CD-player died, need advice
 

On Tue, 18 Dec 2007 21:56:09 -0500, "Arny Krueger"
wrote:

All CD players sound the same.

Nope. The CD player function of the typical computer CD
ROM drive (output via an analog output) is probably so
flawed that you will hear a difference.

You're counting the mini-jack output on the front of a
computer CD drive as a "CD player" in the context of this
thread? Rather a straw man, I'd say :-)

What about the other dozen or so items I listed, or are you going to try to
make out that this was all I said?

Why should I question the points that WEREN'T questionable? Though
you were really only muddying the waters by listing devices not
intended for quality analogue playback, portable toys and damaged
units.

CD-player died, need advice
 

On Wed, 19 Dec 2007 10:31:50 +0000 (GMT), Jim Lesurf
wrote:

I've just announced the details of a new webpage in another thread. However
anyone interested in one way in which players may vary significantly might
find

http://www.audiomisc.co.uk/HFN/OverTheTop/OTT.html

interesting. :-)

He seems to be describing nasty things that could happen if you DON'T
filter the analogue input down to half the sample frequency before
digitising. Or am I missing something?

CD-player died, need advice
 

In article , Laurence Payne
NOSPAMlpayne1ATdsl.pipex.com wrote:
On Wed, 19 Dec 2007 10:31:50 +0000 (GMT), Jim Lesurf
wrote:

I've just announced the details of a new webpage in another thread.
However anyone interested in one way in which players may vary
significantly might find

http://www.audiomisc.co.uk/HFN/OverTheTop/OTT.html

interesting. :-)

He seems to be describing nasty things that could happen if you DON'T
filter the analogue input down to half the sample frequency before
digitising. Or am I missing something?

Not sure who 'he' may be since I wrote the page. :-)

However, yes, you do seem to be missing something. The page specifically
deals with a situation where the input signal *is* being filtered to accord
with the Nyquist requirements. The example used assumes ideal FIR filtering
before sampling (recording) and during reconstruction (replay). The 'click'
used an example, when filtered, gives the sinc-like shape, and this is
a signature that the recording is band limited as required. If this were
not the case, the 'misaligned' case would have all sample values being
zero. :-)

So the samples are then correctly taken for that limited bandwidth. No out
of band components affect the recorded samples.

The point is that when you *do* filter in accord with Nyquist requirements
the *levels* (not the frequency range) of the reconstructed waveforms can
have peaks well above any of the individual samples. People have been
discussing this kind of situation in another thread on digitising LP. The
problem is that there can (and will) be peak waveform levels which occur in
between the sampled instants, and which are higher than the sample levels.

Although if the player is unable to reconstruct the waveform correctly from
the samples, then the result will be distorted, and that may lead to
frequency components that weren't in the original signal that was sampled.

In fact, you can easily show that an input which is clearly within
the Nyquist limit can also give the problem. Consider an 11.025 kHz
sinusoid created by alternating two max +ve and then two max -ve
sample values over and over again. These define an inband single
frequency, but the peaks are well above 0dBFS as the samples aren't
at the peaks of the waveform. From Information Theory, there is
no problem with this, and the samples unambiguously define the
recorded waveform. The snag is that the player may not have been
designed to output waveforms with peaks above 0dBFS as the designer
didn't think to allow for this. That isn't an information theory
problem, but a design problem.


FWIW I've analysed a random selection of rock/pop CDs and it seems quite
common for the reconstructed waveform to show peaks of the order of +1dBFS
to +2dBFS if the player is assumed to use the standard sort of time
symmetric FIR reconstruction filtering. This does not mean the recording
was poorly filtered, but that it has been scaled upwards so that the
maximum sample values then require a waveform with such peak levels.

The JAES paper I reference (and others I did not list) also find similar
results. So this isn't simply an 'academic possibility', but a real
situation.


Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.audiomisc.co.uk/index.html
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html

CD-player died, need advice
 

"Trevor Wilson" wrote in message
...

**I, for one, know full well, that all CD players were/are not created
equal. Far from it. In those days, the Meridians were, indeed, the best
available.


