[David Spear]

"codifus" wrote in message
...
On May 22, 7:12 pm, (Greg Grainger) wrote:
I'm sure this has been asked before, but I saw an industrial unit today
that I hadn't seen before - something by Panasonic that played MP-3s,
DVDs, and everything in between, including (and this was the part that I
found interesting) DIV-X files.

What is the consensus of the group as to what the best-*sounding* unit of
this type might be? (I'm thinking in terms of CDs and DVDs particularly
- the whole DIV-X thing doesn't interest me.)

Let's say two categories - $1000, and 'price no object.' (List price,
before taxes.)

Any input would be appreciated.

Many thanks,
Greg.
--
Greg Grainger grainger(at)vex.net

'What a world of gammon and spinach it is, though, ain't it?'
- Miss Mowcher
If $1000 is your budget, why not buy an inexpensive Panasonic, like
$100 or so, then feed its digital output to a Becnhmark DAC1? That
should bring you infinitely closer to audio nirvana.


I've never understood the whole concept of the "high-end CD
transport". The purpose of such a device is to read a stream of 1's
and 0's off of a spinning disc and output the (relatively
low-frequency) digital bitstream. Period. The bitstream is EXACTLY THE
SAME, BIT FOR BIT, coming out of a

Walmart Electro-Sonic $39 special (assuming it has a digital output)
as out of a Krell. Except for the highest-end (read most expensive)
transports, they all use the same S/PDIF encoder chips which are made
but very few manufacturers.

The DAC is all that matters. It alone contains the magic smoke which
will reduce jitter, provide up-sampling and oversampling, and amplify
the resulting signal just so.

I'm using a DVD player that came free with the last computer I bought
run into an external audiophile DAC via a $6 TOSLINK cable. It sounds
fantastic, although the player itself does generate some mechanical
noise I could live without.

I suppose there are issues of build quality, finish, reliability,
serviceability, but when you can replace a digital source such as a
DVD player for $50 or $100 for a brand-name, who cares? Buy a spare
and STILL save yourself a few hundred or thousand for something else
which DOES affect the sound quality.

I suppose one could argue that my cheap digital source doesn't glow in
the dark, but maybe one day I'll cut a hole in the top and stick a
tube in it.

Dave

[Greg Grainger]

In article ,
David Spear wrote:
"codifus" wrote in message
...
On May 22, 7:12 pm, (Greg Grainger) wrote:

What is the consensus of the group as to what the best-*sounding* unit of
this type might be? (I'm thinking in terms of CDs and DVDs particularly

I've never understood the whole concept of the "high-end CD
transport". The purpose of such a device is to read a stream of 1's
and 0's off of a spinning disc and output the (relatively
low-frequency) digital bitstream. Period. The bitstream is EXACTLY THE
SAME, BIT FOR BIT, coming out of a

Walmart Electro-Sonic $39 special (assuming it has a digital output)
as out of a Krell. Except for the highest-end (read most expensive)
transports, they all use the same S/PDIF encoder chips which are made
but very few manufacturers.

I was thinking in terms of a unit like my old Revox, rather than separate
transport-and-DAC. Several people have pointed out that the latter is the
way to go.

The Revox works fine, thank you very much, but it doesn't play DVDs or
MP-3s, much less DIV-X, so I have ended up with several 'specialist'
units scattered around the living room. I'm looking for a kind of
all-in-one unit that produces audiophile-quality sound and also does a
good job with DVDs.

Still researching,
Greg.
--
Greg Grainger grainger(at)vex.net

'What a world of gammon and spinach it is, though, ain't it?'
- Miss Mowcher

[Simonel]

I've never understood the whole concept of the "high-end CD transport". The purpose of such a device is to read a stream of 1's and 0's off of a spinning disc and output the (relatively low-frequency) digital bitstream. Period. The bitstream is EXACTLY THE SAME, BIT FOR BIT, coming out of a Walmart Electro-Sonic $39 special (assuming it has a digital output) as out of a Krell. Except for the highest-end (read most expensive) transports, they all use the same S/PDIF encoder chips which are made but very few manufacturers.

In the May issue of Stereophile John Atkinson measured the Oppo
DV-970HD S/PDIF jacks and found out that they output 96kHz and 192
kHz, BUT, that they dithered 24-bit audio to 16-bit. Not all players
dither 24 bit to 16 bit (I am referring to the signal you get when you
play non-copyright protected DVD-A's, like privately burned ones.) So
the bitstream is not exactly the same coming out of the S/PDIF jacks
of different players.


