowing review to the
ADC XLM cartridge. By cause or coincidence, its a fact that the XLM
had a flawed magnetic design that resulted in considerable amplitude
modulation of the signal due to minor variations in the distance
between the cartridge body and the groove.
If a record was not perfectly concentric and flat, the XLM would
modulate the channels once per revolution, to the extent of several
dB's. Other vendors, such as Shure were long aware of these effects
and made a policy of eliminating them from their product during the
design stage.
Yet, the XLM was cherished by many for its "imaging". Its open to
speculation whether the "imaging"effects XLM owners cherished were
just locally generated amplitude modulation effects - flaws that
detracted from ideal reproduction of the recorded sound.
There are many opportunities for euphonic coloration in phono systems.
Every tone arm has a major resonance in the 5-25 Hz range based on its
mass and the compliance of the cartridge. Some tone arms attempt to
damp this resonance with flexibly mounted counterweights, and others
don't even try. The results, in any case, are far from perfect
supression of this major resonance which can peak above 10 DB. Is this
the source of the "warmth" mentioned by Paul Schaeffer on the
Letterman show last Monday night?
There is another major resonance in phono systems due to the
compliance of the vinyl and the mass of the stylus assembly. In moving
magnet cartridges, designers often try to reduce it (with pretty fair
success) by resonating the inductance of the pickup coils and any
capacitance in the tonearm and preamp wiring. Due to the small number
of turns of wire and therefore low inductance, this opportunity is
lost for most moving coil cartridges, and large resonances in the
15-45kHz range result. While 45kHz is supersonic, a large moderately
damped resonance there can have measurable (and audible) effects at 15
kHz and below.
If you apply these same criteria to most CD players, you find that
within the normal audible range, they are just not there. And, unlike
the high frequency resononace of most phono cartridges, the low pass
"stonewall" fliters of CD players are carefully designed to not
impinge on response even a few kHz below.
My wife asks me whether you can't just adjust your equalizer to get
these kinds of effects with CD's, and the answer is largely yes - if
your system has an appropriate equalizer. The good news is that
equalizers don't wear out significantly every time you play them, and
if you get tired of one type of equalization, or it becomes
inappropriate with changes in program material, you can just turn some
knobs.
Steven R. Rochlin
> Well, if you bring this up, you had better be prepared to explain
> yourself in more detail. Did you compare the digital master to the CD
>
> 1. in the same system (presumably a recording studio)?
Yes, in the same system. The guys an audio writer/reviewer. He's
from the school of measurements, whereas i'm from the school of
"subjectiveness" (uh oh).
> 2. with high-quality digital tape and CD players?
His CD players wasn't a Levinson combo (thank goodness). My humble
apologies but i forget the exact make-model. He bought it becuase he
told me it measured very well.
> 2. at the same volume?
Look, the difference was SO big that that didn't matter. Though i
might add TDK hired this dude for the $1,000,000 difference giveaway
at the 96 WCES. If ANYONE is familiar with matching levels, etc, he's
among them. At this dude's place i was at is among the King of "same
volume level" "ABX", measure it to death, damn the torpedoes, full
steam ahead.
> 3. (you guessed it) blind?
Again, the differences were THAT dramatic. And although i was not
blind, he did all the stuff behind my back. THIS is why i choose to
not post much here on rec.audio.high-end. If you want ABX,
double-blind... Please seek another person. For if YOU could only
hear how VASTLY DIFFERENT it was between the CD and the master DAT.
And YES, i AM familiar with the basic differences between 20 bit DAT
and 16 bit seedie, er, um, CD.
> I'd say it's completely irrelevant that you heard a difference if any
> of these statements are false.
And i'm ok with that. Really i truly am. There's a saying "The whole
point of Camp is to dethrone the serious" ---Susan Sontag
> Alright, we have an admission that the CD gets at least some aspects
> of the recording better than the LP. To mitigate this horrible turn
> of events, JA then refers to what is possibly a completely
> unscientific test (other than level-matched) with unspecified
> listeners and their associated biases, and claims a victory for LP
> because they TALKED MORE during the CD? Crimony, guys . . . are you
> that stupid, or do you think WE'RE that stupid? I'm scared to learn
> the answer to that one.
Hmmm... maybe JA has a very good point here. Would you rather have
"sound" or "enjoy the music"? Neither is "right" nor "wrong". We all
have our preferences in life. Like good art, what makes a peice of
art good to you? And does you're preferences in art agree with
everyone elses?
> Why do digiphobes insist on chasing their tails to
> avoid accounting for their own preferences?
Personally, i enjoy vinyl, SE tube amps, and as i sit here i also...
Enjoy the music (Rush "Grand Designs" right now... on vinyl),
Steven
"So much poison in power, the principles get left out
So much mind on the matter, the SPIRIT gets forgotten about"
--
John Hamm
Bob Myers wrote:
>
> Curtis Leeds (cle...@mail.idt.net) wrote:
> > Bob Myers wrote:
> > >...the listeners
snip,snip,snip ~ all over the place
I think the following is 'misquoting' the original - probably
unintentionaly.
Wasn't the original statement that the CD was more accurate in a Hi-Fi
sense; frequency response, signal to noise, etc.? That doesn't equate
to 'musically' accurate. Now before you aim-the-flame consider that a
system can "measure good, sound bad" and vice-versa.
Arthur LeGrand Shapiro
Hamish Hubbard pontificates:
--> Spending thousands on hi fi
--> gear is highly amusing, to me, from this perspective. Why not buy a
--> nice instrument and learn to play it, for the same amount of money?
Tell me this is a bad joke!
Well, a good Bosendorfer concert grand is what, these days- $175K? I
wish being a musician was simply a matter of buying the instrument; if
it were, I wouldn't be slaving away here in the office. You think
it's fun here, with the guy out in the hallway rythmically banging a
huge drum and yelling "Code! Code! Code!"
The fact of the matter is that the normal human cannot play the piano
well; one has to be an anatomical freak to do such things as left-hand
4th and 5th finger trills at 20 per second, clocklike runs up and down
the keyboard, and the like. I obviously cannot speak for anyone else,
but my fairly lavish system has to compensate for the reality that,
while I can and do enjoy reading and memorizing scores, I don't have a
bloody prayer of playing 95% of the material. Even if I could,
hearing others' interpretations would still be a worthwhile pursuit.
Friends not as fanatical about the piano as I talk about thingies
called "orchestras", which I understand to be a group of varied
instrumentalists playing together. Should we each buy one of each
instrument and multiplex them?
Art (feeling ornery today)
Arthur L. Shapiro Art...@mpa15c.mv.unisys.com
Rarely an official Unisys corporate spokesman, and certainly not now.
Jim Andrews
Steven R. Rochlin wrote:
> I (Jim Andrews) wrote . . .
>> 3. at the same volume?
> Look, the difference was SO big that that didn't matter.
>> 4. (you guessed it) blind?
> Again, the differences were THAT dramatic. And although i was not
> blind, he did all the stuff behind my back.
Fair enough, but there are those of us who won't give a second thought
about your "test results", inasmuch as you are a clearly biased
listener, who may or may not have an agenda, and who makes very little
attempt to control those biases. I really do support your preference
for LP and all the things it does right for you, but I don't trust for
a minute that there is any difference at all between the CD and 20-bit
master, just because you say so. You're not a "credible witness", as
they say, though you appear to be a passionate one. That's cool.
By the way, I'm not at all surprised that there's a difference between
a CD and its 20-bit master. Whether there's an AUDIBLE difference is
beyond my ability to judge unless I do the test for myself, but there
certainly should be a quantifiable difference. Also, these "tests"
really reveal very little about the capabilities of the two media (LP
and CD) unless they test "state of the art" versions of both. As has
been stated many times before, it's not hard to cut a bad LP or a bad
CD.
> Hmmm... maybe JA has a very good point here. Would you rather have
> "sound" or "enjoy the music"? Neither is "right" nor "wrong". We all
> have our preferences in life. Like good art, what makes a peice of
> art good to you? And does you're preferences in art agree with
> everyone elses?
Sure, we are all entitled to our preferences, and they will quite
obviously vary from person to person. The problem here is that John
Atkinson appears to be making a statement about LPs having some sort
of "musical accuracy" advantage over CDs, based on the results of a
very dubious experiment. The fact that I'd like to see the experiment
done in a more scientific manner has nothing to do with my feelings
about preference, and I resent your implications to the contrary. I
honestly don't see the point of this argument, and so I'll bow out. I
am clearly not helping matters any!
jim andrews
Steven Abrams
my...@hpfcla.fc.hp.com (Bob Myers) writes:
> - Given the above, and from #3 and #4, the most logical hypothesis
> appears
> to be that the preference for the LP sound comes from qualities inherent
> in the LP medium, which could best be characterized as "euphonic"
> distortions. A second possibility exists - that the preference was due to
> the effect of some other component in the signal chain which was NOT
> identical between the two - in which case the test says nothing about the
> inherent superiority of either medium, and most certainly does NOT imply
> that the LP is "more accurate".
Clearly, the LP is *not* more accurate.
However, Mr. Atkinson (who has a habit of throwing articles up here
without engaging in any discussion thereafter) left one question
rather open. Did the LP sound more "musical" than the 20-bit master?
If the "order" of preference is 20-bit Master, LP, then CD, how can
this be attributed to the euphonic distortion of the LP? That would
imply that the loss in S/N ratio in going from 20-bits to 16-bits is
more than compensated by the euphonic distortion of the LP. However,
I don't believe that this is the case.
He says that neither the 16-bit CD nor the LP are indistinguishable
from the original 20-bit master. Gabe has said, in the past, that a
properly dithered 16 bit from a good 20-bit ecording is virtualy
indistinguishable from the original. Is this indicative of a poor job
of doing the 20->16 bit conversion, of the sighted nature of Mr.
Atkinsons's listening "tests" or of one man's "virtually
indistinguishable" being another man's "close but no cigar."
I am leaving this question open because I do not have the answers.
Let me say this, however. Mr. Atkinson is in a very unique position,
having in his posession both 20-bit, 16-bit, and LP versions of the
same recording, having access to state-of-the-art playback equipment,
and an arsenal of people who love to write about audio and,
ostensibly, know how to listen.
How about setting up a blind or double blind listening test, as
Stereophile has done in the past for speakers. You can set all the
variables and avoid all of the things that the typical audiophile
hatses about these sorts of tests. The point is to see if, udner
blind conditions, the order of preference is the same. Is there a
distinguishable difference between the 16- and 20-bit master? Is the
LP really the best of the bunch? Let's find out.
~~~Steve
--
Steven Abrams abr...@cs.columbia.edu
Living is easy with eyes closed, misunderstanding all you see.
-Lennon/McCartney
Bob Myers
John Hamm (jlh...@lawrence.ks.us) wrote:
> Wasn't the original statement that the CD was more accurate in a
> Hi-Fi sense; frequency response, signal to noise, etc.? That
> doesn't equate to 'musically' accurate. Now before you
> aim-the-flame consider that a system can "measure good, sound bad"
> and vice-versa.
Yes, and that's the whole point - the sound that the LP-philes prefer
does NOT seem to be due to any increased OBJECTIVE accuracy on the
part of that medium, but rather due to the specific way in which the
medium introduces INaccuracies. In other words, the results aren't
all that surprising, if the notion that the LP can provide "euphonic"
distortions is accepted. And we have yet to see any evidence which
would suggest that this sort of distortion is NOT occuring.
--
Bob Myers
Kevin Connery (Ker...@cris.com) wrote:
> In article <5cmeg2$3...@canyon.sr.hp.com>,
> Bob Myers <my...@hpfcla.fc.hp.com> wrote:
> > the signal produced by the CD was the more accurate in comparison
> > with that from the master. This can be stated in other terms which
> > some will no doubt object to, but I see no way around it: the LP added
> > more distortion to the signal than did the CD.
> One minor, but critical point: added more *measurable* distortion.
> It's not as though we do measure everything important, so the
> distinction is important. It's possible, for example, that the
> 'distortion' which occured during the transfer from 20-bit digital is
> less than that going from the 20-bit master to analog. I'm not saying
What is often lost in these discussions is that, in discussing the
"objective measurements" form of "accuracy", we're talking about
electrical signals, not sound. If the CD has less measurable
distortion in this regard, then it has less distortion, PERIOD. We
may not be measuring everything to the degree you'd like in terms of
its effect on the final perception - it might be argued that this
CANNOT be done via electrical measurements only, since perception
occurs within the skull of the listener - but we most definitely ARE
measuring well enough to determine which of two electrical signals
most closely resembles a third.
It is not possible, given properly-performed measurements, that there
was actually less distortion in going to the analog format than in
going to the digital, AND have the distortion *measure* worse for the
analog format. You are effectively arguing that there is distortion
of the signal from the digital source which is greater than that from
the analog, and yet which could not be measured by the test equipment.
Given that the signal from the analog source showed distortion which
is clearly within the capabilities of the test equipment to detect,
this is simply not possible.
> it WAS, I'm noting the possibility you preclude. (I'm inclined to
> consider downsampling to be distortion, though it's often a
> handwavable type.)
First, the transformation from 20 bits/sample to 16 does not
constitute "downsampling", at least as I would commonly use that term,
since there is no change in the number of samples per second. Whether
or not all of the information in the 20-bit original is preserved in
the 16-bit version depends on whether or not there actually WAS
information, as opposed to noise, in those lower four bits. (Key
item: simply adding more bits/sample does NOT guarantee that you're
actually dealing with more INFORMATION.) If there was, then yes,
there MUST be distortion in making this transformation, since you have
lost information. But if not, then what the 20-bit master let you do
was work in a digital domain where the math errors won't make it into
those bits where the sound really lives. In short, the 16-bit version
would come out as faithful TO THE ORIGINAL SIGNAL as the 20-bit was,
since there was no real information beyond 16 bits anyway.
> From this, I don't any way around at least the following conclusions:
> >- The preference for the LP cannot in this case be due to something
> > inherently "wrong" about the basic notion of digital recording. The
> > signal on the LP came from a "digital" source in the first place, and
> > if the LP IS the accurate medium that many claim, any faults which have
> > to do with simply "being digital" would have to be preserved.
> Probably true. I do note that most analog fans aren't anti-digital so
> much as anti-the-current-digital-implementation -- there's a big
> difference.
Yes, but I would also submit that many of the supposed faults these
folks find with the "current digital implementation" actually have
very little to do with the fundamentals of the digital format itself.
John Busenitz
On 29 Jan 1997, Kevin Connery wrote:
> Probably true. I do note that most analog fans aren't anti-digital so
> much as anti-the-current-digital-implementation -- there's a big
> difference.
I politely disagree. From what I've seen, the majority of analog
fans harbor an intense distrust and dislike for all things digital,
probably because it is a "new" technology, and they don't understand
it. Witness people throwing about "infinite resolution" and "hearing
below the noise floor" without having a clue what these terms mean.
Now that concepts related to digital audio are becoming more well
known, people are slowly beginning to learn more about it and to
realize that digital audio doesn't "throw out information between
samples". But the distrust is still there, and is thrust upon the CD
medium as heavily as before, and "new digital formats" are now the
"savior", since people know that analog mediums are on their way out, no
matter how much Fremerites deny it. So now we have people like Scull
mindlessly droning the mantra "24-bit/96 kHz sampling" (he does it in
every issue of Stereophile; it's getting a bit old) without having a
clue as to whether it is necessary or better. And JA (though I don't want
to try to speak for him; correct me if I'm wrong) acknowledged in the
"Sonata" article that the noise-shaping "moves" the dither to higher
frequencies (around 20 kHz; I think lower, too) WHERE IT IS INAUDIBLE.
This (along with the vast body of evidence about human hearing limits)
indicates that sampling faster than the current rate does little other
than waste data space that could be better used for more channels: the
biggest problem with audio mediums are NOT the frequency response or
dynamic range, but the use of only two channels which is utterly
inadequate for proper spatial reproduction.
