On Wed, 2 Jul 2003 20:54:43 +0100, Chesney Christ
wrote:

A certain Kurt Hamster, of uk.rec.audio "fame", writes :

Tell us more about this "harmonic distortion". Which harmonics do you
find get distorted ? Have you measured this ? Have you been able to
detect it through the dithering anti-aliasing filter built into all
modern CD players ?

How about the harmonics, from say 25k, being totally missing?

CD can't reproduce more than 22.5kHz, that's a physical limitation (the
Nyquist frequency, where you get just two samples per cycle). But well
below that you get 'reflections' in the frequency domain. To take an
example, imagine sampling a pure sin wave at the Nyquist frequwncy (fn)
less 1Hz. To start with, say that the samples come at the crossover
points, so they are zero. After 0.25s, the points will have shifted
so that you are now sampling at the maximum and minimum and you will
have a maximum amplitude square wave, then after another 0.25s back to
zero, etc. The result is a 'beat' at 1Hz. As the difference from the
Nyquist frequency increases (the signal frequency drops) the 'beat'
frequency increases.

In practice, the effect can be heard down to around a third of the
sampling frequency. For CD sampling rates this is just under 15kHz, and
most normal people can hear up there easily (and modern speaker systems
can reproduce it).

(I don't know about 'modern' CD players, my CDs are played via my PC,
usually into headphones. If the player has to 'dither' then it is
losing information and is dependent on the type of recorded material and
the dithering algorithm how successful it will be at masking the
imperfections.)

That is not what "harmonic distortion" means. You're talking about a
steep frequency rolloff. Can you get 25Khz from an LP ? How do you
manage to hear it ?

Why do modern amplifiers and speakers bother with frequencies above
20kHz, if no one can hear them? Because the human ear does detect
"something wrong" even if it can no longer register the actual
frequencies.

Does this alleged distortion hurt when you listen to a vinyl LP
originally recorded on a digital master tape (most vinyl manufactured
from the mid-80s onwards..) ?

Most? Maybe, maybe not. In the late 80s a lot of studios still preferred
analogue tape.

TBH, I have no easy way of measuring exactly how many of them were
produced digitally. Just doing a quick cross-reference over my own small
vinyl collection.

Many were produced digitally (but at a higher than CD sample rate) but
then went through analogue tape and filtering before getting to vinyl.
And many were ADA so the original recorded signal didn't have anything
significant above 15kHz anyway. The problem comes with modern "straight
to digital" recordings where the signal is "faithfully reproduced" --
right up to the down-sampling to CD quality.

(It is most noticable with organ solos at the top end of the organ's
compass, where even the 3rd harmonic (first overtone of an open pipe) is
above 15k and the higher harmonics have significant presence. Most
people can't hear the harmonics directly, but the 'reflected'
frequencies fall back in the same band as the audible ones and make
horrible discords. It can be 'solved' by filtering before the
down-sampling, but then the frequency response is not noticably better
than that of audio cassette and vinyl...)

Chris C

On 3 Jul 2003 06:25:41 GMT, (Chris Croughton)
wrote:

On Wed, 2 Jul 2003 20:54:43 +0100, Chesney Christ
wrote:

A certain Kurt Hamster, of uk.rec.audio "fame", writes :

Tell us more about this "harmonic distortion". Which harmonics do you
find get distorted ? Have you measured this ? Have you been able to
detect it through the dithering anti-aliasing filter built into all
modern CD players ?

How about the harmonics, from say 25k, being totally missing?

CD can't reproduce more than 22.5kHz, that's a physical limitation (the
Nyquist frequency, where you get just two samples per cycle). But well
below that you get 'reflections' in the frequency domain. To take an
example, imagine sampling a pure sin wave at the Nyquist frequwncy (fn)
less 1Hz. To start with, say that the samples come at the crossover
points, so they are zero. After 0.25s, the points will have shifted
so that you are now sampling at the maximum and minimum and you will
have a maximum amplitude square wave, then after another 0.25s back to
zero, etc. The result is a 'beat' at 1Hz. As the difference from the
Nyquist frequency increases (the signal frequency drops) the 'beat'
frequency increases.