The DAC is all that matters. It alone contains the magic smoke which
will reduce jitter, provide up-sampling and oversampling, and amplify
the resulting signal just so.

I agree that the quality of the DAC is more important than what comes
out of the S/PDIF jacks, but I have 2 players connected to the same
DAC1 through their S/PDIF jacks. One is an old Panasonic S55 and the
other a Denon 2910. The Denon gives me much better sound from DVD-V's.

Simonel

[Steven Sullivan]

Simonel wrote:
I've never understood the whole concept of the "high-end CD transport". The purpose of such a device is to read a stream of 1's and 0's off of a spinning disc and output the (relatively low-frequency) digital bitstream. Period. The bitstream is EXACTLY THE SAME, BIT FOR BIT, coming out of a Walmart Electro-Sonic $39 special (assuming it has a digital output) as out of a Krell. Except for the highest-end (read most expensive) transports, they all use the same S/PDIF encoder chips which are made but very few manufacturers.

In the May issue of Stereophile John Atkinson measured the Oppo
DV-970HD S/PDIF jacks and found out that they output 96kHz and 192
kHz, BUT, that they dithered 24-bit audio to 16-bit. Not all players
dither 24 bit to 16 bit (I am referring to the signal you get when you
play non-copyright protected DVD-A's, like privately burned ones.) So
the bitstream is not exactly the same coming out of the S/PDIF jacks
of different players.

First, how much 'hi rez' audio can be passed from S/PDIF? Neither DVD-A nor SACD can be so
passed (the Oppo can pass them as PCM via HDMI, but that's not S/PDIF; Atkinson didn't
measure the HDMI port, so I can'st say whether dithering goes on there too).

Second, are 24 bits of audio resolution necessary in a home listening environment?

I would guess any effect on SQ would be down to how well the dithering is done,
not 16 vs 24 per se.

The DAC is all that matters. It alone contains the magic smoke which
will reduce jitter, provide up-sampling and oversampling, and amplify
the resulting signal just so.

I agree that the quality of the DAC is more important than what comes
out of the S/PDIF jacks, but I have 2 players connected to the same
DAC1 through their S/PDIF jacks. One is an old Panasonic S55 and the
other a Denon 2910. The Denon gives me much better sound from DVD-V's.

If two transports are giving authentically different sound (verified in blind
trials), at least one of them must be doing something wrong. If one really sounds
'much better', then the other must be doing something VERY wrong.

___
-S
"As human beings, we understand the world through simile, analogy,
metaphor, narrative and, sometimes, claymation." - B. Mason

[Codifus]

On May 31, 10:50 am, Simonel wrote:
I've never understood the whole concept of the "high-end CD
transport". The purpose of such a device is to read a stream of 1's
and 0's off of a spinning disc and output the (relatively
low-frequency) digital bitstream. Period. The bitstream is EXACTLY THE
SAME, BIT FOR BIT, coming out of a
Walmart Electro-Sonic $39 special (assuming it has a digital output)
as out of a Krell. Except for the highest-end (read most expensive)
transports, they all use the same S/PDIF encoder chips which are made
but very few manufacturers.

In the May issue of Stereophile John Atkinson measured the Oppo
DV-970HD S/PDIF jacks and found out that they output 96kHz and 192
kHz, BUT, that they dithered 24-bit audio to 16-bit. Not all players
dither 24 bit to 16 bit (I am referring to the signal you get when you
play non-copyright protected DVD-A's, like privately burned ones.) So
the bitstream is not exactly the same coming out of the S/PDIF jacks
of different players.



The DAC is all that matters. It alone contains the magic smoke which
will reduce jitter, provide up-sampling and oversampling, and amplify
the resulting signal just so.

I agree that the quality of the DAC is more important than what comes
out of the S/PDIF jacks, but I have 2 players connected to the same
DAC1 through their S/PDIF jacks. One is an old Panasonic S55 and the
other a Denon 2910. The Denon gives me much better sound from DVD-V's.

Simonel

It may be nice that DVD-Vs sound better from the Denon, but given that
the sound format from DVD-V is not a serious audiophile format, why
should it matter so much? The priority of DVD-V sound is to transport
multiple channels of whizz bang movie soundtracks which aren't really
produced for the utmost in sound quality. I would think of it more as
an added plus that the DVD-Vs sound better. It's like trying to make
MP3s into and audiophile format.