Which brings my to my second disagreement: there is not a "huge" difference
between CD audio and the "next digital medium". Recordings and listening
rooms probably don't have the dynamic range to take full advantage of
current CD technology, let alone 120 dB (or more). And the extra bandwidth
isn't needed; we can't hear that high, and even if we could, that little
bit wouldn't benefit audio reproduction to the extent more (well implemented)
channels would. Now, all this is NOT to say that I think the CD medium is
"perfect" or anything. It's dynamic range can and should be improved, but
not to the extent proposed by those who seem to have not thoroughly and
pragmatically investigated the topic. After all, what use is two ultra-
(and needlessly) high-resolution channels? It's still just two channels,
and comes negligibly closer to "reality" while taking up precious space
that could be used for better spatial-reproduction systems such as
Ambisonics.
Espen Braathen
Maybe Stereophile should make a HDCD disc of the master as well, and
perform a master vs CD vs HDCD comparission?
EspenB
Chris Ross +44 1383 822131
misc...@cantua.canterbury.ac.nz (Hamish Hubbard) wrote:
>FWIW, I find the encoding of a musical event as a stereo pair of
>electrical signals fairly amusing, since these signals are so totally
>incapable of containing enough information to reproduce physically the
>musical event. Compared to that, arguing over whether to encode in
>analog or digital seems insignificant. Spending thousands on hi fi
>gear is highly amusing, to me, from this perspective.
Your reply sounds very much like the perspective of a musician, are
you a musician?
>Why not buy a nice instrument and learn to play it, for the same
>amount of money?
Why not indeed. I would very much like to be able to play music
myself, in fact I am thinking about learning to play the 'cello.
However, I will be 35 in a few weeks and I think that it is probably
thirty years too late to start. Do you think there is any point in
starting at this age?
Since I don't play, my musical pleasure is gained from either
listening to other people playing, or listening to recordings of other
people playing. Naturally, therefore, I want to get the best
reproduction possible, within constraints of cost and convenience of
course.
> A 'sound system' in its current form is never going to approach reality,
>merely provide ear-candy. Nothing wrong with indulging, though :-)
I basically agree, certainly with current technology. However, I also
feel that unless we as consumers constantly seek better and better
approximations of reality there will be no reason to advance the state
of the art. I think it highly likely that one day it will be possible
to completely recreate recorded sound within a confined space, perhaps
when the total surface of the 'listening room' is in effect a sounder
and not just a couple of boxes at the front. One can see this
beginning to become imaginable with the development of Ambisonics
making full periphony a reality. With the ceramic sounders used in the
aerospace industry (e.g. for anti-noise) being small enough to be
plentiful and unobtrusive, arbitrarily high numbers of audio channels
are theoretically possible now. e.g. 128 full range sounders built
into a room to provide 128 channel full-sphere periphonic sound from a
six-channel ambisonic encoded DVD. Sort of like the audio equivalent
of the 'Holodeck' on Star Trek. Now if only we could de-rail the Dolby
AC-3 bandwaggon and get the ARA proposals incorporated into DVD.
Reproduced music will always be valuable. Even assuming that I were to
master the 'cello, would that stifle my desire to listen to recordings
of great string quartets? Hopefully not! What about, for instance,
those people that continue to enjoy the music of Elvis Presley even
though the poor chap is now dead? Do you feel that they would be
better served by buying the sheet music and just singing the songs
themselves? I think not, it is not just the appeal of the songs but
the precise way in which he performed them that people are paying for.
It is therefore worthwhile to aim to recreate that performance as
fully as possible, even if it takes another century to get there.
Chris Ross,
Edinburgh, Scotland
P.S. Sorry for mentioning Star Trek.
Steve Zipser (Sunshine Stereo)
John Busenitz wrote:
> On Tue, 28 Jan 1997, Steve Zipser (Sunshine Stereo) wrote:
> > No, the Velodyne is not serious because the amp/xover isn't seperate.
> Well, and to think I thought what made a subwoofer serious is their
> low-frequency cutoff, flat response, low distortion, and output
> capability. So it's the old "separates" thing that makes a difference.
> What makes a subwoofer "not serious" when the amp/filter is designed
> into the subwoofer itself?
John:
Look at the quality of parts used in the F-1500 - they aren't very
good. They used to make a ULD-15 that was far superior w/ seperate
x-over/pwr amp. There is also no provision for altering high & low
pass cut offs, and there is no 180 degree phase control, and you have
to run a lot of ong & expensive & sonically degrading cqables because
the amp/xover si so far from the main systems (generally).
Zip
Nils Christian Framstad
Some questions which needs further explanation:
1) What equipment, in particular what DAC, was used for playback? Why
not use the same DAC as when converting for the LP pressing? This
should at least eliminate that difference, true?
2) Why was there an audible difference between digital master and CD?
Possible explanations:
a: The fact that it was not the master, but a copy that is necessarily
not closer to the real thing.
b: Loss of information in the last 4 bits. (Information which on the
other hand was transferred to the LP)
c: The process of "burning" the CD. (A problem related to the medium)
d: The process of reading the data. (Another problem related to the
medium)
e: The process of converting to analog. (If the same DAC is used, it
should be no problem, except matching between the spinner and the
DAC.)
In case of a: One could eliminate this problem by copying the master
once again and introducing it in the test
b: I suppose there is a 16-bit tape acting as master for the CD. One
could listen to this as well. Of course, comparing LP and CD, one
could use this as a master for an LP. (Is it getting expensive now?)
John Atkinson wrote:
[snip-a-lot]
> Is the LP doing something, as of yet unknown,
> right? Or is the CD doing something, as of yet unknown, wrong?
First of all: It is not clear that from your posting whether (the
audience thought that) the LP only "preserved other properties" of the
master, or added something.
Second, if your audience is right, then at least SOMETHING wrong has
happened along the way into and out of the CD.
ncf
Peter Irwin
On 29 Jan 1997, Kevin Connery wrote:
> Probably true. I do note that most analog fans aren't anti-
>digital so much as anti-the-current-digital-implementation
>-- there's a big difference.
John Busenitz replied:
>I politely disagree. From what I've seen, the majority of
>analog fans harbor an intense distrust and dislike for all
>things digital, probably because it is a "new" technology,
>and they don't understand it.
Perhaps this is true, but sometimes people go farther than this and
they decide that the reason audiophiles had trouble with the sound of
CDs in the early days of CD was primarily due to a dislike of things
new, or misconceptions about digital audio.
I am not really disagreeing with John Busenitz here, (nor am I sure I
entirely agree with him about this) but I just thought this would be a
good time to give a reply which included an assessment of CD problems
and digital audio as they seemed to an audiophile writer back in 1983.
From:
Andrew Marshall. Audio Ideas Guide. November 1983. p.28
First of all, I have concluded that, whatever may
be wrong with early CD players, it is not the digital
system itself. The same standards apply to the digital
processors as to the the Compact Disc system, yet the
sound of digital tapes is quite different from that of
early CD machinery. Technics' SV-P100 complete VHS
recorder has proved its ability to make clean,
noiseless tapes that are much better than CD
reproduction from a first generation player. Tapes of
LPs made on the SV-100 sound better in such cases than
the CD disc equivalents do direct from early CD
players. The similarly superb sound from the new group
of processors (Designed to be used with either Beta
or VHS VCRs) reinforces one's awareness that the
digital system is not the problem. It is also
instructive to note that the BBC in Britain has been
using 13 bit digital processing for its network feeds
for years, and listeners continue to rhapsodize about
the Beeb's beautiful Third Program Service sound. The
digital systems work.
I just thought this might be interesting.
Peter.
----
Peter Irwin
pir...@ktb.net
John Busenitz
> > On Tue, 28 Jan 1997, Steve Zipser (Sunshine Stereo) wrote:
>
> > > No, the Velodyne is not serious because the amp/xover isn't seperate.
> John Busenitz wrote:
>
> > Well, and to think I thought what made a subwoofer serious is their
> > low-frequency cutoff, flat response, low distortion, and output
> > capability. So it's the old "separates" thing that makes a difference.
> > What makes a subwoofer "not serious" when the amp/filter is designed
> > into the subwoofer itself?
On Thu, 30 Jan 1997, Steve Zipser (Sunshine Stereo) wrote:
>
> Look at the quality of parts used in the F-1500 - they aren't very
> good. They used to make a ULD-15 that was far superior w/ seperate
> x-over/pwr amp. There is also no provision for altering high & low
I don't disagree, but this has nothing to do with whether, as you said
earlier, having a separate amp/filter makes a subwoofer serious.
> pass cut offs, and there is no 180 degree phase control, and you have
> to run a lot of ong & expensive & sonically degrading cqables because
> the amp/xover si so far from the main systems (generally).
But with a separate amp/filter, the subwoofer is so far away from the
amplifier that you must run a lot of long, more expensive, and more
sonically degrading speaker cable to the subwoofer. Running line-level
is usually preferable to running speaker cable, since the amplifier
contains more gain stages which can compensate for reduced level from
cable resistance, while long speaker cable will increase the subwoofer
system Q and reduce output by increasing the voltage drop across the
cable.
A possible disadvantage is increased capacitance with longer
line-level connections, but I don't know that this is a problem,
especially with a low-pass filter between the preamp and the
amplifier.
Maybe Zipser defines "serious" differently than I do, but I say a
"serious" subwoofer is one with flat output, a well-damped alignment,
high output capability and a low cutoff.
John Busenitz
On Wed, 29 Jan 1997, John Hamm wrote:
> Wasn't the original statement that the CD was more accurate in a Hi-Fi
> sense; frequency response, signal to noise, etc.? That doesn't equate
> to 'musically' accurate. Now before you aim-the-flame consider that a
> system can "measure good, sound bad" and vice-versa.
I don't understand the distinction, and I guess I don't agree with
your last statement. I would tend to think that if something sounds
bad, there is a reason for it, and most likely it can be measured. The
fact that it "measured good" (measured well?!) merely shows that a
complete and properly-interpreted set of measurements was not done.
However, I am curious as to how Mr. Hamm and others would distinguish
betweent technical and "musical" accuracy, and define the latter. I
would think that the definition of "musical accuracy" would change
quite a bit from audiophile to audiophile.
Scott Frankland
John Atkinson wrote:
>
> So if all conventional measurements and even individual subjective
> observations on individual aspects of the sound -- bass extension,
> clarity, HF cleanliness, etc. -- indicate that the noiseshaped CD has
> better "fidelity" to the _sound_, how can the less "accurate" LP be
> closer to the _music_? Is it just the addition of groove noise, pre-
> echo, and distortion, as has been proclaimed by some of the denizens
> of rec.audio.high-end? Is the LP doing something, as of yet unknown,
> right? Or is the CD doing something, as of yet unknown, wrong?
Indeed, how can the LP be closer to the music if it is less accurate?
Well, here's a story for you. RCA cut a record in the '50s known as
"Also Sprach Zarathustra". Only about 1000 to 1500 copies were
produced according to informed sources. Of these, all were mastered
from the same acetate (1S) and from the same stamper (A1) at the
Indiana pressing plant (I). On numerous occasions, we have noticed
marked differences in sound quality from apparently mint copies. How
to account for this? Well, the only difference I can think of is due
to stamper wear. One would expect the early pressings to have more
palpable detail than later pressings. This is indeed the case
(although it is impossible to know which is which, for obvious
reasons). Now, the pressings with less detail also seem to have
considerably better "ability to envelop the listener in the music" and
to "better preserve the 'wholeness' of the soundstage". This seems to
bear out the notion that a certain amount of groove polishing and
concomitant information loss is conducive to a more involving musical
experience. Go figure. All I can say is, musicality is an elusive
creature.
~SF~
Kevin Connery
In article <5cqvu8$k...@agate.berkeley.edu>,
Steve Zipser (Sunshine Stereo) <z...@netrunner.net> wrote:
>John Busenitz wrote:
>> On Tue, 28 Jan 1997, Steve Zipser (Sunshine Stereo) wrote:
>> > No, the Velodyne is not serious because the amp/xover isn't seperate.
>> Well, and to think I thought what made a subwoofer serious is their
>> low-frequency cutoff, flat response, low distortion, and output
>> capability. So it's the old "separates" thing that makes a difference.
>> What makes a subwoofer "not serious" when the amp/filter is designed
>> into the subwoofer itself?
>Look at the quality of parts used in the F-1500 - they aren't very
>good. They used to make a ULD-15 that was far superior w/ seperate
>x-over/pwr amp. There is also no provision for altering high & low
>pass cut offs, and there is no 180 degree phase control, and you have
>to run a lot of ong & expensive & sonically degrading cqables because
>the amp/xover si so far from the main systems (generally).
Um, actually, while I'm not thrilled with the _quality_ of the
crossover, it does have a variable frequency low-pass filter (40-100
Hz), though with a fixed rate. And the high pass IS fixed.
Scott Frankland
Arnold B. Krueger wrote:
> Is this the source of the "warmth" mentioned by Paul Schaeffer on
> the Letterman show last Monday night?
The problem with this type of coloration is that it is pandemic, i.e.,
not all recordings will benefit from this type of coloration.
> While 45kHz is supersonic, a large moderately damped resonance there
> can have measurable (and audible) effects at 15 kHz and below.
I'm not sure how, unless the resonance is also non-linear. In that
case, IM products can appear as difference frequencies. Is this what
you mean?
~SF~
Kevin Connery
In article <5cqkgn$d...@canyon.sr.hp.com>, John Busenitz
<buse...@ecn.purdue.edu> wrote:
> On 29 Jan 1997, Kevin Connery wrote:
>> Probably true. I do note that most analog fans aren't anti-digital
>> so much as anti-the-current-digital-implementation -- there's a big
>> difference.
> I politely disagree. From what I've seen, the majority of analog
> fans harbor an intense distrust and dislike for all things digital,
> probably because it is a "new" technology, and they don't understand
> it. Witness people throwing about "infinite resolution" and "hearing
> below the noise floor" without having a clue what these terms mean.
Fair enough; it sounds like your sample is very different than mine.
Of the 7 people I know who might be considered audiophiles, 3 are
pretty much in the analog camp, either not buying CDs or only getting
on CD what isn't available on LP. (I'm in the first of those camps.)
None of them/us are anti-digital; truth be told, all three of us are
in the computer field, which is about as digital as it gets (one tech
writer, one programmer, one QA guy). The programmer also built his
passive pre-amp, and isn't a stranger to electronics, having also been
a ham operation in his younger days.
But we don't like the way the _current_ CDs sound. And, frankly, all
the explanations as to the reason haven't 'fixed' that. It has
improved markedly, as implementation caught up with theory, but 3-of-7
prefer the existing LP to the existing CD. And two of the others still
have their turntables (though I haven't any idea how often they're
used.)
With the new, proposed standards giving some breathing room for
mastering errors, implementation limitation relative to theoretical
max, etc., I'm hoping to be able to convert--for two reasons:
The current LP is a pain-in-the-anatomy to set up and maintain
I'd love something which sounded better and was more convenient.
IN other words, I'm stuck agreeing to disagree; either your data
doesn't match mine, or your interpretation of the same data is
different.
> Which brings my to my second disagreement: there is not a "huge"
> difference between CD audio and the "next digital
> medium". Recordings and listening rooms probably don't have the
> dynamic range to take full advantage of current CD technology, let
> alone 120 dB (or more)
Were the issue inherently dynamic range, I'd probably agree. But I
don't know *what* the problem is, but *something's wrong* with the way
it sounds today. Nowhere near as much as it did 10 years ago, when
people were saying what they're saying now; It's Good Enough, and You
Can't Hear The Difference, and so on, but *to me* it's not right.
I'm *HOPING* a new standard will be sufficient. Whether that's a 1
channel, 2 channel, 8 channel, or any combination isn't as important
as the way it works in the real world.
--kdc
Michael Ford
> recording. The CD was released last year at HI-FI '96, and was
> noiseshaped from the master's 20-bit resolution to the 16 bits of the
> CD standard using the Meridian 518. The LP was cut by veteran
> mastering engineer Stan Ricker from the 20-bit master (played back on
> a Nagra-D and decoded to analog with a 20-bit Mark Levinson No.30.5
> D/A processor) without any compression or equalization (other than
> RIAA). (The LP is cut "hot" and has high levels of vertical modulation
If the mastering consisted only of setting a high level and leaving it
alone, what does it matter who did it?