In practice, the effect can be heard down to around a third of the
sampling frequency. For CD sampling rates this is just under 15kHz, and
most normal people can hear up there easily (and modern speaker systems
can reproduce it).

(I don't know about 'modern' CD players, my CDs are played via my PC,
usually into headphones. If the player has to 'dither' then it is
losing information and is dependent on the type of recorded material and
the dithering algorithm how successful it will be at masking the
imperfections.)

That is not what "harmonic distortion" means. You're talking about a
steep frequency rolloff. Can you get 25Khz from an LP ? How do you
manage to hear it ?

Why do modern amplifiers and speakers bother with frequencies above
20kHz, if no one can hear them? Because the human ear does detect
"something wrong" even if it can no longer register the actual
frequencies.

Does this alleged distortion hurt when you listen to a vinyl LP
originally recorded on a digital master tape (most vinyl manufactured
from the mid-80s onwards..) ?

Most? Maybe, maybe not. In the late 80s a lot of studios still preferred
analogue tape.

TBH, I have no easy way of measuring exactly how many of them were
produced digitally. Just doing a quick cross-reference over my own small
vinyl collection.

Many were produced digitally (but at a higher than CD sample rate) but
then went through analogue tape and filtering before getting to vinyl.
And many were ADA so the original recorded signal didn't have anything
significant above 15kHz anyway. The problem comes with modern "straight
to digital" recordings where the signal is "faithfully reproduced" --
right up to the down-sampling to CD quality.

(It is most noticable with organ solos at the top end of the organ's
compass, where even the 3rd harmonic (first overtone of an open pipe) is
above 15k and the higher harmonics have significant presence. Most
people can't hear the harmonics directly, but the 'reflected'
frequencies fall back in the same band as the audible ones and make
horrible discords. It can be 'solved' by filtering before the
down-sampling, but then the frequency response is not noticably better
than that of audio cassette and vinyl...)

Chris C

I would like to thank Chris for demonstrating, once and for all, that
there is no such thing as unutterable rubbish.

And yes, I looked at his web site - and yes, I am actually rather
scared. I really didn't know such people were also eligible for the
"care in the community" programme.

d

_____________________________

http://www.pearce.uk.com

In article ,
Chris Croughton wrote:
Why do modern amplifiers and speakers bother with frequencies above
20kHz, if no one can hear them?

In much the same way as to make a car perform ok at the legal top speed it
will be capable of somewhat more. To make a speaker or amp with a sharp
cutoff at the limit of hearing would be fairly pointless, although it's
good practice not to let it go too far beyond.

Because the human ear does detect "something wrong" even if it can no
longer register the actual frequencies.

This hasn't been conclusively proved.

--
*Suicidal twin kills sister by mistake.

Dave Plowman London SW 12
RIP Acorn

In message , Dave Plowman
writes
In article ,
Chris Croughton wrote:
Why do modern amplifiers and speakers bother with frequencies above
20kHz, if no one can hear them?

In much the same way as to make a car perform ok at the legal top speed it
will be capable of somewhat more. To make a speaker or amp with a sharp
cutoff at the limit of hearing would be fairly pointless, although it's
good practice not to let it go too far beyond.

Because the human ear does detect "something wrong" even if it can no
longer register the actual frequencies.

This hasn't been conclusively proved.

Don't harmonics in the hearable range also depend on the interaction of
frequencies, some of which may be in the ranges outside the hearable
range? And harmonics contribute to the "flavour" of the whole?
--
Paul B The irony of the Information Age is that it has given new
respectability to uninformed opinion. - John Lawton

On 3 Jul 2003 06:25:41 GMT, (Chris Croughton)
wrote:

On Wed, 2 Jul 2003 20:54:43 +0100, Chesney Christ
wrote:

A certain Kurt Hamster, of uk.rec.audio "fame", writes :

Tell us more about this "harmonic distortion". Which harmonics do you
find get distorted ? Have you measured this ? Have you been able to
detect it through the dithering anti-aliasing filter built into all
modern CD players ?