How does the same DVD-A sound out of your Panny when compared to the
Denon, both palying through the DAC1 of course?

CD

[Steven Sullivan]

codifus wrote:
On May 31, 10:50 am, Simonel wrote:
I've never understood the whole concept of the "high-end CD
transport". The purpose of such a device is to read a stream of 1's
and 0's off of a spinning disc and output the (relatively
low-frequency) digital bitstream. Period. The bitstream is EXACTLY THE
SAME, BIT FOR BIT, coming out of a
Walmart Electro-Sonic $39 special (assuming it has a digital output)
as out of a Krell. Except for the highest-end (read most expensive)
transports, they all use the same S/PDIF encoder chips which are made
but very few manufacturers.

In the May issue of Stereophile John Atkinson measured the Oppo
DV-970HD S/PDIF jacks and found out that they output 96kHz and 192
kHz, BUT, that they dithered 24-bit audio to 16-bit. Not all players
dither 24 bit to 16 bit (I am referring to the signal you get when you
play non-copyright protected DVD-A's, like privately burned ones.) So
the bitstream is not exactly the same coming out of the S/PDIF jacks
of different players.



The DAC is all that matters. It alone contains the magic smoke which
will reduce jitter, provide up-sampling and oversampling, and amplify
the resulting signal just so.

I agree that the quality of the DAC is more important than what comes
out of the S/PDIF jacks, but I have 2 players connected to the same
DAC1 through their S/PDIF jacks. One is an old Panasonic S55 and the
other a Denon 2910. The Denon gives me much better sound from DVD-V's.

Simonel

It may be nice that DVD-Vs sound better from the Denon, but given that
the sound format from DVD-V is not a serious audiophile format, why
should it matter so much? The priority of DVD-V sound is to transport
multiple channels of whizz bang movie soundtracks which aren't really
produced for the utmost in sound quality. I would think of it more as
an added plus that the DVD-Vs sound better. It's like trying to make
MP3s into and audiophile format.

All that's required for an MP3 to be 'audiophile' is for the MP3 to be
indistinguishable from the source by the 'audiophile' listener. Which is well
within the range of possible.

Opera , symphonies, etc have been released on DVD-V, so DD/DTS are not necessarily
for 'whizz bang' only.

___
-S
"As human beings, we understand the world through simile, analogy,
metaphor, narrative and, sometimes, claymation." - B. Mason

[bob]

On Jun 1, 10:33 am, codifus wrote:

It may be nice that DVD-Vs sound better from the Denon, but given that
the sound format from DVD-V is not a serious audiophile format, why
should it matter so much? The priority of DVD-V sound is to transport
multiple channels of whizz bang movie soundtracks which aren't really
produced for the utmost in sound quality.

The Last Waltz?

Stop Making Sense?

ANY opera or ballet?

C'mon. I have my doubts about the real sonic basis of Simonel's
perceptions, but the desire for high-fidelity soundtracks is quite
legitimate.

bob

[[email protected]]

On Jun 1, 10:29 am, Steven Sullivan wrote:
Simonel wrote:
In the May issue ofStereophile John Atkinson measured the Oppo
DV-970HD S/PDIF jacks and found out that they output 96kHz and
192 kHz, BUT, that they dithered 24-bit audio to 16-bit. ...

The Stereophile review is now reprinted at
http://www.stereophile.com/hirezplayers/507oppo/

First, how much 'hi rez' audio can be passed from S/PDIF?
Neither DVD-A nor SACD can be so passed...

True for SACD, but with DVD-A, it depends on the specific disc
authoring. The DVD-A's I burn of my own hi-rez recordings will
result in a bit-accurate datastream appearing at the output of
DVD players other than the Oppo.

the Oppo can pass them as PCM via HDMI, but that's not
S/PDIF; Atkinson didn't measure the HDMI port, so I can't
say whether dithering goes on there too.

Note that I suspect the Oppo doesn't redither 24-bit data to
16, but merely truncates.

John Atkinson
Editor, Stereophile

[Arny Krueger]

"Simonel" wrote in message


In the May issue of Stereophile John Atkinson measured
the Oppo DV-970HD S/PDIF jacks and found out that they
output 96kHz and 192 kHz, BUT, that they dithered 24-bit
audio to 16-bit.