Several possibilities occur to me.
1) No audible sonic differences between the formats actually
exists. The perceived difference being due to elements of the
presentation and personal bias that the listening tests failed to
eliminate.
2) The added filtering and euphonic distortion of the LP process sound
better than the raw signal as recorded on the master tape.
3) The Meridian 518 degraded the sound.
How did the sound of the Meridian 518 16 bit output compare to the 20
bit master?
4) Some other element of the CD process degraded the sound.
> In careful level-matched listening tests, neither the CD nor the LP
> were found to sound identical to the mastertape. As Larry Archibald
> describes in the February's issue's "Final Word" column: "The master
> tape is effortless -- the sound just happens, without trying hard,
> ineffably."
You forgot to add, "when I know which source is being used in the
test." Do you think the master tape might have been a little less
effortless or happening if you were not visually sure of the source?
Prove it.
Chris Ross (GEC Marconi Avionics)
my...@hpfcla.fc.hp.com (Bob Myers), on 29 Jan 1997 16:43:48 GMT wrote:
> First, the transformation from 20 bits/sample to 16 does not
> constitute "downsampling", at least as I would commonly use that term,
> since there is no change in the number of samples per second. Whether
> or not all of the information in the 20-bit original is preserved in
> the 16-bit version depends on whether or not there actually WAS
> information, as opposed to noise, in those lower four bits.
Hmmm, I don't think that this is exactly how it works. It bothers me
because it's obvious from this and previous posts of yours that you
clearly know what you're talking about, so I assume I've either
misunderstood what you're trying to say or my understanding of digital
signal processing is flawed. (It's true that I've never worked
directly with audio DSPs, the closest I've got is to write to the
firmware for CD-ROM drives). Either way I'd really welcome
clarification.
Your statement "depends on whether or not there actually WAS
information, as opposed to noise, in those lower four bits" implies
that the extra four bits in 20-bit resolution are just tacked on the
end and that you can safely throw them away if they're not actually
used. At the risk of un-technicallifying to the point of absurdity, my
understanding is this: they don't 'go on the end' as such, the effect
of increasing the resolution is to 'move' all the available values
closer together. Trying for a non- techie example, imagine that a set
of things can be scored marks out of 100, and that you need to change
the scheme to award everything marks out of 10 instead -
i.e. transform to a lower resolution. Given two marks, say 52/100 and
54/100, these will need to be assigned new marks in the new scheme,
e.g 5/10. With dithering, 54/100 might in fact be assigned 6/10 rather
than the obvious, but I won't get into that here.
> (Key item: simply adding more bits/sample does NOT guarantee that you're
> actually dealing with more INFORMATION.) If there was, then yes,
> there MUST be distortion in making this transformation, since you have
> lost information. But if not, then what the 20-bit master let you do
> was work in a digital domain where the math errors won't make it into
> those bits where the sound really lives. In short, the 16-bit version
> would come out as faithful TO THE ORIGINAL SIGNAL as the 20-bit was,
> since there was no real information beyond 16 bits anyway.
What it looks like you're saying is that putting the extra on the end
would mean that moving from the 5/10 to marks out of 100 would give
5/100, and that so long as we never had any marks higher than 10 we
could convert back to marks out of 10 simply by throwing way any marks
higher than 10 because they only contained 'noise'. As I said, I don't
think it works like that. You are going to lose some accuracy unless
by chance all the marks happened to be exactly divisible by 10.
I'm looking forward to your clarification on this.
Chris Ross,
Edinburgh, Scotland
Disclaimer: I am not representing the views of GEC Marconi Avionics.
Kazushi Endoh
my...@hpfcla.fc.hp.com (Bob Myers) wrote:
>- Given the above, and from #3 and #4, the most logical hypothesis
> appears to be that the preference for the LP sound comes from
> qualities inherent in the LP medium, which could best be
> characterized as "euphonic" distortions. A second possibility
> exists - that the preference was due to the effect of some other
> component in the signal chain which was NOT identical between the
> two - in which case the test says nothing about the inherent
> superiority of either medium, and most certainly does NOT imply that
> the LP is "more accurate".
>
>If you care to offer an alternative hypothesis, Curtis, I'm all ears.
Since I have no analog front end (just because of inconvenience:-), it
is difficult to participate in, for me. But, most of my friends are LP
lover, I have lot of opportunity to listen to analog system. So I
would like to add some observations.
The Laser turntable that utilizes photo-transducer preserves LP's
"juicy" sound. The photo-transducer may not suffer from mechanical or
magnetic distortion. Since the pickup is optical, vinyl material
itself is not relevant to "juicy" sound. The "euphonic" distortions
may not be introduced in this step.
Some RIAA pre-enphasis/de-enphasis circuit for CD tweaking is
available, and I felt it does not have much meaning, just a bit
roll-off in highs. RIAA circuit may not help "juicy" sound. It may be
interesting to try a digital RIAA by using DSP. I know DSP-RIAAed CD
though, I have ever seen, but never listened to.
Concerning frequency domain, LP lovers claim "bass movement". Fairly
well set-upped analog vs. Mid-Fi CDP is something like full-body wine
vs. diluted vodka (alcohol content or calorie are the same, though).
The difference was in mid to mid-bass range (my observation). Good CDP
emphasizes instrument (in the music) placement , while LP emphasizes
"in the room" feeling. The subsonic noise may not be relevant, since
some digitally recorded CD also contains subsonic noise that does not
sound like LP. Does the modulation test track in a test CD sound like
LP?
Since some people claimed that DAT could preserve LP sound, isn't it
recordable CD vs. LP on DAT? (Sorry I don't have DAT)
If a full digital LP cutting with DSP-RIAA could sound euphonic, is
the "euphonic" distortion introduced in a cutting process? or is it
due to high gain head amp??
Just wondering which step could add euphonic effect...
--
Kazushi Endoh Med.Univ.Yamanashi
ken...@res.yamanashi-med.ac.jp
PSiu89449
vdoran (vdo...@videohomesearch.com) wrote:
> I agree that replies to your original post have not addressed your
> most interesting question, which I take to be "What will give me the
> best sound from a subwoofer integrated with my main speakers?", so I
> would like to blather on about some stuff you probably already know,
> and then pose a question to you.
This is thus far the most intelligent and direct reply to my original
post. No offense to other knowledgable contributors in this
thread. Hope I can learn much from Mr. Doran and others.
> Placing the sub in the middle of the room it seems to me will only
> increase its tendency to drive these fundamental response. So IF
> one assumes flat freq response is the desideratum, the corner wins
> for a small room. (if your room is 30x30, on the other hand, and no
> elephants listen to your music, then surely middle placement for
> major 18hz would be more fun.)
Fortunately or unfortunately, pending on how one look at this, my
dedicated music room is a spacious loft measuring 25'L x 17'W x 11'H
open on one side to a similar size living room down stairs which gives
a total ceiling height of 25'. Furthermore, both rooms benefits from a
compound sloping ceiling. To maintain some form of front of the room
symmetry, I have a large office partition treated with acoustic foam
erected on the open side to serve as a stand-in wall. Since I also use
the living room as a home theater, both room are moderately treated
with acoustic treatments not excluding various Room Tunes and Bass
Traps. Perhaps an occasion bear or two knocks on my front door. No
elephants apply.
> It surely seems preferable that a bass attack should be rendered as
> "THUMP" rather than "TH" (mains) followed several ms later by "UMP"
> (sub) (or in your placement "UMP...TH"). This would argue for sub
> placement at the same distance from your ears as the mains (assuming
> no delays in the rest of the system).
It is precisely "THUMP" that brings forward the most accurate attack
assuming that frequence response is properly maintained. Thus far, the
closest I can get is by moving my system out to my spacious deck but
neighbor's complain aside, the whole system sounds subjectively
unnatural. And what about rain and snow?
> And so my question: You say that you have done many many
> measurements /variations. Presumably you have also done much
> listening. WHAT SOUNDED THE BEST? Could you discern an improvement
> from a time-aligned setup? Even if it impaired the frequency
> response? Could you make any improvement in sound of the
> time-aligned setup by varying the phase?.........but for
> transients/steps/attack which LIE IN THE CROSSOVER REGION phase will
> matter.
Being a wan-na-be speaker builder/designer for the past 10 years, I
had done some semi-educated research in the way loud speaker interacts
with a room. Measurements are relatively easy to make, but the
interpretation of such takes experience and deep acoustic
understanding, both which I do not pretend to acquire. Henceforth my
question to the public. To answer Mr. Doran's questions, I have to
present a few of my inconclusive studies.
Reading back from my notes. Using primary the LAUD, no SigTech, the
first setup I did was to time align one subwoofer (PF-1800) with my
main speakers (Alon V Mk II). The mains were 1st order highpassed at
70Hz via Bryston 10B and the sub was 6th order internally lowpassed at
55Hz with 0 degree phase. Main speakers were 8' from back wall and
3.5' from side walls/partition. Sub was strategically placed against
the railing which opens down to my living room. Strategically here
bing furtherest away from any walls for longest possible reflected
arrivals. Measuring/listening position is 8' from the back wall making
this 7' from the main speaker's center line. The sub ended up
approximately 7' 9" from the same. Subjectively, with this set-up, I
lost all sence of deep bass but mid-bass and up was very well
integrated. Bass attack was tight and immediate unfortunately the
"THUMP" turn into "THump". Disconnecting the sub while running the
mains full range brought back the foundation somewhat. A stong hump at
45Hz was preceived as the function of Alon V and room but extension
was not up to my standards.
Using the above set-up with SigTech gave me almost everything I could
wish for except the PF-1800 was crying out in pain bottoming at loud
transient attack. My ULD 18-II didn't fare much better.
Second configuration involved the same setup but moving the PF-1800
against and immediately behind my listening chair. With sub's phase at
0 and volume adjusted to match the main, extension was great but I do
preceive a "UMP...TH" problem. Again with the help of LAUD to
approximate a better, but never obtainable, minimum phase system
produced a subjectively smooth integration. However, what I miss at
this point was the turn on the dime quickness of the time aligned
setup. Inserting the SigTech would improve the overall system
response, especially in controlling colorations, but never the less,
attack is still wanting.
Third configuration involved again the same setup but moving the
PF-1800 against the far corner behind the main speakers. To cut this
short, bass response was all over the place. Adjusting phase will
further delay the "TH..........UMP" but did improve crossover
integration. Inserting the SigTech again improve system response
tremendously but the attack is more like a retreat.
Fourth configuration involved a total of 5 subs, 2x12" + 2x15" + 1x18"
all time aligned and spread all over the room. What an ugly sight!!!
Purpose of spread multiple subs was to lower individual sub's exertion
thus lower distortion and the possibility of bottoming. Another
purpose is to excite multiple room modes and thus obtain a better
average. Did this work? Yes wonderfully in most aspect but horribly in
others. Attack was good but not great. Extension? Need you ask? The
whole floor if not the whole house move along with the
music. Inserting the SigTech almost put everything in tight control
but the problem I get was the phasy bass. As one could imagine, the
phase got extremely complicated. I didn't even care to measure
it. Another problem is that walking into my music room is like walking
into a maze. One active sub plus four passive subs along with vertical
bi-amp main speakers makes for a lot of power amps and cobweb full of
wires. My neighbor thought I was crazy let alone my wife. I agree. :)
Back to Mr. Doran. Yes, I can definitely perceive an improvement with
time-aligned setup even if it impair frequency response. No, I cannot
duplicate the subjective improvement I obtain from time-alignment by
varying the phase. And finally, the best was running the main speakers
solo with no subwoofers but giving up the bottom octave is an
acceptable compromise for me.
> I am interested in your experiences, although I realize they cannot
> be quantified unless you have an unusual amount of test equipment in
> your home.
Even if I have an unusual amount of test equipment, I would far from
consider myself as quantifiable. But at least I hope to sturr up some
interest in this subject and hope to learn more from experts in this
field.
> Perhaps if we asked politely we could get John Dunlavy to comment on
> this question.
The subject of politeness I always practice. Just hope that I'm up to
standard.
Best Regard
Paul Siu
Bob Myers
Kazushi Endoh (ken...@res.yamanashi-med.ac.jp) wrote:
> The Laser turntable that utilizes photo-transducer preserves LP's
> "juicy" sound. The photo-transducer may not suffer from mechanical or
> magnetic distortion. Since the pickup is optical, vinyl material
> itself is not relevant to "juicy" sound. The "euphonic" distortions
> may not be introduced in this step.
Sure they can. The euphonic distortions of the LP medium don't
necessarily all occur at the moment the stylus is tracking the groove.
A lot of them are due to the way vinyl takes the information stamped
into it in the first place, the nature of the material itself, and
what goes on when the master disc is cut. Whether the final product
is played with a laser, a stylus, or a pocketknife has no effect at
all on the latter item, and the first two could be picked up by a
non-contact transducer just as well as by a mechanical stylus.
(Either the laser pick-up follows the surface as well as the stylus,
or it doesn't. If it doesn't, then I would question its use in the
first place - but if it DOES, it WILL pick up the vinyl surface noise
and so forth. The big advantage is that it won't be embedding all
those dust particles in the groove walls or chopping off all the sharp
edges as they're played.)
Robert Orban
In article <5cntjj$4...@eyrie.graphics.cornell.edu>,
enjoyt...@top.monad.net says...
[snips 'o plenty]
> double-blind... Please seek another person. For if YOU could only
> hear how VASTLY DIFFERENT it was between the CD and the master DAT.
> And YES, i AM familiar with the basic differences between 20 bit DAT
> and 16 bit seedie, er, um, CD.
..ummm... Am I missing something here? There ain't no such thing as a
'20 bit DAT'. The DAT format is 16-bit, PERIOD.
There are relatively few means available to the professional to put a
greater-than-16-bit signal on tape at reasonable cost. One is the
Nagra-D. There are also adapter that use a pair of tracks on an
inexpensive 8-track digital recorder like the Alesis ADAT or the
Tascam to encode a 20-bit signal. And there are digital workstations
that will encode 20-bit signals on hard disk relatively
inexpensively. But no DAT will do it.
Bob Myers
Kevin Connery (Ker...@cris.com) wrote:
> Were the issue inherently dynamic range, I'd probably agree. But I
> don't know *what* the problem is, but *something's wrong* with the way
> it sounds today. Nowhere near as much as it did 10 years ago, when
> people were saying what they're saying now; It's Good Enough, and You
> Can't Hear The Difference, and so on, but *to me* it's not right.
OK, but the fact that you have heard a marked improvement over the
last ten years suggests that at least MOST of what you don't like has
nothing to do with the CD standard itself - that HASN'T changed over
this period. So looking to a new standard to cure the problems you're
hearing is very likely barking up the wrong tree. The implementation
of the standard is not the standard itself, and in all likelihood the
problems have VERY little to do with being "analog" or "digital".
John Atkinson
To answer some of the questions that have been raised in this thread:
All listeners preferred the 20-bit master to the LP -- when they could
do a direct comparison.
All listeners except one preferred the 20-bit master to the
noise-shaped CD.
All listeners felt the CD was closer in "hi-fi terms" to the 20-bit
master than the LP.
All listeners preferred the LP for the quality of the listening
experience.
The tests I referred to were mainly sighted, with some single-blind.
Peak levels were matched. Audible background noise was not a factor in
the case of the LP. Ticks and pops of course would give identification
clues but there were very few of these per LP side.
The same D/A processor was used for the playback of the 20-bit master
and the CD. It was also the same as used for the LP mastering.
The Meridian 518's noise-shaped 16-bit output, taking its feed from
the 20-bit master, sounded very close. It sounded closer than the CD
cut from the same data. However, it was still distinguishable from the
master on some program some of the time.
Mastering an LP involves rather more than just "setting a high level
and leaving it alone." As this is a craftsman-like operation involving
skill and experience, it matters very much who does the cut.