How about the harmonics, from say 25k, being totally missing?

CD can't reproduce more than 22.5kHz, that's a physical limitation (the
Nyquist frequency, where you get just two samples per cycle).

Correct.

But well
below that you get 'reflections' in the frequency domain.

Incorrect.

To take an
example, imagine sampling a pure sin wave at the Nyquist frequwncy (fn)
less 1Hz. To start with, say that the samples come at the crossover
points, so they are zero. After 0.25s, the points will have shifted
so that you are now sampling at the maximum and minimum and you will
have a maximum amplitude square wave, then after another 0.25s back to
zero, etc. The result is a 'beat' at 1Hz. As the difference from the
Nyquist frequency increases (the signal frequency drops) the 'beat'
frequency increases.

Nice fun theory, unfortunately not borne out by reality. Real
spoilsport, reality....................

Check it out, you will *not* see a reflection of 22.05kHz in the
output of any CD player. There are plenty of test CDs around with
20kHz 0dB FS test tones on them, which according to you should produce
a 2.05kHz 'reflection', right in the ear's most sensitive region. It
doesn't happen.

In practice, the effect can be heard down to around a third of the
sampling frequency. For CD sampling rates this is just under 15kHz, and
most normal people can hear up there easily (and modern speaker systems
can reproduce it).

Absolute garbage! You are making up this nonsense.

(I don't know about 'modern' CD players, my CDs are played via my PC,
usually into headphones. If the player has to 'dither' then it is
losing information and is dependent on the type of recorded material and
the dithering algorithm how successful it will be at masking the
imperfections.)

Dithering is done at the A/D stage, not in the player.

That is not what "harmonic distortion" means. You're talking about a
steep frequency rolloff. Can you get 25Khz from an LP ? How do you
manage to hear it ?

Why do modern amplifiers and speakers bother with frequencies above
20kHz, if no one can hear them? Because the human ear does detect
"something wrong" even if it can no longer register the actual
frequencies.

This is not proven.

Does this alleged distortion hurt when you listen to a vinyl LP
originally recorded on a digital master tape (most vinyl manufactured
from the mid-80s onwards..) ?

Most? Maybe, maybe not. In the late 80s a lot of studios still preferred
analogue tape.

TBH, I have no easy way of measuring exactly how many of them were
produced digitally. Just doing a quick cross-reference over my own small
vinyl collection.

Many were produced digitally (but at a higher than CD sample rate) but
then went through analogue tape and filtering before getting to vinyl.
And many were ADA so the original recorded signal didn't have anything
significant above 15kHz anyway. The problem comes with modern "straight
to digital" recordings where the signal is "faithfully reproduced" --
right up to the down-sampling to CD quality.

What is the above garbage supposed to mean? What 'problem'?

(It is most noticable with organ solos at the top end of the organ's
compass, where even the 3rd harmonic (first overtone of an open pipe) is
above 15k and the higher harmonics have significant presence.

No, they don't. They are in fact inaudible.

Most
people can't hear the harmonics directly, but the 'reflected'
frequencies fall back in the same band as the audible ones and make
horrible discords. It can be 'solved' by filtering before the
down-sampling, but then the frequency response is not noticably better
than that of audio cassette and vinyl...)

This is just more of the same garbage. Your claimed 'reflections'
simply *do not exist*, as can easily be proven by the simple
experiment I described above.
--

Stewart Pinkerton | Music is Art - Audio is Engineering

A certain Chris Croughton, of uk.rec.audio "fame", writes :
On Wed, 2 Jul 2003 20:54:43 +0100, Chesney Christ
wrote:

A certain Kurt Hamster, of uk.rec.audio "fame", writes :

Tell us more about this "harmonic distortion". Which harmonics do you
find get distorted ? Have you measured this ? Have you been able to
detect it through the dithering anti-aliasing filter built into all
modern CD players ?