Actually, John Atkinson said:

"I found that the DV-970's digital output truncated the word length to 16
bits, even with true 24-bit audio (such as my own DVD-As, burned with
Minnetonka Software's DiscWelder Bronze program)."

Truncation is *not* dithering.

Not all players dither 24 bit to 16 bit
(I am referring to the signal you get when you play
non-copyright protected DVD-A's, like privately burned
ones.) So the bitstream is not exactly the same coming
out of the S/PDIF jacks of different players.

Do you know of a review of a DVD-A player that says it outputs the full 24
bits from 24 bit DVD-As?

The DAC is all that matters. It alone contains the magic
smoke which
will reduce jitter, provide up-sampling and
oversampling, and amplify
the resulting signal just so.

In this case, very significant because John Atkinson found *huge* amounts of
jitter in the analog outputs of the Oppo.

I agree that the quality of the DAC is more important
than what comes out of the S/PDIF jacks, but I have 2
players connected to the same DAC1 through their S/PDIF
jacks. One is an old Panasonic S55 and the other a Denon
2910. The Denon gives me much better sound from DVD-V's.

Level-matched, bias-controlled, time-synched listening test?

[Kalman Rubinson]

On 1 Jun 2007 14:29:21 GMT, Steven Sullivan wrote:

First, how much 'hi rez' audio can be passed from S/PDIF? Neither DVD-A nor SACD can be so
passed

24/96 - 2 channel is no problem over S/PDIF. The reason you cannot
send DVD-A/SACD over S/PDIF has to do with copy protection and
with the extra bandwidth demands of multichannel.

Kal

[Steven Sullivan]

wrote:
On Jun 1, 10:29 am, Steven Sullivan wrote:
Simonel wrote:
In the May issue ofStereophile John Atkinson measured the Oppo
DV-970HD S/PDIF jacks and found out that they output 96kHz and
192 kHz, BUT, that they dithered 24-bit audio to 16-bit. ...

The Stereophile review is now reprinted at
http://www.stereophile.com/hirezplayers/507oppo/

First, how much 'hi rez' audio can be passed from S/PDIF?
Neither DVD-A nor SACD can be so passed...

True for SACD, but with DVD-A, it depends on the specific disc
authoring. The DVD-A's I burn of my own hi-rez recordings will
result in a bit-accurate datastream appearing at the output of
DVD players other than the Oppo.

Do-it-yourself DVD-A hobbyists constitute at best a minisucle share of the
Oppo (or any other player) market, I suspect. That leaves commercial
DVD-A, of which only a very few in my experience have ever offered S/PDIF
playback of the DVD-A folders (an those were either downsampled, or mono,
or downmixed).

the Oppo can pass them as PCM via HDMI, but that's not
S/PDIF; Atkinson didn't measure the HDMI port, so I can't
say whether dithering goes on there too.

Note that I suspect the Oppo doesn't redither 24-bit data to
16, but merely truncates.

I'd call that bad design then. Yet, the '24-bit' material I play over the
Oppo seems to sound fine. This could be because the dynamic range of teh
source material or the mastering is already well within 16, much less
24bit, limits, or because the extra 8 bits were just 'padding'
(implying a lower-bit stage somewhere in recording/production), or
because I'm not set up to easily properly compare it to a player that
doesn't truncate.

___
-S
"As human beings, we understand the world through simile, analogy,
metaphor, narrative and, sometimes, claymation." - B. Mason

[Steven Sullivan]

Kalman Rubinson wrote:
On 1 Jun 2007 14:29:21 GMT, Steven Sullivan wrote:

First, how much 'hi rez' audio can be passed from S/PDIF? Neither DVD-A nor SACD can be so
passed

24/96 - 2 channel is no problem over S/PDIF. The reason you cannot
send DVD-A/SACD over S/PDIF has to do with copy protection and
with the extra bandwidth demands of multichannel.

Yes, my question should have been more precisely 'how much *commercially
avaialble* hi-rez audio can be passed from S/PDIF?' I'd say it narrows
down rather quickly to the Classic Records HDADs and the odd two-channel
(resampled/downmixed) or mono output from DVD-As.

___
-S
"As human beings, we understand the world through simile, analogy,
metaphor, narrative and, sometimes, claymation." - B. Mason

[campdog]

hi all,

i use a toshiba carousel dvd/cd/sacd player that was recommended as a
fairly low-price alternative to high-end players. i've compared it to
a couple $1000 players and have heard no difference in sound quality.
i payed all of $250.00 for this unit and am quite satisfied. i would
recommend it. i'm currently listening to it hooked to the trend 10.1
amp and the klipschorns seem to love it...i am too.

pete

[Simonel]

On Jun 2, 10:53 am, "Arny Krueger" wrote:
Do you know of a review of a DVD-A player that says it outputs the full 24
bits from 24 bit DVD-As?