Yes, it was too expensive to have masters cut and test LPs pressed
from the 16-bit CD as well as from the 20-bit master. This would have
added a hefty 4-figure sum to the costs of the project for no benefit
other than satisfying our curiosity. And as I pointed out, I am not
even sure if we will break even on this project.
My point was that if you examine each individual parameter of
performance, either objectively or subjectively, with the exception of
the mid-treble region where a 16-bit digital system's error floor is
not low enough to be inaudible to all people on all recordings all of
the time (given a realistic playback spl), the LP is a less accurate
facsimile than a CD made from the same hi-rez master. Yet it is
preferred overall to the CD. Again I ask is this because of what it is
additively doing wrong -- the euphony argument -- or because of what
it does right (including psychoacoustic factors such as the stochastic
effects of inaudible noise)? Speaking personally, I have no idea which
is correct. Or even if the answer involves both!
John Atkinson, Editor, Stereophile
John Ongtooguk
Arnold B. Krueger (ar...@concentric.net) wrote:
: During reproduction, a sonic image could become smeared if anything
: changed the phase or amplitude relationships in the signal for each
: channel. Additional smearing can be expected if the phase or
: amplitude relationships were changed between the channels. The major
: source of sonic smearing in most systems are reflections and
: inconsistencies in the acoustics of the listening room.
I agree not only with the above but also with the rest of your
nicely written summary of some of the problems with LP playback.
When doing cartridge reviews for 'Audio' awhile back Long used
some of the techniques mentioned but I've not seen anyone else
do so. Due to such problems one would expect the LP to be a
less accurate reproduction of the recording since so much care
is needed in component selection and setup in order to minimize
the problems mentioned, a process that in part can be at odds
with the preference for 'good sound'. Still, I think it's
obvious that it will take more than pointing out some of the
differences in currently measured performance between LPs and
CDs in order to put the issue to rest.
Even when using senses that we tend to have more confidence in,
such as our eyes when viewing images, it seems common to fall
back on subjective evaluation of quality in spite of the
number of methods that are available for measuring performance.
Perhaps the differences between decent CD and LP systems are
more like the choice of film color palette as opposed to
format and lens selection, and perhaps the appeal of LP is
similar to that of film with a high color saturation; the
composition doesn't change, image details are still well
recorded, but the colors have a bit more snap. It can be
easy to at least subjectively compare a recorded image with
the subject, something that is hard to do with audio but
something that needs to be done.
I've read testimonials that state that the CD seems to do a
better job of capturing what's on the master tape but there
are still a number of comments that suggest that LP does a
better job of something; imaging, soundstaging, presence,
or whatever it is. My guess is that it's what I would call
imaging, the cues that reinforce the illusion of a 3D
soundstage, and that such cues when poorly recorded or when
played back thru systems that don't image well tend to only
enhance 'presence'; the instrument sounds more like an
instrument even if the soundstage isn't fully formed. To
state that the LP is merely euphonic may require carefully
controlling the variables that contribute to imaging, and
demonstrating that the LP is enhancing or distorting the
recording.
John Ongtooguk (jo...@vcd.hp.com)
Kevin Connery
In article <5d2ukm$t...@agate.berkeley.edu>,
John Ongtooguk <jo...@hp-vcd.vcd.hp.com> wrote:
> Even when using senses that we tend to have more confidence in,
> such as our eyes when viewing images, it seems common to fall
> back on subjective evaluation of quality in spite of the
> number of methods that are available for measuring performance.
I think that is due to the current situation where something which
measures as 'perfect' as we can measure doesn't necessarily sound
good--indicating that we're missing at least one critical measurement.
Certainly, the quantifiable measures can be used to 'disqualify'
devices which are NOT good enough, due to clear, measureable, defects,
but the converse is not yet true.
Hearken back to the early days of the CD, where all the tests being
done--the ones based on earlier models--showed the sound to be as good
as it could get. Over time, other measurements were identified which
showed differences in sonic (and other) behavior.
We don't have a complete set of quantifiable tests which can be run to
reliably identify the "good" equipment. Yet.
--kdc
Steve Jones
Scott Frankland wrote:
> Well, here's a story for you. RCA cut a record in the '50s known as
> "Also Sprach Zarathustra". Only about 1000 to 1500 copies were
> produced according to informed sources. Of these, all were mastered
> from the same acetate (1S) and from the same stamper (A1) at the
> Indiana pressing plant (I). On numerous occasions, we have noticed
> marked differences in sound quality from apparently mint copies. How
> to account for this? Well, the only difference I can think of is due
> to stamper wear.
Or process thermal variation. If the LP run is of varying increamental
pressings, the mechanical press may (or may not) have a chance to
fully stabilize thermally. How would you account for the breaks,
lunch, or shift changes that disrupt the process? Could this be
infered as analog process jitter?
BTW, I've seen this on CD's too (a cloudy swirl pattern in the
polycarbonate). Haven't noticed any sonic problems with such to date.
> All I can say is, musicality is an elusive
> creature.
IMHO, more real than Bigfoot!
-Steve Jones
Scott Drysdale
In article <5d2kto$p...@agate.berkeley.edu>, John Atkinson <7447...@CompuServe.COM> wrote:
>To answer some of the questions that have been raised in this thread:
i understand that making an LP using the CD as a master would be big
$$$, but it should be very easy to make a 20-bit tape of the LP, and
make a CD-R from that. or just A/D->D/A the LP playback, and see if
anyone can tell the difference. if the digitized version of the LP
still sounds like the LP, then something in the LP process is altering
the signal. and if you prefer that altered signal, that's cool - sell
CD's made from LP playback.
>My point was that if you examine each individual parameter of
>performance, either objectively or subjectively, with the exception of
>the mid-treble region where a 16-bit digital system's error floor is
>not low enough to be inaudible to all people on all recordings all of
>the time (given a realistic playback spl), the LP is a less accurate
>facsimile than a CD made from the same hi-rez master.
please expand upon this "error floor" thing. i smell i rat.
// Scott Drysdale // Visual Networks Inc // Software Engineer
// sdry...@visual.mctec.com // 60 HD FLH & 57..69 HD FLH // AMIGA!
Arnold B. Krueger
John Atkinson <7447...@CompuServe.COM> wrote in article
> Mastering an LP involves rather more than just "setting a high level
> and leaving it alone." As this is a craftsman-like operation involving
> skill and experience, it matters very much who does the cut.
IMO, this paragraph tells only half the story. The "missing paragraph"
would run something like this:
> Setting up a LP Playback system involves rather more than just
"hooking up some > components and leaving it alone." As this is a
craftsman-like operation involving > skill and experience, it matters
very much who does the setup.
The bottom line is that cutting and playing back a LP is an artistic
venture that involves changes to sound character that is at least the
same order as mixing down the original master. Therefore, when a
preference is granted for LP playback, it easy to question whether the
preference is caused by the technical properties of the LP media
(which are basically a list of limitations) or the consequences of
additional steps of attention to the subjective impression that the
recording creates.
Since LP technology basically limits bandwith and dynamic range, it
makes life easier for other system components, like amplifiers and
speakers if they happen to have limited frequency response and dynamic
range.
Looking at the systems of my closest audiophile friends (who almost
universally embraced digital technology) the big change to handle
digital was greatly increased capacity for cleanly handling program
material with dramatically greater frequency response and dynamic
range.
Its my experience that if you just hook a CD player up to a LP-centric
audio system, there is a good chance that you are not going to like
the sound, and you may even find yourself breaking a few things.
Its my experience that if you hook a LP player to a CD-centric system,
the absence of broadband dynamic range in commercial LP's becomes
quickly and even painfully aparrent.
One friend of mine who routinely listens to LP's on a system designed
to exploit CD's has simply added a filter to eliminate excess low
frequency noise from the LP side. In the old days we called these
"rumble filters". In those days there was not much program material
down there. By CD standards, the cutoff point of these filters roughly
corresponds to where things are just starting to get interesting.
sdura...@aol.com
>My point was that if you examine each individual parameter of
>performance, either objectively or subjectively, with the exception of
>the mid-treble region where a 16-bit digital system's error floor is
>not low enough to be inaudible to all people on all recordings all of
>the time (given a realistic playback spl), the LP is a less accurate
>facsimile than a CD made from the same hi-rez master. Yet it is
>preferred overall to the CD. Again I ask is this because of what it is
>additively doing wrong -- the euphony argument -- or because of what
>it does right (including psychoacoustic factors such as the stochastic
>effects of inaudible noise)? Speaking personally, I have no idea which
>is correct. Or even if the answer involves both!
I think you'll find that those who rely on measurements will support
the idea that the preference for LPs is based on its non-linearities
being more euphonic (though it's wholly ironic that those who believe
this don't, as a rule, share in the preference for that euphony).
Those relying more on subjective observation can't seem to shake the
feeling that analog reproduction is more "right" because of certain,
as yet undocumented, subtractive effects in digital playback.
The motives for preference need to be better understood. What causes
one person to prefer euphonics, another accuracy? I have observed
that those coming from a strong technical background will prefer
accuracy, while those with weaker technical backgrounds will allow for
more euphony. For example, there are a lot more EEs strongly in the
digital, solid-state camp than in the LP, single-ended tube camp. My
personal view is that their expectation is driven by their knowledge
and experience.
On the other hand, I have also found that many who prefer LPs have
huge (and I mean huge) music collections. This is NOT to say that EEs
have no music collections. But, on balance, those driven by the sense
that the hardware merely serves the software are much more liberal in
the types gear they end up with. Notions of technical accuracy are
secondary to how well the gear brings forth the musical treasures of
their collection.
So what it comes down to really is diff'rent strokes for diff'rent
folks. It appears that those present for your listening sessions had
a preference for LP playback because of their expectation of what is
good reproduced sound. I suspect if you brought together a different
group, with different expectations (say folks who hang with Tom
Nousaine or Peter Aczel), you'd end up with a strong preference for
digital playback.
In the final analysis, it appears there is no "correct." And it looks
like we'll have to leave it at that, even though we may be sitting in
front of our turntables thinking "how in the world can anyone prefer
CDs?"
Jim Andrews
sdura...@aol.com wrote:
> On the other hand, I have also found that many who prefer LPs have
> huge (and I mean huge) music collections. This is NOT to say that EEs
> have no music collections. But, on balance, those driven by the sense
> that the hardware merely serves the software are much more liberal in
> the types gear they end up with. Notions of technical accuracy are
> secondary to how well the gear brings forth the musical treasures of
> their collection.
This one little paragraph is packed with so much subtle snobbery and
innuendo that it's difficult to figure out where to begin . . .
1. The implication that many LP lovers are TRUE music lovers, as
evidenced by their large LP collections (which brings to mind
some rather puerile contests of another sort), completely
ignores the fact that a lot of us simply cannot AFFORD a huge
music collection. I'm pleased that I have managed to acquire
maybe 600-700 recordings in my life (about a 50/50 split CD/LP),
but noone I know of would think of me as anything BUT a music
lover. Do I still get to play?
2. I'm sure that if you surveyed the truly "huge" privately-owned
music collections in the world, you'd find that nearly all of
them are dominated by LP. So what? Given that very few of us
can afford to acquire recordings at a rate that would allow one
to build such a collection in a short time (i.e., with CDs), it
stands to reason that such collections have been acquired over
a very long period of time. Let's say a "huge" collection is
one containing at least 10,000 recordings. At a rather prodigious
consumption rate of 500 recordings a year, that would take 20
years . . . and I'd be surprised if CDs accounted for 1/3 of such
a collection, even if the collector "got on board" relatively
early in the digital age. I'm willing to bet that in most cases,
it's economics that have driven the collection, not preference.
3. Please stop referring to your antagonists as "EEs". Unless you
have a very legitimate reason to speak to that very particular
subset of the "objectivist" camp, it is far too specific, and
it doesn't take an EE degree to figure out what accuracy means.
It also insultingly implies that a technical education somehow
affects one's ability to hear the "truth", but that your own
preference isn't constrained by the blinders of technology.
jim andrews
Nousaine
John Atkinson asks
"My point was that if you examine each individual parameter of
performance, either objectively or subjectively, with the exception of
the mid-treble region where a 16-bit digital system's error floor is
not low enough to be inaudible to all people on all recordings all of
the time (given a realistic playback spl), the LP is a less accurate
facsimile than a CD made from the same hi-rez master. Yet it
ispreferred overall to the CD. Again I ask is this because of what it
is additively doing wrong -- the euphony argument -- or because of
what it does right (including psychoacoustic factors such as the
stochastic effects of inaudible noise)? Speaking personally, I have no
idea which is correct. Or even if the answer involves both!"
Of course the LP is a less accurate reproducer of the master. The
reason listeners "prefer" the LP to the real sound is because they
hold a psychoacoustic bias toward what they "already know." They
prefer what they have already come to regard as good sound and are
reluctant to give up the reference even faced with overwhelming
evidence that it is wrong (sounds "more" like the master.)
Worse performance is only accepted as "preferable" by champions of the
old guard who have such strong psychologial ties ties to the existing
performance level that they are willing to reject reality and "vote"
on emotion...not performance.
Why is they question even asked by professionals? No one, except
aplologists for the dead technology, evers bothers to ask these
question....unless they expect to get advertising revenues from
someone. Again...if even your special listeners "hear" the digital
format as delivering the most accurate rendition of the master WHY are
you asking the "preference" question? If it is more accurate; it is
more accurate. Perhaps your listeners need more training!
Kevin Connery
In article <5d2r5v$3...@canyon.sr.hp.com>,
Bob Myers <my...@hpfcla.fc.hp.com> wrote:
>Kevin Connery (Ker...@cris.com) wrote:
[cut -- rgd]
>OK, but the fact that you have heard a marked improvement over the
>last ten years suggests that at least MOST of what you don't like has
>nothing to do with the CD standard itself - that HASN'T changed over
>this period. So looking to a new standard to cure the problems you're
>hearing is very likely barking up the wrong tree. The implementation
>of the standard is not the standard itself, and in all likelihood the
>problems have VERY little to do with being "analog" or "digital".
A tech-writer I used to work for had a phrase which applies to this
situation:
you mistake me for someone who cares about the reason it doesn't work.
He would use it whenever we'd run into a bug and the development team
would explain *why* it wasn't working. From his standpoint (as one of
the customer representatives), it's a perfectly valid answer.
I suggested that *if* the standard were more stringent, there would be
more room for implementation error without that error being
audible. This was based on what happened the last time -- do you
really want to go for a decade before a new standard's implementation
is an improvement over what is currently available?
As for your last item, I agree completely, and said so in earlier
posts; the *current implementation* is what I have audible trouble
with. I don't know if the limiting factor is the standard, or just the
implementation; I do know that it seems toe a lower limit than what
was (and is) available on the 1950's analog standard.
Were I on the team developing the next version, I'd do research into
the causes. As it stands, even if I knew where the problems were,
fixing it is outside my control.
--kdc
* From the QA standpoint, I care *why*. From the customer standpoint,
where I am as a consumer of recorded music and live performances, I
really *don't* care *why*. What I do care about is whether it's good
(enough) or not--just that.
L.E.Sixma
Ker...@cris.com (Kevin Connery) wrote:
>We don't have a complete set of quantifiable tests which can be run to
>reliably identify the "good" equipment. Yet.
Maybe we keep on measuring the wrong entity; maybe we should start
again at the points where the Just Notable Differences are measured.
Interesting in this field would be the energy-behaviour of the brain
just below the levels of the JND's. Maybe we can find some
subconscoius brainactivities responsible for not feeling comfortable
with a so and so audio product, allthough objectively it is measuring
right.
Anybody knows if anything like this has been done?
Kind greetings, L.E.Sixma
"Take care not to have any ideas, because they prevent you from
thinking"
Spinoza
Bob Myers
Chris Ross (GEC Marconi Avionics) (chris...@gecm.com) wrote:
> my...@hpfcla.fc.hp.com (Bob Myers), on 29 Jan 1997 16:43:48 GMT wrote:
>> First, the transformation from 20 bits/sample to 16 does not
>> constitute "downsampling", at least as I would commonly use that
>> term, since there is no change in the number of samples per second.