I have acess to a number of fairly standard Sony, Panasonic,
Technics etc machines, a Quad 66, and also a couple of
broadcast Denons, plus the legendary Studer D730 broadcast
machine. The Studer, Denon broadast machine and the
Quad 66 have accurate sync, within ten subframes.

One cannot reliably compare them by moving the same CD
from one machine to another, but only by AB switched comparison.
As Arny points out, tight sync is crucial, so in a multi-machine
comparison each machine must be playing the same CD and be
carefully level matched. You have to be able to start all machines
simultaneously, with sub-frame accuracy, and also be able
to make allowances (start offset) for the differences in
start-up time.

Is this what you did Trevor?

I am interested to know in what respect you thought
the HK outperformed the others. More details please.
The HK which you praise so highly, is a cheapie
cheapo, available at huge discounts everywhere,
so one could be forgiven for wondering if you had bought
a vanload of them at retail minus 70%, and your post
was just a pre-sales pitch:-))

In a six machine test which I attended earlier this year,
units like the ones I have listed above were compared in
DBT. Some very perceptive people took part.
None could distingiush any audible difference machine
to machine.

However, there are *huge* differences in build quality and
facilities offered. I am told that the HK does not have even
a pitch control. Many serious listeners, especially those with
an accurate sense of pitch require this.

I believe that with CD players, as with most things, you get
what youy pay for.

Regards to all
Iain

CD-player died, need advice
 

"Laurence Payne" NOSPAMlpayne1ATdsl.pipex.com wrote in
message
On Tue, 18 Dec 2007 21:56:09 -0500, "Arny Krueger"
wrote:

All CD players sound the same.

Nope. The CD player function of the typical computer CD
ROM drive (output via an analog output) is probably so
flawed that you will hear a difference.

You're counting the mini-jack output on the front of a
computer CD drive as a "CD player" in the context of
this thread? Rather a straw man, I'd say :-)

I suspect that it is not widely known that the audio sections of CD ROM
drives are generally among the worst forms of CD player in general use. This
is because they have some of the worst 44/16 digital-to-analog converters
that are in any piece of modern audio equipment.

I know many people who use them daily to listen to music.

The good news is that since Win98 or so, MS's media player has avoided the
audio circuitry in the CD ROM drive, and captured audio via the IDE port.
However, the captured audio still usually had to negotiate the
on-motherboard sound chip.

In the early days of computer audio, the headphone jack on the CD ROM was
the cleanest audio source in the whole computer! It wasn't until sometime
around the turn of the millenium that the on-board converters in most PCs
were as good.

CD-player died, need advice
 

On Tue, 18 Dec 2007 21:55:14 -0500, Arny Krueger wrote:

In the case of a CD player, you compare the signal at its output
terminals to the signal that was used to burn the CD being played.

That's not strictly true as the human ear hears different types of
distortion differently. Some high level distortion may go unnoticed and
vice versa. Of course, no one could argue with a zero distortion
measurement, but nearly zero may not be good enough for some who are
particularly sensitive.

CD-player died, need advice
 

"Arny Krueger" wrote in message
. ..
"Trevor Wilson" wrote
in message

**That would be a reasonable assumption. HK have clearly
put a lot of effort into building a quality analogue
section in that machine.

An ultimate quality analog section for a CD player - a 5532, a few penny
resistors, and a couple of caps.

**No, but that is an adequate one. More crucially, however, many cheap
players use vastly inferior OP amps. I've seen 4558 class OP used in many
cheap players. Cheap DVD players are guilty of this. Further, correct
implemented muting can make a big difference. Cheap players use a
transistor, whilst better quality ones use relays. For the record: The HK
uses a transistor. Mine will soon have relays fitted.

Trevor Wilson

CD-player died, need advice
 

"Iain Churches" wrote in message
i.fi...

"Trevor Wilson" wrote in message
...

**I, for one, know full well, that all CD players were/are not created
equal. Far from it. In those days, the Meridians were, indeed, the best
available.