Good point, exactly what bothered me. I addressed this question to
John Atkinson, who gave me a list of players reviewed on Stereophile
that he measured to output 24 bits from 24 bit DVD-As, although for
some reason he never published the results. However, they were all
very expensive and I think do not justify the price as transports. I
am still awaiting a review of a consumer grade player that outputs 24
bits from 24 bit DVD-As.

Simonel

(BTW - although DVD-V's sound better on my Denon, CD's and DVD-A's
sound the same from the old Panasonic. Go figure.)

[[email protected]]

On Jun 2, 8:29 pm, Steven Sullivan wrote:
wrote:
On Jun 1, 10:29 am, Steven Sullivan wrote:
First, how much 'hi rez' audio can be passed from S/PDIF?
Neither DVD-A nor SACD can be so passed...

True for SACD, but with DVD-A, it depends on the specific disc
authoring. The DVD-A's I burn of my own hi-rez recordings will
result in a bit-accurate datastream appearing at the output of
DVD players other than the Oppo.

Do-it-yourself DVD-A hobbyists constitute at best a miniscule
share of the Oppo (or any other player) market, I suspect.

I agree, but I should point out that Minnetonka Software's
excellent, low-cost Discwelder Bronze program does
make it very easy for audiophiles to burn their own
hi-rez DVDs.

That leaves commercial DVD-A, of which only a very few in
my experience have ever offered S/PDIF playback of the
DVD-A folders...

AIX, HiRez Music, and Classic Records releases are the ones
that come to mind, where the disc authoring has allowed the
recording's full-resolution bit stream to be output from the
player's S/PDIF jack.

Note that I suspect the Oppo doesn't redither 24-bit data to
16, but merely truncates.

I'd call that bad design then. Yet, the '24-bit' material I play over the
Oppo seems to sound fine...

A feeling shared by the Stereophile reviewer. It is possible that
the recorded material has enough analog noise of the right spectrum
to be self-dithering. And some listeners have also preferred
truncation
to dithering in listening tests.

The argument may be moot, in any case, given the commercial failure
of DVD-A. I am told that the next-generation players' HDMI outputs
will
be able to provide a 24-bit/88k2 LPCM datastream from an SACD's DSD
layer, BTW.

John Atkinson
Editor, Stereophile

[David Spear]

"Simonel" wrote in message
...
I've never understood the whole concept of the "high-end CD transport".
The purpose of such a device is to read a stream of 1's and 0's off of
a spinning disc and output the (relatively low-frequency) digital
bitstream. Period. The bitstream is EXACTLY THE SAME, BIT FOR BIT,
coming out of a Walmart Electro-Sonic $39 special (assuming it has a
digital output) as out of a Krell. Except for the highest-end (read

most expensive) transports, they all use the same S/PDIF encoder chips
which are made but very few manufacturers.

In the May issue of Stereophile John Atkinson measured the Oppo
DV-970HD S/PDIF jacks and found out that they output 96kHz and 192
kHz, BUT, that they dithered 24-bit audio to 16-bit. Not all players
dither 24 bit to 16 bit (I am referring to the signal you get when you
play non-copyright protected DVD-A's, like privately burned ones.) So
the bitstream is not exactly the same coming out of the S/PDIF jacks
of different players.


The DAC is all that matters. It alone contains the magic smoke which
will reduce jitter, provide up-sampling and oversampling, and amplify
the resulting signal just so.

I agree that the quality of the DAC is more important than what comes
out of the S/PDIF jacks, but I have 2 players connected to the same
DAC1 through their S/PDIF jacks. One is an old Panasonic S55 and the
other a Denon 2910. The Denon gives me much better sound from DVD-V's.


Okay, here's one for you:

The output digital bitstream CANNOT VARY from transport to transport
unless gross errors are introduced by inaccurate Red Book audio CD
error detection, error correction, and concealment. It's 1's and 0's,
read directly off of the disk. Unless it is dithered, which is a whole
can of worms we won't open here. Any noise or other analog output on
the S/PDIF line is IRRELEVANT!