>> Whether or not all of the information in the 20-bit original is
>> preserved in the 16-bit version depends on whether or not there
>> actually WAS information, as opposed to noise, in those lower four
>> bits.
> Hmmm, I don't think that this is exactly how it works. It bothers me
> because it's obvious from this and previous posts of yours that you
> clearly know what you're talking about, so I assume I've either
> misunderstood what you're trying to say or my understanding of
> digital signal processing is flawed. (It's true that I've never
> worked directly with audio DSPs, the closest I've got is to write to
> the firmware for CD-ROM drives). Either way I'd really welcome
> clarification.
OK, here it is.
My point in the above is that there is a difference in information
content between the 16 bit and 20 bit versions of this recording ONLY
IF there was actually 20 bits' worth of information to be had from the
original signal; if these lower bits are actually just encoding random
noise, then nothing will be lost in transferring this recording
(through proper techniques, of course) into 16-bit storage, since
there really wasn't 20 bits of INFORMATION per sample to begin with.
There is a fundamental difference between providing 20 bits per sample
in terms of raw data, and there actually being 20 bits of meaningful
information about the signal in each and every sample.
20-bit systems are used professionally to give a bit of extra space in
each package of data such that processing in the digital domain can
occur without errors (from, for lack of a better term, "rounding the
results" in the math) creeping into the 16-bit final result. 20
bits/sample can also give you more headroom to play with, but that by
itself doesn't say that the full 108 dB or so of range is actually
USED, in that you'll have to toss something out to fit this into a
16-bit product. I can't say anything specific about the recording in
question - and would certainly like to hear more from John or anyone
else who CAN provide knowledgable comment; I just wanted to make the
point that going from 20 to 16 bits/sample doesn't necessarily HAVE to
have any negative impact on the sound.
--
Hamish Hubbard
Chris Ross +44 1383 822131 (chris...@gecm.com) wrote:
: Your reply sounds very much like the perspective of a musician, are
: you a musician?
Yes, but only recently. I play guitar, and my sound is completely
dependent on my tube amp. (I only find tubes acceptable when fidelity
is to be avoided, but that's a different thread :-) ).
: Why not indeed. I would very much like to be able to play music
: myself, in fact I am thinking about learning to play the 'cello.
: However, I will be 35 in a few weeks and I think that it is probably
: thirty years too late to start. Do you think there is any point in
: starting at this age?
This is somewhat off-topic, but being too old to learn an instrument
is a myth. Small children usually progress much more slowly than
people who are older, FWIW. You just have to have patience, an open
mind, and a teacher who does not think you are too old (this one is
really important). Try it, you may find you have been missing out on
something you find more enjoyable than hi-fi.
: Reproduced music will always be valuable. Even assuming that I were to
: master the 'cello, would that stifle my desire to listen to recordings
: of great string quartets? Hopefully not! What about, for instance,
: those people that continue to enjoy the music of Elvis Presley even
: though the poor chap is now dead? Do you feel that they would be
: better served by buying the sheet music and just singing the songs
: themselves? I think not, it is not just the appeal of the songs but
: the precise way in which he performed them that people are paying for.
: It is therefore worthwhile to aim to recreate that performance as
: fully as possible, even if it takes another century to get there.
Of course that's all true, it's just that a recording of Elvis or
anyone else for that matter is a very poor approximation of an actual
'musical event'. Recorded music is vital, it's such a shame about
those we entrust it with, i.e. big record companies.
Cheers,
Hamish
Anthony Clarke
In article <5co98i$5...@eyrie.graphics.cornell.edu>, Bob Myers
<URL:mailto:my...@hpfcla.fc.hp.com> wrote:
> John Hamm (jlh...@lawrence.ks.us) wrote:
>> Wasn't the original statement that the CD was more accurate in a
>> Hi-Fi sense; frequency response, signal to noise, etc.? That
>> doesn't equate to 'musically' accurate. Now before you
>> aim-the-flame consider that a system can "measure good, sound bad"
>> and vice-versa.
> Yes, and that's the whole point - the sound that the LP-philes prefer
> does NOT seem to be due to any increased OBJECTIVE accuracy on the
> part of that medium, but rather due to the specific way in which the
> medium introduces INaccuracies. ^^^
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Um .... Are you saying that CD does
NOT introduce INaccuracies or dare I say even fewer than the analogue
medium, if so why does the Digital world have devices such as error
correction, filtering, quantitisation, and the latest
craze... anti-jitter !!
> all that surprising, if the notion that the LP can provide "euphonic"
^^^^^^^^
Geez.... this could go back to the argument of first order V's second
order harmonic distortions debate !!
> distortions is accepted. And we have yet to see any evidence which
> would suggest that this sort of distortion is NOT occuring.
Or "IS" occuring. ??? perhaps :-)
--
Anthony Clarke | Oak Seed Computers Pty. Ltd. |
mailto:ant...@om.com.au | Phone / Fax (066) 246 200 |
vince doran
Mr. Siu:
Considering the effort you have already put into this problem, I
suspect I profit more from this conversation than you do. And to
maximize my gain, I have some more questions:
Several posts to other threads have noted that they obtained best
integration of sub wi. mains by not hi-passing the mains, instead
running them full range while low-passing the sub. I have also found
this to be the case, with mains that are nominally -3db@30hz, but
which I adjusted using their integral eq to be -12db@30hz. Since the
s'phile reviewer of your mains noted "bass extension to 25hz" and you
mention "losing the bottom octave", I assume they begin to roll off
below ~40hz, but slowly. Have you tried this setup with the sub
low-passed at or below 40hz?
With respect to time-alignment, when I said "assuming no delays in the
rest of the system" I was of course neglecting the likely difference
between the delay of the subs low-pass (given for a recently reviewed
M&K sub as ~11ms) and the delay of the low-pass part of your mains'
crossover. Are you able, with the equipment and room you have, to
measure the actual delay before driver output for both systems and
then adjust the speaker-to-ear distance to compensate? And is this the
alignment you found to have the best "THUMP" response?
With respect to my partially-in-jest suggestion that we might induce
Mr. Dunlavy to comment on these issues; I have no idea of his
willingness to educate the public, but I have noted that Ken Kantor of
NHT has been very generous of his time and knowledge on the DIY Bass
mailing list in the past, perhaps we could hook him in.
Seriously, in fact, I think we should put this proposition to the
newsgroup: Some of you must have positions/aquaintances among the
designers and manufacturers or personal expertise. Many of you must,
in this age of home theater, have subs with decent extension below
20hz, and like Mr. Siu, very good mains which do not go quite that
low. From the designers' point of view, the "best" way to obtain that
extra bass extension is with something like the Dunlavy SC-VI, where
the designer/builder can control most parameters. From the user's pov,
however, spending that additional >$20k may not be the "best" idea
their wife/accountant/psychiatrist ever heard. So: Given the
preceeding asssumptions, could a package of crossover capability,
testing regime, and placement advice be developed which would more
reliably optimize the benefit of sub integration, and is there liable
to be net benefit?
Finally, Mr. Siu, it was your comment about your personal
unquantifiability (sp?) that provoked the "EUREKA!" This is what's
been missing all along! The AUDIOPHILOMETER! With such a device in
hand, we would be able to factor out reviewer vagary and finally have
meaningful measurement of the entire audio chain! Hm, let's see, we'll
need a log scale, to accomodate the Hirsch-Scull range, and some
brainwave pickups to provide real quantifiable measurements, and then
just a tad of calibration research, and... well, it's beyond me, but I
think it's just the project for some of the magicians, er, scientists
in the audio field. Krytonite wiring, of course, in Mobius-strip
configuration. Until then...
Enjoyable listening
Vince Doran
Richard D Pierce
In article <5d542d$n...@canyon.sr.hp.com>, Kevin Connery
<Ker...@cris.com> wrote:
>In article <5d2ukm$t...@agate.berkeley.edu>, John Ongtooguk
><jo...@hp-vcd.vcd.hp.com> wrote:
>> Even when using senses that we tend to have more confidence in,
>> such as our eyes when viewing images, it seems common to fall
>> back on subjective evaluation of quality in spite of the
>> number of methods that are available for measuring performance.
>I think that is due to the current situation where something which
>measures as 'perfect' as we can measure doesn't necessarily sound
>good--indicating that we're missing at least one critical measurement.
Interesting premise, unfortunately rendered pretty useless by the fact
that absolutely NOTHING measures perfect in ANY objective measurement.
>Certainly, the quantifiable measures can be used to 'disqualify'
>devices which are NOT good enough, due to clear, measureable, defects,
>but the converse is not yet true.
Again, NO MEASUREMENTS available today show perfect results.
>Hearken back to the early days of the CD, where all the tests being
>done--the ones based on earlier models--showed the sound to be as good
>as it could get.
And, praytell, exactly what measurements were these and where is such
data published?
I ask because ALL of the literature in the professional press at the
time completely contradict your assertion that "all tests being done
showed the sound to be as good as it could get." People in 1984 were
measuring gross group delay errors due to analog anti-imaging filters,
non-monotonic behavior of ADCs and DACs and a whole raft of other
problems. Jitter was a phenomenon understood 3 decades prior to the
introduction of CDs.
Basically, your premise that things did or now measured "perfect" is
completely unsupported by ANY objective data or history for that
matter, unless we want to drag out the old, tired strawman of "perfect
sound forever" which was a marketing and press declaration and a
position NEVER held in the engineering community.
>Over time, other measurements were identified which
>showed differences in sonic (and other) behavior.
No, over time, the popular press began to understand issues and
measurement criteria that were in regular use by the professional and
engineering community decades earlier
>We don't have a complete set of quantifiable tests which can be run to
>reliably identify the "good" equipment. Yet.
Maybe, maybe not. Rather, the high-end press is still attempting to
use the strawman of manufacturers "specifications" (which are NOT
measurements) not representing sound as a whipping boy.
That we may or may not fully understand the mechanisms behind hearing
and preference is one issue that bears intelligent discussion. On the
other hand, assertions like "measurements show 'perfect' performance"
is simple nonsense that is unsupported by the real data.
I have NEVER measured ANY piece of equipment that on ANY measurement
(pick any one you want, frequency response, THD, S/N, separation,
spectral composition of distortion, group delay, whatever) has even
REMOTELY measured perfectly.
Such measurement have not, do not and cannot exist and no professional
or engineer has ever seriously made auch an assertion.
--
| Dick Pierce |
| Loudspeaker and Software Consulting |
| 336 Broadway Hanover, MA 02339 |
| (617) 826-4953 (Voice and FAX) |
Kevin Connery
>Of course the LP is a less accurate reproducer of the master. The
>reason listeners "prefer" the LP to the real sound is because they
>hold a psychoacoustic bias toward what they "already know." They
>prefer what they have already come to regard as good sound and are
>reluctant to give up the reference even faced with overwhelming
>evidence that it is wrong (sounds "more" like the master.)
The question I took as key to the original post was whether or not the
conventional measurements were sufficient--but the answer given here
changes that into "of course, x is less accurate"--without any
reasoning provided to show that maybe, just maybe, we're missing one
or more important quantifiable measurements.
It's just as easy--and just as valid--to claim that people like CDs
because they're new-and-exciting, or because they're new-technology,
or whatever, as it is to make the unsupported claim that people hold a
psychoacoustic bias towards what they already know. More valid, in
fact, for there are many people who grew up with CDs who are just now
switching to analog.
Whether either claim is *correct*, I don't know. But to hand-wave away
the key question in favor of unsupported rhetoric isn't going to
educate anyone.
If we consider two media, both perfect except for *one* detail each,
it's not at all unlikely that both would have supporters, because what
is important to some listeners isn't so for others. (reference
imaging, bass slam, rhythm, impact, pace, frequency response, etc.;
which of these are most important to you--so much so that you
knowingly choose to ignore flaws in one or more of the others when
you're auditioning equipment?)
Please note; I'm not saying either LPs or CDs are perfect except for
one difference; neither is anywhere near that good. But it is clear
that the distortions are *different*. Which means that, even if we had
perfect and complete measurements (which we do NOT), there's no way to
state that X is better than Y for all listeners.
>Worse performance is only accepted as "preferable" by champions of the
>old guard who have such strong psychologial ties ties to the existing
>performance level that they are willing to reject reality and "vote"
>on emotion...not performance.
See above; a lot of new buyers of vinyl are in their early 20's--
hardly 'old guard'. If they have such psychological ties to the analog
medium, it's not through long association, for most never heard LPs
much past childhood. And switching is expensive. (Which argument can
also be used the other way, granted.)
--kdc
Curtis Leeds
Arnold B. Krueger wrote:
> The bottom line is that cutting and playing back a LP is an artistic
> venture that involves changes to sound character that is at least the
> same order as mixing down the original master.
I don't agree with this at all. A mixdown from an original master to
two track does indeed involve artistic judgement. On the other hand,
the creation of an LP master from tape - while fraught with
difficulties and limitations - should not be an artistic exercise. In
the end, the LP should sound like the master tape. If the job is well
executed, the LP can be very close to that ideal.
> One friend of mine who routinely listens to LP's on a system designed
> to exploit CD's has simply added a filter to eliminate excess low
> frequency noise from the LP side. In the old days we called these
> "rumble filters".
They're still known as rumble filters. With today's best turntables
and arms, they are just not necessary.
--
********************************************************
Curtis Leeds cle...@mail.idt.net
"A man hears what he wants to hear, and disregards
the rest."
********************************************************
Arnold B. Krueger
[Moderator's Note - PLEASE note that Mr. Krueger is very carefully
talking about EARLY digital. Don't, for god's sake, post a lot of
blather indicating that it's not that way anymore. That is not the
point. Thank you. -- bt]
Peter Irwin <pir...@ktb.net> wrote
> From:
> Andrew Marshall. Audio Ideas Guide. November 1983. p.28
>
> First of all, I have concluded that, whatever may
> be wrong with early CD players, it is not the digital
> system itself. The same standards apply to the digital
> processors as to the the Compact Disc system, yet the
> sound of digital tapes is quite different from that of
> early CD machinery...
In retrospect, there were many effects in play here:
(1) Some early CD's were later alleged to have been made from tapes
that were "cutting masters". These tapes were preprocessed with what
it takes to get the best possible sound out of vinyl, and therfore
inappropriate for digital encoding if best possible subjective
impression is the goal. Some titles were remastered, either in secret
or public claims were made.
(2) Some early CD's were digitally encoded on equipment that was
substandard. I recollect a well known authority who was investigating
the issue, telling me, at the time, that analysis of the digital data
on some CD's indicated that only 12 bits were encoded, and the
remianing lower order bits were always the same in every sample on the
disc.
(3) Some early CD's had only one DAC that was timeshared between two
channels, introducing an approximate 22 microsecond delay between
them. If you had program material with lots of high frequency
information and a center channel speaker (or its equivalent), then
audible frequency response variations would result.
(4) Some audiophile systems had the bandpass, but not the power
handling capacity to handle the far broader power spectra that was
possible with CD's, and not previously possible with LP's. They would
distort or destruct. Subwoofers and high power capacity tweeters were
popularized after CD's were introduced.
(5) Some audiophile systems had their frequency response tailored to
compensate for the uneven response of LP playback. Some cartridges had
audible peaks and others were audibly rolled off. Playing truely flat
responding equipment such as CD players through sytems tuned for vinyl
could result in unbalanced sound.
John Busenitz
On 3 Feb 1997 sdura...@aol.com wrote:
> I think you'll find that those who rely on measurements will support
> the idea that the preference for LPs is based on its non-linearities
> being more euphonic (though it's wholly ironic that those who believe
> this don't, as a rule, share in the preference for that euphony).
I agree, but not just "those who rely on measurements" would say this.
What other discourse is there? LPs are less faithful and thus less
accurate than CDs, but they are still the medium of choice for a small
but vocal group. Thus the only choice is that LPs have more
distortion, but a pleasant kind that fools people into thinking they
are more realistic.