How about the harmonics, from say 25k, being totally missing?

CD can't reproduce more than 22.5kHz, that's a physical limitation (the
Nyquist frequency, where you get just two samples per cycle).

I understand what you are trying to say, but you have the terminology
wrong. 22.5khz is not known as "the Nyquist frequency". In fact there is
no such thing as "the Nyquist frequency". Nyquist merely stipulates the
relationship between the sample rate and the highest frequency that will
be accurately recorded.

There are now digital recording systems that can record up to 192khz
(96khz audio). I very much doubt that such recording systems would
placate vinyl enthusiasts such as your good self. I'd say you'll still
perform what you describe as the warm, natural and lively sound of a
good old LP. That is entirely your prerogative and no-one can take that
away from you - but let's not insult each other's intelligence by
talking about sampling frequencies and harmonic distortion - they are
nothing to do with it (aside from the fact that LP has a much lower
frequency limit than 44.1khz digital does, and it introduces far more
distortion).

[And to clear it up : yes, I'm pretty much in the school of thought that
digital recording systems with a resolution beyond 44.1khz are
completely pointless. 192khz recording is merely a marketing ploy. I'd
be willing to bet that if you take that same recording and dither it
down to 44.1, it would be extremely difficult to detect a difference
between that and the original reliably.]

That is not what "harmonic distortion" means. You're talking about a
steep frequency rolloff. Can you get 25Khz from an LP ? How do you
manage to hear it ?

Why do modern amplifiers and speakers bother with frequencies above
20kHz, if no one can hear them?

You could build a low pass filter into an amp to stop it, but why bother
to add expensive extra components in the audio path to remove something
that can't be heard ? The high frequencies don't do any harm, you just
can't detect them.

Because the human ear does detect
"something wrong" even if it can no longer register the actual
frequencies.

For a start, there is no scientific evidence that this is the case. But
let's play fair and assume for a moment that it has been conclusively
proven. There are still two good reasons why I think this is irrelevant
:

(1) Vinyl isn't capable of exceeding 15-16khz at best - the frequencies
are above are rolled off using an LPF during mastering. How often does
this need to be repeated ? The original master tape gets *extensively*
doctored before it can be used to master an actual vinyl disc.

(2) Vinyl fans never seem to complain about digitally mastered LPs,
which would already have all of these supposed frequencies missing, even
before being mastered to the LP format. This leads me to believe that
vinyl fans are simply people who enjoy the warm distortion that vinyl
mastering and playback produces. As I continuously say, this is entirely
the prerogative of the listener. But it is still an unalterable fact
that distortion is the issue here, not complicated (and often
nonsensical) stuff about which frequencies or harmonics can or cannot be
reproduced.

TBH, I have no easy way of measuring exactly how many of them were
produced digitally. Just doing a quick cross-reference over my own small
vinyl collection.

Many were produced digitally (but at a higher than CD sample rate)

I disagree; digital recorders operating at frequencies above 44.1khz
were uncommon during vinyl's final years as a mainstream format (the
late 80s/early 90s). We can settle this if we can find when the first
one was produced.

but
then went through analogue tape and filtering before getting to vinyl.
And many were ADA so the original recorded signal didn't have anything
significant above 15kHz anyway.

ADA suggests analogue recording, digital mastering and an analogue
reproduction medium. According to your theory, the digital mastering
would have removed all the goodness from the original analogue recording
so this would have been detectable to the listener. Why don't vinyl fans
complain about such recordings ?

The problem comes with modern "straight
to digital" recordings where the signal is "faithfully reproduced" --
right up to the down-sampling to CD quality.