It's not an analog signal, the DAC is recovering the digital data and
converting it!

How can one possibly expect a "much better sound" from one transport
vs. another going into a common DAC? A good DAC should convert and
amplify the same digital signal the same way, how could it be
otherwise?

I had a long discussion about this awhile back on rec.audio.tech,
here's a snippet that sums it up:

"This is one of the areas in which the market, and its perceptions,
can be a bit backwards from reality. Some DAC-boxes have a reputation
for "revealing" the differences between different transports, and this
is often touted as a good and impressive characteristic. I see it
otherwise... a good DAC-box should be entirely immune to timing
jitter, noise on the S/PDIF signal, etc., and should always sound at
its best. DAC-boxes which "reveal" transport-related differences in
the S/PDIF signal are, I think, showing that their clock recovery
circuits are not robustly designed. "

Dave

[Steven Sullivan]

Simonel wrote:
On Jun 2, 10:53 am, "Arny Krueger" wrote:
Do you know of a review of a DVD-A player that says it outputs the full 24
bits from 24 bit DVD-As?

Good point, exactly what bothered me. I addressed this question to
John Atkinson, who gave me a list of players reviewed on Stereophile
that he measured to output 24 bits from 24 bit DVD-As, although for
some reason he never published the results. However, they were all
very expensive and I think do not justify the price as transports. I
am still awaiting a review of a consumer grade player that outputs 24
bits from 24 bit DVD-As.

Again, few if any consumers are ever going to get 'full' DVD-A from
S/PDIF outputs, even if said outputs can deliver 24 bit DVD-A playback...that's
because the vast majority of DVD-A releases forbid such S/PDIF output .
Only the tiny fraction who burn their own DVD-As will achieve it.

A perhaps more germane question is whether non-S/PDIF outputs like Denonlink
and ilink and HDMI deliver 24 bit DVD-A output on a given player.

This leaves aside of course the question of whether 24-bit delivery
formats even makes a difference for the consumer, given typical listening
environment noise levels and source dynamic ranges.

___
-S
"As human beings, we understand the world through simile, analogy,
metaphor, narrative and, sometimes, claymation." - B. Mason

[Steven Sullivan]

wrote:
On Jun 2, 8:29 pm, Steven Sullivan wrote:
wrote:
On Jun 1, 10:29 am, Steven Sullivan wrote:
First, how much 'hi rez' audio can be passed from S/PDIF?
Neither DVD-A nor SACD can be so passed...

True for SACD, but with DVD-A, it depends on the specific disc
authoring. The DVD-A's I burn of my own hi-rez recordings will
result in a bit-accurate datastream appearing at the output of
DVD players other than the Oppo.

Do-it-yourself DVD-A hobbyists constitute at best a miniscule
share of the Oppo (or any other player) market, I suspect.

I agree, but I should point out that Minnetonka Software's
excellent, low-cost Discwelder Bronze program does
make it very easy for audiophiles to burn their own
hi-rez DVDs.

From what sources, pray tell? Other than a live recording
the hobbyist has made him/herself at 24 bits? Burning
an LP-to-digital transfer at 24 bits is overkill. Burning
a CD, ditto.

That leaves commercial DVD-A, of which only a very few in
my experience have ever offered S/PDIF playback of the
DVD-A folders...

AIX, HiRez Music, and Classic Records releases are the ones
that come to mind, where the disc authoring has allowed the
recording's full-resolution bit stream to be output from the
player's S/PDIF jack.

Again, a teensy fragment of the market...notwithstanding
that the necessity of 24-bit home delivery itself is dubious.
I'd be more concerned about the lack of dithering, than the
lack of 24-bit output per se.

Note that I suspect the Oppo doesn't redither 24-bit data to
16, but merely truncates.

I'd call that bad design then. Yet, the '24-bit' material I play over the
Oppo seems to sound fine...

A feeling shared by the Stereophile reviewer. It is possible that
the recorded material has enough analog noise of the right spectrum
to be self-dithering. And some listeners have also preferred
truncation
to dithering in listening tests.

Also, in my case, the 24-bit sources are only being listened to via
the HDMI output, whose handling of 24-bit sources has not been
described. As for truncation being subjectively
*preferable* to dithering (and let's assume it was *good* dithering)
....oy vey. Do you have a reference for this?

The argument may be moot, in any case, given the commercial failure
of DVD-A. I am told that the next-generation players' HDMI outputs
will
be able to provide a 24-bit/88k2 LPCM datastream from an SACD's DSD
layer, BTW.