> Those relying more on subjective observation can't seem to shake the
> feeling that analog reproduction is more "right" because of certain,
> as yet undocumented, subtractive effects in digital playback.
But not all those who rely on subjective observation. Just some. And
those that do feel that "analog reproduction is more right" can easily
have their "feelings" attributed to things other than
fidelity. Inherent psychological bias, for example.
> The motives for preference need to be better understood. What causes
> one person to prefer euphonics, another accuracy? I have observed
This seems to me to be the basis for two schools of audiophilia. Some
seek faithful audio reproduction, whether it sounds "euphonic" or not.
Others wish for a pleasant sound; entertainment.
> that those coming from a strong technical background will prefer
> accuracy, while those with weaker technical backgrounds will allow for
A general statement which is often, but not always, true.
> more euphony. For example, there are a lot more EEs strongly in the
> digital, solid-state camp than in the LP, single-ended tube camp. My
> personal view is that their expectation is driven by their knowledge
> and experience.
Would you mean that they expect vinyl and tubes to be less accurate,
and so they have a bias or prejudice when they listen to these more
euphonic systems? Would this mean that their subjective evaluations
are polluted by their preconceived notions? Would this in turn suggest
that subjective evaluation is an inherently colored and flawed method
of evaluation, because they perceive what they want to or think they
will hear?
> On the other hand, I have also found that many who prefer LPs have
> huge (and I mean huge) music collections. This is NOT to say that EEs
> have no music collections. But, on balance, those driven by the sense
> that the hardware merely serves the software are much more liberal in
> the types gear they end up with. Notions of technical accuracy are
If the software is unfaithful to the original, who cares what the
hardware is? I agree, those who think vinyl is the only way are
equipment lovers; they can only seriously listen to music on the
"right" $10000 record player.
> secondary to how well the gear brings forth the musical treasures of
> their collection.
Well, for someone who wants "the musical treasures" to be reproduced
faithfully, the two are one and the same.
> So what it comes down to really is diff'rent strokes for diff'rent
> folks. It appears that those present for your listening sessions had
> a preference for LP playback because of their expectation of what is
> good reproduced sound. I suspect if you brought together a different
> group, with different expectations (say folks who hang with Tom
> Nousaine or Peter Aczel), you'd end up with a strong preference for
> digital playback.
So does this mean that you agree that subjective evaluation is subject
to the listener, and thus the evaluation is inherently and hopelessly
colored? Obviously, as you point out, the recording and equipment
wouldn't change, but with different listeners the evaluation is
different. There is no other recourse than to proclaim the invalidity
of subjective evaluation. Thanks for the help!
> In the final analysis, it appears there is no "correct." And it looks
> like we'll have to leave it at that, even though we may be sitting in
> front of our turntables thinking "how in the world can anyone prefer
> CDs?"
And we'll be sitting in front of our CD players wondering, "How in the
world can anyone call themselves an audiophile and be merely
interested in pleasant, euphonic sound rather than a faithful
reproduction of the original performance?"
_____________________________________________________________
John Busenitz buse...@ecn.purdue.edu
P.U. ECE http://cernan.ecn.purdue.edu/~busenitz
Disclaimer: My statements do not represent Purdue University.
F. Blaine Dickson
Nousaine <nous...@aol.com> wrote:
> Of course the LP is a less accurate reproducer of the master. The
> reason listeners "prefer" the LP to the real sound is because they
> hold a psychoacoustic bias toward what they "already know." They
> prefer what they have already come to regard as good sound and are
> reluctant to give up the reference even faced with overwhelming
> evidence that it is wrong (sounds "more" like the master.)
That would be the case only if the people who preferred the LP
listened to it over a period of time for it to "engrain" in their
accoustical memory as the "right" or "accurate" sound. However, this
does not account for the fact that there are many people like myself
who have grown up on CDs, but still prefer the LP sound because it
appears, to me at least, to be "richer" and have more "presence" than
my $550 Denon DCD 620 player and any of the CDs I own.... and I own
300+ CDs :-)
Of course this would depend on what system you were listening to since
in my experience I have found a few CD players that give an LP's
"presence," but they are +$4000 (ie. Linn Numerik/Karik and the YBA
2).
>Again...if even your special listeners "hear" the digital
>format as delivering the most accurate rendition of the master WHY are
>you asking the "preference" question? If it is more accurate; it is
>more accurate.
But that does not make it *better* I'm afraid. What might be a more
important question is if the digital format was more accurate
relatively speaking, why do people, especially those who never got
*used* to the LP, *prefer* the LP sound and regard it to be more like
the live performance?
*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
F. Blaine Dickson, B.A.
CPGA ELITE Golf Professional
Kelowna BC Canada
Take time to dream.
*~*~*~*~*~*~*~*~*~*~**~*~*~*~
Kurt Strain
Nousaine (nous...@aol.com) wrote:
: Of course the LP is a less accurate reproducer of the master. The
: reason listeners "prefer" the LP to the real sound is because they
: hold a psychoacoustic bias toward what they "already know." They
[cut -- rgd]
This of course discounts the thousands that have switched from their
old preference of CD's to LP's. Why let some evidence stand in the
way of absolutism?
Kurt
Michael Ford
In article <5d7rhs$g...@eyrie.graphics.cornell.edu>, DPi...@world.std.com
(Richard D Pierce) wrote:
> In article <5d542d$n...@canyon.sr.hp.com>, Kevin Connery
> <Ker...@cris.com> wrote:
>
> >In article <5d2ukm$t...@agate.berkeley.edu>, John Ongtooguk
> ><jo...@hp-vcd.vcd.hp.com> wrote:
>
> >> Even when using senses that we tend to have more confidence in,
> >> such as our eyes when viewing images, it seems common to fall
> >> back on subjective evaluation of quality in spite of the
> >> number of methods that are available for measuring performance.
I think the fundamental difference between a painting and a photograph
may also apply to sonic events. Both have images they can bring the
viewer into some recreation of the original event, one with very
selective fidelity relying more on artificial creations that are
intended for our brains to decode into more complete illusions than
the stricter fidelity of the other media may be able to provide.
> >I think that is due to the current situation where something which
> >measures as 'perfect' as we can measure doesn't necessarily sound
> >good--indicating that we're missing at least one critical measurement.
What you say isn't fact, it is the point of most of the debate.
> Interesting premise, unfortunately rendered pretty useless by the fact
> that absolutely NOTHING measures perfect in ANY objective measurement.
Maybe a better term would be "perfect" enough for audio, and the
limits recognized as typical of what may or may not be audible.
> That we may or may not fully understand the mechanisms behind hearing
> and preference is one issue that bears intelligent discussion. On the
> other hand, assertions like "measurements show 'perfect' performance"
> is simple nonsense that is unsupported by the real data.
I agree, the whole of the issue is what is audible. Connecting direct
measurements with subjective perception.
Peter Irwin
John Busenitz wrote:
>The fact that it "measured good" (measured well?!) merely shows
>that a complete and properly-interpreted set of measurements was not
>done.
To say that a piece of equipment "measures good" may not be very good
English, but it is from a very old audio saying.
C. J. LeBel said back in 1938, "If it measures good, but sounds bad -
it is bad." (I have not seen the original of this, but the quote was
given this way in Audio in the mid 1950s by C. G. McProud and I am
sure that McProud got the quote right.) You may find it interesting
that C. J. LeBel was the founding president of the Audio Engineering
Society.
You are right that much of what this statement originally meant was
that simplistic objective measurements do not always get to the heart
of the matter and that complex objective measurements which take into
account both how equipment performs under various conditions and the
fact that the ear is sensitive to different types of distortions in
wildly varying degrees.
It also meant that controlled subjective testing with trained
listeners was necessary.
The phrase is clearly NOT meant to mean that we don't need controls.
I think that perhaps the bungled Chinn and Eisenberg bandwidth
preference tests (Proc IRE Sept 1945) were a major contributing factor
in the U.S. audio industry's general disregard for proper controlled
testing. The BBC appears to have done more in the way of controlled
subjective testing in audio during the 50s and 60s than the entire
U.S. audio industry.
>However, I am curious as to how Mr. Hamm and others would distinguish
>betweent technical and "musical" accuracy, and define the latter. I
>would think that the definition of "musical accuracy" would change
>quite a bit from audiophile to audiophile.
You might try this quote from H. A. Hartley
When I go to a concert I hear things which please me or
annoy me, but whatever I hear is the result of people
consciously making music. What I want in my home is as
close an approach as possible to the sensation created in me
when I go to a concert."
From: H. A. Hartley. "The great loudspeaker mystery" Audio
Engineering. January 1953.
Peter.
----
Peter Irwin
pir...@ktb.net
Michael Ford
> Scott Frankland wrote:
>
> > Well, here's a story for you. RCA cut a record in the '50s known as
> > "Also Sprach Zarathustra". Only about 1000 to 1500 copies were
> > produced according to informed sources. Of these, all were mastered
> > from the same acetate (1S) and from the same stamper (A1) at the
> > Indiana pressing plant (I). On numerous occasions, we have noticed
> > marked differences in sound quality from apparently mint copies. How
> > to account for this? Well, the only difference I can think of is due
> > to stamper wear.
It is also possible that they were only "apparently" mint
copies. Having "cleaned up" a number of old dirty LPs I can say that
most of the time after a good wash the Vinyl looks as good as
new. Without a microscopic examination (which I don't have a clue
about), the way I would judge a LP is how does it sound after
cleaning, not what does it look like. Visual bias you know, terrible
stuff.
Steven Abrams
sdura...@aol.com writes:
> I think you'll find that those who rely on measurements will support
> the idea that the preference for LPs is based on its non-linearities
> being more euphonic (though it's wholly ironic that those who believe
> this don't, as a rule, share in the preference for that euphony).
> feeling that analog reproduction is more "right" because of certain,
> as yet undocumented, subtractive effects in digital playback.
You're missing the primary conclusion of Mr. Atkinson's post -- namely
that all people, regardless of their background -- prefered the vastly
more accurate 20-bit rendition, "even" over the LP. So all arguments
about accuracy vs. euphony based on this test are off. If you buy
that his methodology can be used to draw any conclusions, and if you
assume that he accurately represented the preferences expressed by his
listeners, there is only one possible conclusion to be drawn: The most
accurate playback was prefered by everyone present.
So much for "undocumented subtractive effects in digital playback."
That being said, I must take issue with your repeated
mischaracterizations of a group of people which includes myself.
You *insist* on using the phrase "those who rely on measurements" to
represent those who prefer the sound of a more accurate rendition;
similarly, you use the phrase "those relying on subjective
observation" to represent those who prefer LPs. This is an incorrect
characterization, and one which I'm getting pretty tired of hearing.
Those who prefer LPs are completely characterized by the following
phrase: "those who prefer LPs." Similarly, Those who prefer a more
accurate rendition are characterized by the phrase "those who prefer a
more accurate rendition." Since this is a high-end audio group, we
should dispense with the notion that anybody here "relys on
measurements" to determine whether or not they will like something.
We all listen; we all use our ears; We're ALL interested in the sound.
Some of us prefer it one way, and some prefer another. That's all.
[...]
> huge (and I mean huge) music collections. This is NOT to say that EEs
> have no music collections. But, on balance, those driven by the sense
> that the hardware merely serves the software are much more liberal in
> the types gear they end up with. Notions of technical accuracy are
> secondary to how well the gear brings forth the musical treasures of
> their collection.
Oh, here we go again.
I venture to say that if you take the N listeners who participated in
Atkinson's "test" and took the average size of their music
collections, it would dwarf what we consider to be a "big" record
collection. It would probably fall in to the "huge" category. Yet
they ended up prefering the most accurate rendition. Ooops.
> So what it comes down to really is diff'rent strokes for diff'rent
> folks. It appears that those present for your listening sessions had
> a preference for LP playback because of their expectation of what is
> good reproduced sound. I suspect if you brought together a different
> group, with different expectations (say folks who hang with Tom
> Nousaine or Peter Aczel), you'd end up with a strong preference for
> digital playback.
You missed the point. There WAS a preference for digital playback.
20-bit digital playback. If you accept that the listening conditions
were such that conclusions could be drawn, this is the conclusion for
that recording, by that group, on that equipment.
> In the final analysis, it appears there is no "correct." And it looks
> like we'll have to leave it at that, even though we may be sitting in
> front of our turntables thinking "how in the world can anyone prefer
> CDs?"
You say that your conclusion is that there is no "correct" -- although
your implication *clearly* is that those who own music, like music,
and actually listen prefer it YOUR way, and those who don't listen,
take measurements, and don't care about the music prefer it the other
way. Bullshit.
The one thing you said that makes sense is that, in matters of
preference, there is no right or wrong.
Some like it this way, some like it that way, some like it the other
way.
~~~Steve
--
Steven Abrams abr...@cs.columbia.edu
Living is easy with eyes closed, misunderstanding all you see.
-Lennon/McCartney
Scott Frankland
Michael Ford wrote:
>
> It is also possible that they were only "apparently" mint
> copies. Having "cleaned up" a number of old dirty LPs I can say that
> most of the time after a good wash the Vinyl looks as good as
> new. Without a microscopic examination (which I don't have a clue
> about), the way I would judge a LP is how does it sound after
> cleaning, not what does it look like. Visual bias you know, terrible
> stuff.
Well, you can be sure that we cleaned the bejeezus out of each copy
before we listened to it. I doubt that we were listening to a worn
copy in either case because there were very few spindle trails or
other tell-tale signs of use.
~SF~
Curtis Leeds
Steven Abrams wrote:
> You're missing the primary conclusion of Mr. Atkinson's post -- namely
> that all people, regardless of their background -- prefered the vastly
> more accurate 20-bit rendition, "even" over the LP. So all arguments
> accurate playback was prefered by everyone present.
>
> So much for "undocumented subtractive effects in digital playback."
Sorry, but this is faulty reasoning. It is incorrect to conclude from
this one test that there are no subtractive distortions in digital
playback. The test only proves that - in this instance and with this
recording reproduced on the equipment at hand - all the listeners
preferred the 20 bit digital to the LP.
Bob Myers
Kevin Connery (Ker...@cris.com) wrote:
> I suggested that *if* the standard were more stringent, there would be
> more room for implementation error without that error being
> audible. This was based on what happened the last time -- do you
> really want to go for a decade before a new standard's implementation
> is an improvement over what is currently available?
You miss the point. The problem is that people are calling for a new
standard without realizing that the evidence suggests there's nothing
wrong with the old one. We hear loud and long cries for 24
bits/sample at 96 kHz sampling rate, without ANY acknowledgement on
the part of the people proposing this that (a) the problems they're
reporting with the current standard, such as the sonic performance of
DACs and clock jitter and so forth will, if anything, be MORE
DIFFICULT to address under such a standard, and (b) these problems
weren't caused by, and have no relation to, the parameters of the
current standard in the first place.
> As for your last item, I agree completely, and said so in earlier
> posts; the *current implementation* is what I have audible trouble
> with. I don't know if the limiting factor is the standard, or just the
> implementation; I do know that it seems toe a lower limit than what
> was (and is) available on the 1950's analog standard.
Well, at least you're acknowledging that you don't know what limits
current performance. That's more than a lot of digiphobes are doing.
Given this position, I would hope that you're NOT among those calling
for a change to the current standard. Or do you think that a new
standard would be easier to implement properly than the current one
simply because it's new? If anything, we should expect the opposite
to be true. I'd rather fix the remaining problems with current
implementations - which again AREN'T related to the standard - than to
have to start all over from zero.
> Were I on the team developing the next version, I'd do research into
> the causes. As it stands, even if I knew where the problems were,
> fixing it is outside my control.
And what if those problems come from things not addressable in the
standard for the recording format? Not EVERYTHING is covered by such
a standard, or should be.
sdura...@aol.com
In article <5d5r0q$7...@agate.berkeley.edu>, Jim Andrews
<jand...@activepower.com> writes:
>This one little paragraph is packed with so much subtle snobbery and
>innuendo that it's difficult to figure out where to begin . . .
I'm sorry if that's your interpretation - snobbery and innuendo wasn't
the intent.