I have acess to a number of fairly standard Sony, Panasonic,
Technics etc machines, a Quad 66, and also a couple of
broadcast Denons, plus the legendary Studer D730 broadcast
machine. The Studer, Denon broadast machine and the
Quad 66 have accurate sync, within ten subframes.

One cannot reliably compare them by moving the same CD
from one machine to another, but only by AB switched comparison.

**Wrong.

As Arny points out, tight sync is crucial, so in a multi-machine
comparison each machine must be playing the same CD and be
carefully level matched. You have to be able to start all machines
simultaneously, with sub-frame accuracy, and also be able
to make allowances (start offset) for the differences in
start-up time.

Is this what you did Trevor?

**In a few cases, yes. In most, no.


I am interested to know in what respect you thought
the HK outperformed the others. More details please.
The HK which you praise so highly, is a cheapie
cheapo, available at huge discounts everywhere,
so one could be forgiven for wondering if you had bought
a vanload of them at retail minus 70%, and your post
was just a pre-sales pitch:-))

**Nope. I have not made a single cent from selling any HK machines. I simply
sugested to people that they were excellent machines at a very good price.


In a six machine test which I attended earlier this year,
units like the ones I have listed above were compared in
DBT. Some very perceptive people took part.
None could distingiush any audible difference machine
to machine.

**Please list the machines tested, the amplifiers and speakers used. Please
paint a word picture of the room used.


However, there are *huge* differences in build quality and
facilities offered.

**Build quality is largely irrelevant. I've listened to Studer CD players.
Their build quality is quite good. Their sound quality is no better than a
$200.00 machine.

I am told that the HK does not have even
a pitch control. Many serious listeners, especially those with
an accurate sense of pitch require this.

**Are you suggesting that the recordings are faulty? That is the only way a
CD player can play at the wrong speed. Pitch controls are of no use for
domestic listening. I've seen DJs use them, however.


I believe that with CD players, as with most things, you get
what youy pay for.

**Believe what you will. However, I have a better idea: Borrow an HK and
listen to it. Provided your ancillary equipment is up to snuff, you will
hear how good it is. I know your speakers are certainly adequate. I doubt
you have good enough amplification however. You seem to have a preference
for highly distorted (both linear AND non-linear distortion) amplifiers.
Through such rubbish, you will NEVER hear a difference in CD players.

Trevor Wilson

CD-player died, need advice
 

"Eeyore" wrote in message
...


Trevor Wilson wrote:

"Eeyore" wrote
Trevor Wilson wrote:
"Arny Krueger" wrote

Doing a decent job of comparing optical disc players is a often lot
of
work - the big problem is getting and keeping the discs in synch.
I've
done it, and the results were amazingly small audible differences,
some
of
which were found to be imaginary when the statistical analysis was
complete.

**Agreed. And some differences were audible, when used with suitably
high
quality speakers. In some cases, the differences are not subtle,
though
these instances are rare. In the case of the HK, I've compared with
some
obscenely high priced players and some very cheap players. In most
instances, the HK has won hands-down.

WON by what criteria though ?

**Measurements and sound quality. Is there any other criteria of
relevance?

Sound quality as adjudged by whom and/or what measure ?

**By myself and everyone who has listened to it. BTW: I had my first claim
that the HK is no better than a modestly priced Denon yesterday. A client
borrowed my machine and returned, saying that it did not sound any better
than his machine.

Trevor Wilson

CD-player died, need advice
 

Trevor Wilson wrote:
"Iain Churches" wrote in message
i.fi...

I am told that the HK does not have even
a pitch control. Many serious listeners, especially those with
an accurate sense of pitch require this.

**Are you suggesting that the recordings are faulty? That is the only way a
CD player can play at the wrong speed. Pitch controls are of no use for
domestic listening. I've seen DJs use them, however.

The speed of a CD player is only as accurate as its quartz oscillator.
I wouldn't be surprised if a player could gain or lose a tenth of a
second during an hour, and vary with temperature. Not that it matters
for normal listening but it makes comparisons difficult.

--
Eiron.