That's pretty much all of them. Apart from the extremely high end and
low end cassette-based kit, no analogue multitrack or professional tape
recorders are made these days. It's all digital. It's very strange that
the extremely discerning artists including world-class orchestras all
seem quite happy to record their works using modern multitrack digital
recorders, and don't complain about the artifacts that you're
describing. Why do you reckon that is ?

(It is most noticable with organ solos at the top end of the organ's
compass, where even the 3rd harmonic (first overtone of an open pipe) is
above 15k and the higher harmonics have significant presence.

Except you can't hear them.

It can be 'solved' by filtering before the
down-sampling, but then the frequency response is not noticably better
than that of audio cassette and vinyl...)

Audio cassette ? That now-dead medium which gets dubbed through 3-4
generations (at several times normal speed) is somehow equivalent to a
digital recording ? Please tell me you're joking. Not even the vinyl
enthusiasts try out this madness.

--

"Jokes mentioning ducks were considered particularly funny." - cnn.com

Message-ID: from Chesney Christ contained
the following:

(1) Vinyl isn't capable of exceeding 15-16khz at best - the frequencies
are above are rolled off using an LPF during mastering. How often does
this need to be repeated ? The original master tape gets *extensively*
doctored before it can be used to master an actual vinyl disc.

In the early 80s I was a director of a 24 track recording studio and was
often able to compare quarter inch masters with the vinyl pressing using
the same studio equipment. There is no doubt that the sound was changed
(and in our view, degraded) by the process.

Clearly the vinyl enthusiasts enjoy the medium but the fact remains that
the vinyl disc will always be inferior to the studio master. We noticed
no such differences when we started using digital masters and I imagine
those early devices were no better than CD quality.



--
Geoff Berrow
It's only Usenet, no one dies.
My opinions, not the committee's, mine.
Simple RFDs http://www.ckdog.co.uk/rfdmaker/

In inside of
uk.net.news.config, 'Geoff Berrow' wrote:

Message-ID: from Chesney Christ contained
the following:

(1) Vinyl isn't capable of exceeding 15-16khz at best - the frequencies
are above are rolled off using an LPF during mastering. How often does
this need to be repeated ? The original master tape gets *extensively*
doctored before it can be used to master an actual vinyl disc.

In the early 80s I was a director of a 24 track recording studio and was
often able to compare quarter inch masters with the vinyl pressing using
the same studio equipment. There is no doubt that the sound was changed
(and in our view, degraded) by the process.

Clearly the vinyl enthusiasts enjoy the medium but the fact remains that
the vinyl disc will always be inferior to the studio master. We noticed
no such differences when we started using digital masters and I imagine
those early devices were no better than CD quality.

The experiment I would like to try would be to digitise and burn onto
CD the output of a high quality record deck, then see if vinyl lovers
can spot the difference between the two.

I've no preconeptions about the result, but it would be a fair
experiment that would answer most of these 'superiority' issues once
and for all.

All I know is that my preference leans toward analogue for home hifi,
it may change if the prices of higher end digital stuff keep falling.



--
Dave J. - Back from the grey netherworld..

In article ,
Dave J. wrote:
The experiment I would like to try would be to digitise and burn onto
CD the output of a high quality record deck, then see if vinyl lovers
can spot the difference between the two.

I've done this and they can't. And neither can I.

--
*Elephants are the only mammals that can't jump *

Dave Plowman London SW 12
RIP Acorn

"Dave J." wrote


The experiment I would like to try would be to digitise and burn onto
CD the output of a high quality record deck, then see if vinyl lovers
can spot the difference between the two.


Then try it. See http://www.apah69.dsl.pipex.com/keit...ng/ripping.htm
for some testing I've done.

I'll tell you this much - I prefer a CD rip of the vinyl to the actual CD of
the same piece of music


I've no preconeptions about the result, but it would be a fair
experiment that would answer most of these 'superiority' issues once
and for all.


Oh no it wouldn't.............. ;-)



All I know is that my preference leans toward analogue for home hifi,
it may change if the prices of higher end digital stuff keep falling.


And if prices of 'higher end' analogue keep falling??