>1. The implication that many LP lovers are TRUE music lovers, as
> evidenced by their large LP collections (which brings to mind
> some rather puerile contests of another sort), completely
> ignores the fact that a lot of us simply cannot AFFORD a huge
> music collection. I'm pleased that I have managed to acquire
> maybe 600-700 recordings in my life (about a 50/50 split CD/LP),
> but noone I know of would think of me as anything BUT a music
> lover. Do I still get to play?
I think it's really a question of desire. I have been acquiring used
LPs at an average price of about $4 per record for over twenty years.
Not a week goes buy when I don't visit a record store or garage sale
and score $20-30 worth of records. Am I a snob because I spend $100
month on records and you choose not to?
>2. I'm sure that if you surveyed the truly "huge" privately-owned
> music collections in the world, you'd find that nearly all of
> them are dominated by LP. So what? Given that very few of us
> can afford to acquire recordings at a rate that would allow one
> to build such a collection in a short time (i.e., with CDs), it
> stands to reason that such collections have been acquired over
> a very long period of time. Let's say a "huge" collection is
> one containing at least 10,000 recordings. At a rather prodigious
> consumption rate of 500 recordings a year, that would take 20
> years . . . and I'd be surprised if CDs accounted for 1/3 of such
> a collection, even if the collector "got on board" relatively
> early in the digital age. I'm willing to bet that in most cases,
> it's economics that have driven the collection, not preference.
Well CDs have been available for going on 15 years now. Does anyone
have a 10,000 CD collection? 5,000? No, people buy them, get bored
with the sound and trade them in for new ones. I don't see any
changes to this trend.
>3. Please stop referring to your antagonists as "EEs". Unless you
> have a very legitimate reason to speak to that very particular
> subset of the "objectivist" camp, it is far too specific, and
> it doesn't take an EE degree to figure out what accuracy means.
> It also insultingly implies that a technical education somehow
> affects one's ability to hear the "truth", but that your own
> preference isn't constrained by the blinders of technology.
The "blinders" of technology work both ways. I could equally assert
that your expectation of good sound is colored by advance knowledge
that your playback gear is technically accurate. This argument
doesn't get anywhere.
You might take a moment to ponder, though, that we are fifteen years
into the introduction of the CD and yet the LP is alive and
flourishing, with new adherents every day (just today I discovered
Stretch Records releasing Chick Corea's latest on "High Density
Vinyl"). By stark contrast, fifteen years after the introduction of
the LP (ca 1963), you were hard-pressed to find any serious listeners
sticking to 78s. Can you take YOUR blinders off for a second and
contemplate the implications of this fact?
Siegfried
John Ongtooguk
Steven Abrams (abr...@cs.columbia.edu) wrote:
: You're missing the primary conclusion of Mr. Atkinson's post -- namely
: that all people, regardless of their background -- prefered the vastly
: more accurate 20-bit rendition, "even" over the LP. So all arguments
[cutting -- rgd]
: So much for "undocumented subtractive effects in digital playback."
But most of us will have to be content to listen to 16 bit digital
as played back thru a CD player instead of 20 bit off of a tape,
and since the LP was preferred to the 16 bit CD the primary
conclusion is the question why ? What's missing on the 16 bit CD
and/or what's been added to the LP ? Also, why did the 16 bit CD
sound different than the 16 bit feed ? Again what's being lost on
the CD ?
John Ongtooguk (jo...@vcd.hp.com)
Michael Ford
In article <5d542d$n...@canyon.sr.hp.com>, Ker...@cris.com (Kevin Connery)
wrote:
> In article <5d2ukm$t...@agate.berkeley.edu>,
> John Ongtooguk <jo...@hp-vcd.vcd.hp.com> wrote:
>
> > Even when using senses that we tend to have more confidence in,
> > such as our eyes when viewing images, it seems common to fall
> > back on subjective evaluation of quality in spite of the
> > number of methods that are available for measuring performance.
>
> I think that is due to the current situation where something which
> measures as 'perfect' as we can measure doesn't necessarily sound
> good--indicating that we're missing at least one critical measurement.
I suggest the critical thing missing is that "stereo", regardless of
the quality isn't reality, it is a fragment of the real acoustic event
that our brains can "pretend" is real and imagine what the real event
might have been like. Our brains fill in gaps in the information, and
ignore artifacts of the recording VERY well. No stereo speakers have
an accurate image, they have a "credible" illusion of imaging.
I think we make a VERY false assumption if we assume ACCURACY in some
arbitrary fragment of the acoustic event will always recreate the best
illusion of the full event. A simple example is a minimonitor. Most
people prefer the sound of such low end limited speakers with a slight
bump in the frequency response at the low end they can create to
compensate for the lower low end they can't. Why we can't just accept
some level of nonfidelity as a preference and move on is beyond me.
Michael R. Hobaugh
>>The reason listeners "prefer" the LP to the real sound is because they
>>hold a psychoacoustic bias toward what they "already know."
>It's just as easy--and just as valid--to claim that people like CDs
>because they're new-and-exciting, or because they're new-technology,
>See above; a lot of new buyers of vinyl are in their early 20's--
>hardly 'old guard'. If they have such psychological ties to the analog
>medium, it's not through long association, for most never heard LPs
>much past childhood.
I'm 27 yo and used to listen to records as a kid on my brother's Sears
all-in-one piece of garbage and my own Panasonic a-i-o-p-o-g. It was with
great excitement that in 1986 at the age of 16 I bought my first CD player
and started buying CDs. At that time I was "Digital, totally;" completely
sold on the superiority of CDs and could quote line and verse why they
were better than LPs.
From age 16 to 24 I bought only CDs and loved their rainbow gleam in
the light, the crystalline purity of the sound. Then I got hold of a
used Toshiba belt drive TT to listen to some of my own records
and... the bass was bloated and imprecise. Some of my records popped
and clicked, but despite the obvious sonic limitations of this set up,
I enjoyed listening to music in a way I hadn't in years. I was still
listening to CDs most of the time though. The next year, I bought a
Rega Planar 3 with BPS and now the bass was tight and full, pops and
clicks, as well as overall "hum" was _greatly_ reduced and records
have since come to dominate my listening choices. Pretty much
whenever I'm actually going to sit and just listen to music, it is
vinyl. I don't know why, but it is much more satisfying to listen to
for me. "Maybe a yearning for the innocence of childhood," you say?
Maybe.
But I still think that everyone out there owes it to themselves to
hear a good turntable set up... it makes music magical again.
MRH
Arnold B. Krueger
Arnold B. Krueger wrote:
> While 45kHz is supersonic, a large moderately damped resonance there
> can have measurable (and audible) effects at 15 kHz and below.
Scott Frankland <audi...@ix.netcom.com wrote:
> I'm not sure how, unless the resonance is also non-linear. In that
> case, IM products can appear as difference frequencies. Is this
> what you mean?
Nonlinearity is not requried. A reasonably well damped simple
resonance will have substantial frequency response effects even two
octaves away. In the case of 45 kHz, that is 11.25 kHz, which is in
an area where many ears are quite sensitive to small changes.
The same effect relates to simple roll-off's. I recall doing some ABX
tests showing that a simple R-C low pass filter that measured 3 dB
down at 35 kHz was clearly audible with musical program material.
Later on we found that sharp cut-offs at 22 kHz were hard to
detect. This indicates that the 35KHz filter's effect at 35 KHz was
probably not what people were hearing. It was the effect at lower
frequencies, which while probably less than a dB, affected a broad
range below 18 kHz.
Other ABX tests showed that deep (many dB's) dips over a narrow range
were not nearly as audibly problematical as shallow (fractional dB)
dips over an octave or more.
Michael Ford
In article <5d7tm5$h...@eyrie.graphics.cornell.edu>, Curtis Leeds
<cle...@idt.net> wrote:
> Arnold B. Krueger wrote:
>> One friend of mine who routinely listens to LP's on a system
>> designed to exploit CD's has simply added a filter to eliminate
>> excess low frequency noise from the LP side. In the old days we
>> called these "rumble filters".
> They're still known as rumble filters. With today's best turntables
> and arms, they are just not necessary.
Tell that to your now EX-subwoofer the next time you misscue an LP, or
the tail out groove ends with a clunk. Actually if you are playing an
LP might as well turn off a "true" subwoofer anyway, right?
BTW I think it is surprising how good and expensive a turntable needs
to be before a rumble filter doesn't improve the basic sound. Many
amps and speakers really don't care for subsonic noise at all.
Michael Ford
In article <5d5dq8$3...@agate.berkeley.edu>, nous...@aol.com (Nousaine)
wrote:
> Of course the LP is a less accurate reproducer of the master. The
> reason listeners "prefer" the LP to the real sound is because they
> hold a psychoacoustic bias toward what they "already know." They
> prefer what they have already come to regard as good sound and are
> reluctant to give up the reference even faced with overwhelming
> evidence that it is wrong (sounds "more" like the master.)
>
> Worse performance is only accepted as "preferable" by champions of the
> old guard who have such strong psychologial ties ties to the existing
> performance level that they are willing to reject reality and "vote"
> on emotion...not performance.
I think it is also possible that the sound of the LP may compensate or
compliment some elements of current recording practices. Almost all
recording practices where created in the age where a LP system could
be expected in the playback chain. It is likely from this that those
practices would have some consideration for the final playback and
adjustments would be made to tailor the sound for it. The heart of
this would be the notion that the mastering engineer would not try to
make the recording sound its "best" in the control room, but would aim
for some target sound quality that would sound best in the expected
playback chain. I think it might even be possible that over time
engineers might be selected based on a natural preference for a target
sound that resulted in the "best" LP playback sound.
Substitute the word, "reverb" for LP and see if this logic doesn't
hold. If all engineers knew a reverb unit was part of most playback
chains, and all critical ones, then a slightly "dry" mix would be a
desirable target.
The hard part I think is to define what an LP does to the sound. A
carefull analysis of the whole LP process might be instructive to the
nature of this debate. Once defined then the question would be, do we
add this change to the CD process, or try to remove the complimentary
aspects from the recording process?
Scott Drysdale
>Well CDs have been available for going on 15 years now. Does anyone
>have a 10,000 CD collection? 5,000?
let's see. 5000 CDs divided by 15 years is 333 1/3 CDs per year (hey,
isn't that 10 times faster than an LP spins?). divide by 12 and you
get 27.77_ CDs per month. multiply by $15 (a WAG as to average CD
prices) and you get $416.67 a month. how many people can throw this
much money away per month on recordings? would they have time to ever
*LISTEN* to them more than once? do think you could actually pick 28
music selections per month that you would actually WANT to listen to?
>No, people buy them, get bored with the sound and trade them in for
>new ones. I don't see any changes to this trend.
please, oh all-knowing speaker for "people," tell me where the data
for this assertion came from. you obviously have your fingers on the
pulse of the industry, having polled everyone putting a CD in the USED
rack as to their reasons.
words cannot express (at least not in this newsgroup) how *TIRED* i am
of you presuming to represent some kind of majority all the time.
>The "blinders" of technology work both ways. I could equally assert
>that your expectation of good sound is colored by advance knowledge
>that your playback gear is technically accurate. This argument
>doesn't get anywhere.
sure it does. that's why there's BLIND TESTING. test master tape vs.
CD and master tape vs. LP. which is easier to tell from the master
tape, the CD or the LP? if you don't like the sound of the master
tape, fire the recording engineers and producers.
>You might take a moment to ponder, though, that we are fifteen years
>into the introduction of the CD and yet the LP is alive and
>flourishing
a relatively tiny number of expensive specialty releases does not
constitute "flourishing." nor does the bin of 12-inch singles
intended for DJ "scratching." nor do the small number of vinyl
releases by various (mostly grunge) bands - though i have a few of
these because i like the bigger artwork.
>Can you take YOUR blinders off for a second and contemplate the
>implications of this fact?
i haven't seen a straight fact yet from which to draw an implication.
// Scott Drysdale // Visual Networks Inc // Software Engineer
// sdry...@visual.mctec.com // 60 HD FLH & 57..69 HD FLH // AMIGA!
Bob Myers
Anthony Clarke (ant...@om.com.au) wrote:
>> Yes, and that's the whole point - the sound that the LP-philes prefer
>> does NOT seem to be due to any increased OBJECTIVE accuracy on the
>> part of that medium, but rather due to the specific way in which the
> ^^^
>> medium introduces INaccuracies.
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> Um .... Are you saying that CD does NOT introduce INaccuracies or
> dare I say even fewer than the analogue medium, if so why does the
> Digital world have devices such as error correction, filtering,
> quantitisation, and the latest craze... anti-jitter !!
The CD format itself has no INHERENT inaccuracies or limitations
beyond (a) the bandwidth limit imposed by the sampling rate, and (b)
the dynamic range limit imposed by the number of bits per sample.
Unlike the LP, the CD disc itself - the physical medium which carries
the information - does not NECESSARILY interact with that information.
Filtering and quantization are NOT necessarily sources of additional
inaccuracy beyond the two limits already cited for the format
standard. Error correction is a part of the standard, to allow for
the possibility that the read of the information from the disc may not
be 100% error free, but the correction circuits themselves do NOT
introduce additional inaccuracy unless the disc or the transport are
damaged (or improperly designed) such that the correction can no
longer restore the original data. And finally, jitter is NOT an
inherent inaccuracy of the medium itself. I've never said that there
aren't various things, in the physical implementation of those things
that play back CDs, which need improvement - but we were talking about
those limitations which are fundamental in the two media.
> > distortions is accepted. And we have yet to see any evidence which
> > would suggest that this sort of distortion is NOT occuring.
> Or "IS" occuring. ??? perhaps :-)
No, the distortions of the LP medium are pretty well understood, and
we also know how many of these translate into more pleasant - even if
less objectively accurate - sound. Check the archives for details.
--
Jeff Swauger
> with the sound and trade them in for new ones. I don't see any
> changes to this trend.
I have to disagree with the above. CD has been available for 15 years
now, but it was a significant time after introduction before a large
amount of previously recorded material was available on CD. So, unless
you were going to buy only new music it was hard to get your tunes on
CD. Also, the fact that reissuing old recordings on CD is as
profitable as it is speaks volumes with respect to how many people are
replacing their old vinyl. I think the presumption that people get
bored with a CD because of the way it sounds is incorrect. What I
believe is happening is that, with the CD, for the first time you have
a medium that can be played without degradation by almost anyone, and
that if you buy a used one you have a reasonable chance of getting a
copy in as good a condition as a new one. I believe this has allowed
people to swap out music they don't like and focus in on things they
do, as well as explore a wider variety of music than they might
otherwise have been able to by trading out and buying used. I know
when records were the only option, I had tons of them that I never
listened to, whereas with CDs ones I am not wild about can be traded
for credit towards other music.
> You might take a moment to ponder, though, that we are fifteen years
> into the introduction of the CD and yet the LP is alive and
> flourishing, with new adherents every day (just today I discovered
> Stretch Records releasing Chick Corea's latest on "High Density
> Vinyl"). By stark contrast, fifteen years after the introduction of
> the LP (ca 1963), you were hard-pressed to find any serious listeners
> sticking to 78s. Can you take YOUR blinders off for a second and
> contemplate the implications of this fact?
I think the statement that LP is "flourishing" is a bit of an
overstatement. It still accounts for an extremely small portion of the
market for prerecorded music. Granted it's more than I would have
expected, but this is mainly due to the fact that there are a small,
but loyal, group of audiophiles that are interested in keeping it
alive, and a cottage industry has sprung up to supply them. In the
overall picture of the "average" music lover it is a non-issue. I
still have what I consider a good turntable, but I've not bought vinyl
in years, and use it only for playing back very obscure things that
are not out on CD. All the music I really enjoy that's available has
been replaced with CD versions.
Kevin Connery
In article <5d7rhs$g...@eyrie.graphics.cornell.edu>,
Richard D Pierce <DPi...@world.std.com> wrote:
>In article <5d542d$n...@canyon.sr.hp.com>, Kevin Connery
><Ker...@cris.com> wrote:
[cut -- rgd]
>Again, NO MEASUREMENTS available today show perfect results.
Fine. Let me restate the premise with even more qualifications.
The measurements we use(1) today are incomplete. No matter how well(2)
any piece of equipment measures(3), we cannot state conclusively from
those measurements *alone* if the equipment will permit(4) the music
reproduced to sound "good", or "accurate", or "convincing" enough.
ALL the measurements show *differences*. But nothing I've seen has
isolated a set of measurements which, BY THEMSELVES, is sufficient to
give a thumbs-up as for their suitability for effective playback.
I repeat my earlier statement that we do have measurements which can
DISqualify equipment, but, while that is a good starting point, it is
clearly insufficient.
If you disagree, please include the list of quantifiable measurements
you feel to be sufficient to so identify either hardware or software.
--kdc
(1) As available from the reviewing magazines, product manufacturers
and distributors -- or even 'widely known' to professionals in the
field (though this last is difficult to determine, unless such
information is at least made known, if not distributed, to the
inquisitive consumer)
(2) but clearly 'not perfect'
(3) Using the measurements available via #1
(4) or create, if you prefer
Jim Andrews
sdura...@aol.com wrote:
[snip]
> I think it's really a question of desire. I have been acquiring used
> LPs at an average price of about $4 per record for over twenty years.
> Not a week goes buy when I don't visit a record store or garage sale
> and score $20-30 worth of records. Am I a snob because I spend $100
> month on records and you choose not to?
No, you're a snob because you seem to think that there is some
correlation between the amount of money one spends on their record
collection and one's status as a "music lover".
> Well CDs have been available for going on 15 years now. Does anyone
> have a 10,000 CD collection? 5,000? No, people buy them, get bored
> with the sound and trade them in for new ones. I don't see any
> changes to this trend.
I have no idea. For someone to acquire 10,000 CDs in 15 years, they'd
have to buy 667 CDs per year. At an average purchase price of, oh,
$13.00 per CD, that comes to nearly $9,000 PER YEAR spent on CDs. Do
you have that kind of cash? I don't. That's my original point, that
CDs haven't been out long enough for many people to acquire prodigous
collections without either being independently wealthy or taking out a
second mortgage. On the other hand, acquiring a similarly-sized
collection of LPs over a period of 30 years (333 LPs per year), at an
average price of $4.00 per LP (your number), one spends a little less
than $1,400 PER YEAR. Do you understand the difference?
And for the record, I know very few people that actually trade in
their CDs. The exchange rate is terrible, and as far as I can tell,
most of those who trade are younger buyers with very little
disposeable income. I seriously doubt they are "bored" with the
sound. I suppose most of them just want to keep current.
> The "blinders" of technology work both ways. I could equally assert
> that your expectation of good sound is colored by advance knowledge
> that your playback gear is technically accurate. This argument
> doesn't get anywhere.
Oh, for pity's sake, Sigfried. The only reason this argument doesn't
get anywhere is because you won't let it get anywhere. Look, I argued
against CDs for years, not because Harry Pearson told me to, but
because my roommate's first-generation Sony player sounded absolutely
horrible compared to my Oracle Alexandria, using "identical"
recordings. In retrospect, I don't know how much of that horrible
sound was due to the player or to poorly-mastered CDs, but when I
started to hear WITH MY OWN EARS that CDs were "catching up" to LPs, I
switched over. That doesn't mean that I've thrown away my LPs, only
that now I usually buy CDs. It's unfortunate that they cost 2.5-3
times what I used to pay for an LP, but there's not much I can do
about that.
> You might take a moment to ponder, though, that we are fifteen years
> into the introduction of the CD and yet the LP is alive and
> flourishing, with new adherents every day (just today I discovered
> Stretch Records releasing Chick Corea's latest on "High Density
> Vinyl"). By stark contrast, fifteen years after the introduction of
> the LP (ca 1963), you were hard-pressed to find any serious listeners
> sticking to 78s. Can you take YOUR blinders off for a second and
> contemplate the implications of this fact?
<blinders off>
Perhaps it just means that LPs sound better than 78s? And before you
go off half-cocked again, there is a big difference between the two
events. In one case, one format (vinyl) was REPLACED by a different
format (CD) that sounded substantially different in most respects. In
the other, a particular format (vinyl) was simply improved, which
apparently resulted in widespread agreement about LPs sonic
superiority to 78s. Do you see the difference?
I'm not getting my hopes up . . .
jim andrews
Steven R. Rochlin
Nousane,
Nousaine wrote:
> Of course the LP is a less accurate reproducer of the master.
Though which is closer to the sound of the live event itself?
> The
> reason listeners "prefer" the LP to the real sound is because they
> hold a psychoacoustic bias toward what they "already know."
After over 10 years of strictly CD, during the later stages a
top-of-the-line Theta rig including the single-mode laser digital data
transfer and balanced circuitry, i now admit my enjoyment of the music
is greater on vinyl then CD. Why? To this musicians ears the MUSIC
sounds more REALISTIC on vinyl then on CD. By the way, MANY years of
my life have been spent on psychotic-acoustic experiments. Sorry and
all, during my studies that wasn't the answer to why folks enjoy music
on vinyl vs CD. May i humbly suggest you listen to live acoustic
music in a good hall. Please pay close attention to the way, say, the
violins sound. Now listen to a CD vs. vinyl. Same goes for a nice
grand piano.
QUESTION: What turntable and CD player setup do you use as a
reference?
> Again...if even your special listeners "hear" the digital
> format as delivering the most accurate rendition of the master WHY are
> you asking the "preference" question? If it is more accurate; it is
> more accurate. Perhaps your listeners need more training!
Again, which is closer to the REALISYIC music reproduction of the
live event, vinyl or CD? Perhaps more folks should be musicians and
better still, attend more live acoustic concerts. Since your in the
car audio field i believe, when will IASCA demand their SQ judges to
attend live acoustic music? In the end what REALLY matters to me is
that YOU...
Enjoy the music,
Steven
Music lover, Head toilet bowl scrubber, workaholic, drivin' maniac,
Baron of Grey Matter, and just one kooky dude who's havin' a
goooood time as he travels through this time/space continuum.
Curtis Leeds
> > Arnold B. Krueger wrote:
> >> One friend of mine who routinely listens to LP's on a system
> >> designed to exploit CD's has simply added a filter to eliminate
> >> excess low frequency noise from the LP side. In the old days we
> >> called these "rumble filters".
and I answered...
> > They're still known as rumble filters. With today's best turntables
> > and arms, they are just not necessary.
...and Krueger rresponds:
> Tell that to your now EX-subwoofer the next time you misscue an LP, or
> the tail out groove ends with a clunk. Actually if you are playing an
> LP might as well turn off a "true" subwoofer anyway, right?
No, I get a lot of LF off LP on my system. For the, uh, "record", I
don't actually use a subwoofer; instead, I have a full-range speaker
system. (Infinity IRS Beta) As for "misscuing" an LP: my pick-up arm
is damped both directions, so it's not a problem. As for the
"tail-out" grooves - I've never had any problems there, either.
jj, curmudgeon and all-around grouch
In article <5d542d$n...@canyon.sr.hp.com> Ker...@cris.com (Kevin
Connery) writes:
> I think that is due to the current situation where something which
> measures as 'perfect' as we can measure doesn't necessarily sound
> good--indicating that we're missing at least one critical measurement.
Exactly what is it that you're talking about here. I've seen
absolutely NO such problem, when using any sensible measurment method.
If you're talking about antiquated measures like MSE/SNR/THD, as
measured on the bench, as opposed to in-situ, they've been debunked so
many times here that I'm loath to repeat myself.
If you're talking about the new perceptual measures, they predict very
well what sounds different and what doesn't. They were cited in the
midst of a controversy a few months ago when someone dear to all our
hearts claimed that they didn't exist.
Go read.
--
Copyright alice!jj 1997, all rights reserved, except transmission by USENET
and like facilities granted. This notice must be included. Any use by a
provider charging in any way for the IP represented in and by this article
and any inclusion in print or other media are specifically prohibited.
John Busenitz
On 5 Feb 1997, Kevin Connery wrote:
> The measurements we use(1) today are incomplete. No matter how well(2)
> any piece of equipment measures(3), we cannot state conclusively from
> those measurements *alone* if the equipment will permit(4) the music
> reproduced to sound "good", or "accurate", or "convincing" enough.
That is because the above adjectives are dependant upon the person
listening and their bias. They are not objective quantifiable
parameters that exist in equipment. They are in someones mind. Thus
it is useless discussing them, at least in this case.
> ALL the measurements show *differences*. But nothing I've seen has
> isolated a set of measurements which, BY THEMSELVES, is sufficient to
> give a thumbs-up as for their suitability for effective playback.
I don't think I've seen a complete set of measurements either, much
less know how to properly interpret them. But that doesn't mean they
don't exist, or can't reliably predict how something will objectively
sound.
> If you disagree, please include the list of quantifiable measurements
> you feel to be sufficient to so identify either hardware or software.
I list of references has been posted numerous times by jj and others.
I don't think anyone on the list has the time or patience to type them
all specifically in. I know I don't. If you really are interested,
look through the JAES.
John Busenitz
On 5 Feb 1997, John Ongtooguk wrote:
> Steven Abrams (abr...@cs.columbia.edu) wrote:
>
> : You're missing the primary conclusion of Mr. Atkinson's post -- namely
> : that all people, regardless of their background -- prefered the vastly
> : more accurate 20-bit rendition, "even" over the LP. So all arguments
>
> [cutting -- rgd]
>
> : So much for "undocumented subtractive effects in digital playback."
>
> But most of us will have to be content to listen to 16 bit digital
> as played back thru a CD player instead of 20 bit off of a tape,
> and since the LP was preferred to the 16 bit CD the primary
> conclusion is the question why ? What's missing on the 16 bit CD
Listener bias.
> and/or what's been added to the LP ? Also, why did the 16 bit CD
Distortion, noise, wow & flutter.
> sound different than the 16 bit feed ? Again what's being lost on
> the CD ?
It hasn't been proven that they did sound different.
John Busenitz
On 6 Feb 1997, Steven R. Rochlin wrote:
> Nousaine wrote:
> > Of course the LP is a less accurate reproducer of the master.
>
> Though which is closer to the sound of the live event itself?
Assuming the recording is an accurate one, the CD would be closer to
the sound of the live event.
> > The
> > reason listeners "prefer" the LP to the real sound is because they
> > hold a psychoacoustic bias toward what they "already know."
>
> After over 10 years of strictly CD, during the later stages a
> top-of-the-line Theta rig including the single-mode laser digital data
> transfer and balanced circuitry, i now admit my enjoyment of the music
> is greater on vinyl then CD. Why? To this musicians ears the MUSIC
> sounds more REALISTIC on vinyl then on CD. By the way, MANY years of
But is this because it really IS more realistic, or is this the result
of something psychological? When analysis says the CD is more
realistic, and opinions are split between CD and vinyl, it is pretty
obvious which is the more realistic medium.
> my life have been spent on psychotic-acoustic experiments. Sorry and
> all, during my studies that wasn't the answer to why folks enjoy music
> on vinyl vs CD. May i humbly suggest you listen to live acoustic
> music in a good hall. Please pay close attention to the way, say, the
> violins sound. Now listen to a CD vs. vinyl. Same goes for a nice
> grand piano.
Done that, as I assume most have. You aren't going to change any
opinions with this. Most people (especially the digiphobes) have
already made up their minds as to which is the more realistic medium.
To me, noise, ragged frequency response, distortion and tics aren't
realistic. And the fact remains that the CD is more accurate.
> QUESTION: What turntable and CD player setup do you use as a
> reference?
Ah, the Curtis Leeds approach. If someone opines that vinyl sounds
inferior to CD, use the "inferiority" of their LP equipment as the
reason they prefer CD.
> > Again...if even your special listeners "hear" the digital
> > format as delivering the most accurate rendition of the master WHY are
> > you asking the "preference" question? If it is more accurate; it is
> > more accurate. Perhaps your listeners need more training!
>
> Again, which is closer to the REALISYIC music reproduction of the
> live event, vinyl or CD? Perhaps more folks should be musicians and
CD, obviously. A few vocal opinions don't change that.
> better still, attend more live acoustic concerts. Since your in the
> car audio field i believe, when will IASCA demand their SQ judges to
> attend live acoustic music?
The same time that audio reproduction becomes more important to them
than bikini chicks, neon lights, RTA readings, and SPL contests.
Nousaine
Kurt Strain refers to data that doesn't exist:
"This of course discounts the thousands that have switched from
their old preference of CD's to LP's. Why let some evidence stand
in the way of absolutism?"
What evidence? What surveys or studies do you have that supports this
assertion? Actually even a few thousand "converts" does not a trend
make. There are still people who move "back to the woods" eschewing
modern civilization every year...and other guys who drive antique cars
around...so what?
Bob Myers
Steven R. Rochlin (enjoyt...@top.monad.net) wrote:
> Nousaine wrote:
>> Of course the LP is a less accurate reproducer of the master.
> Though which is closer to the sound of the live event itself?
Well, this brings up some interesting questions. Do you mean -
Closer to the ACTUAL sound of the live event? (And in the case of
studio recording, just what does this mean?)
or do you mean
Closer to what you EXPECTED was the sound of the real event?
--
Bob Myers | "Man will occasionally stumble over the truth, but
my...@fc.hp.com | most of the time he will pick himself up and continue."
O- | - Winston Churchill
Steven Abrams
"Steven R. Rochlin" <enjoyt...@top.monad.net> writes:
> May i humbly suggest you listen to live acoustic music in a good
> hall. Please pay close attention to the way, say, the violins
> sound. Now listen to a CD vs. vinyl. Same goes for a nice grand
> piano.
May I humbly suggest that many people have done this and come to the
exact opposite conclusion. Myself included.
While we're exchanging personal anecdotes, I thought I'd mention that
I did, at one time prefer LPs/tubes. During the time that I lived on
the corner of 57th St. and Sixth Ave in NYC -- literally down the
block from Carnegie Hall, I went to more live concerts there than you
can shake a stick at. I began to realize more and more that the
things which warm the LP sound and sometimes make it seem more natural
are really colorations that aren't there in real life. This also
coincides with the time that some *really* good CDs were coming out.
Sure, I've heard "bad digital" but I'e also heard outstanding stuff
from a digital rig.
Again, to avoid a tit for tat thing, I should add that I've been a
musician nearly all my life -- I began taking piano lessons when I was
4, I think. So I, too, am pretty familiar with the sound of a good
grand piano. The one in my living room sopunds pretty nice, in fact.
I am not, however, trying to convince you that my way is the right
way. What I am trying to convince you of is that it is simply a
matter of preference. You can like it one way, I can like it another
way, and neither of us are "righter" than the other. We have,
apparently, similarities in our backgrounds, i.e. being musicians who
have heard lots of live music, and yet we have different preferences.
Big deal.
> QUESTION: What turntable and CD player setup do you use as a
> reference?
Have heard many setups. My current setup is an AR ES1/AudioQuest
pT5/Sumiko BP, and a Yamaha changer feeding a PS Audio DAC. Neither
of these are the best examples of either technology that I've heard.
Curtis Leeds
John Busenitz wrote:
> Ah, the Curtis Leeds approach. If someone opines that vinyl sounds
> inferior to CD, use the "inferiority" of their LP equipment as the
> reason they prefer CD.
Really John, this is a cheap shot. If a listener claims to have
chronic problems with LP playback - problems which many other
listeners have long since been able to banish - it is only reasonable
to inquire about the equipment (turntable/arm/cartridge/preamp) he is
using and how it is set-up. It certainly is a more reasonable response
than me saying "gee, John Doe is plagued with no LF response, I guess
I better through out my turntable."
Unfortunately, many problems connected with LP playback are the result
of poor equipment or - even more commonly - poor set-up. This has
nothing to do with "the Curtis Leeds approach". Still, I suppose it's
nice to have something named after you. ;)
--
********************************************************
Curtis Leeds cle...@idt.net
"A man hears what he wants to hear, and disregards
the rest."
********************************************************