HDMI 1.2 , from 2005, already can handle 8-channel DSD . But afaik,
no player has ever implemented that, and we're already up to HDMI 1.3.

___
-S
"As human beings, we understand the world through simile, analogy,
metaphor, narrative and, sometimes, claymation." - B. Mason

[Codifus]

On Jun 3, 11:30 am, Simonel wrote:
On Jun 2, 10:53 am, "Arny Krueger" wrote:

Do you know of a review of a DVD-A player that says it outputs the full 24
bits from 24 bit DVD-As?

Good point, exactly what bothered me. I addressed this question to
John Atkinson, who gave me a list of players reviewed on Stereophile
that he measured to output 24 bits from 24 bit DVD-As, although for
some reason he never published the results. However, they were all
very expensive and I think do not justify the price as transports. I
am still awaiting a review of a consumer grade player that outputs 24
bits from 24 bit DVD-As.

Simonel

(BTW - although DVD-V's sound better on my Denon, CD's and DVD-A's
sound the same from the old Panasonic. Go figure.)

This seems definitely plausible. The DVD-V soundtrack is compressed,
lossy, multi-channel audio data. The Panasonic and Denon probably have
different types of circuits to expand/extract that data to 0s and 1s
before pushing it out to the DAC. The difference in the sound probably
lies there, and the fact that you observe the Denon And Panny sounding
the same when playing DVD-As and CDs makes complete sense. They're
only transporting the 0s and 1s to the DAC, that's it.

CD

[Codifus]

On Jun 2, 10:47 am, Steven Sullivan wrote:
codifus wrote:
On May 31, 10:50 am, Simonel wrote:
I've never understood the whole concept of the "high-end CD
transport". The purpose of such a device is to read a stream of 1's
and 0's off of a spinning disc and output the (relatively
low-frequency) digital bitstream. Period. The bitstream is EXACTLY THE
SAME, BIT FOR BIT, coming out of a
Walmart Electro-Sonic $39 special (assuming it has a digital output)
as out of a Krell. Except for the highest-end (read most expensive)
transports, they all use the same S/PDIF encoder chips which are made
but very few manufacturers.

In the May issue of Stereophile John Atkinson measured the Oppo
DV-970HD S/PDIF jacks and found out that they output 96kHz and 192
kHz, BUT, that they dithered 24-bit audio to 16-bit. Not all players
dither 24 bit to 16 bit (I am referring to the signal you get when you
play non-copyright protected DVD-A's, like privately burned ones.) So
the bitstream is not exactly the same coming out of the S/PDIF jacks
of different players.

The DAC is all that matters. It alone contains the magic smoke which
will reduce jitter, provide up-sampling and oversampling, and amplify
the resulting signal just so.

I agree that the quality of the DAC is more important than what comes
out of the S/PDIF jacks, but I have 2 players connected to the same
DAC1 through their S/PDIF jacks. One is an old Panasonic S55 and the
other a Denon 2910. The Denon gives me much better sound from DVD-V's.

Simonel
It may be nice that DVD-Vs sound better from the Denon, but given that
the sound format from DVD-V is not a serious audiophile format, why
should it matter so much? The priority of DVD-V sound is to transport
multiple channels of whizz bang movie soundtracks which aren't really
produced for the utmost in sound quality. I would think of it more as
an added plus that the DVD-Vs sound better. It's like trying to make
MP3s into and audiophile format.

All that's required for an MP3 to be 'audiophile' is for the MP3 to be
indistinguishable from the source by the 'audiophile' listener. Which is well
within the range of possible.

Opera , symphonies, etc have been released on DVD-V, so DD/DTS are not necessarily
for 'whizz bang' only.

___
-S
"As human beings, we understand the world through simile, analogy,
metaphor, narrative and, sometimes, claymation." - B. Mason

Fair enough, but you have to admitt that the whizz bang soundtracks
comprise a significant majority of soundtracks on DVD-V.

CD

[Codifus]

On Jun 2, 10:50 am, bob wrote:
On Jun 1, 10:33 am, codifus wrote:

It may be nice that DVD-Vs sound better from the Denon, but given that
the sound format from DVD-V is not a serious audiophile format, why
should it matter so much? The priority of DVD-V sound is to transport
multiple channels of whizz bang movie soundtracks which aren't really
produced for the utmost in sound quality.

The Last Waltz?

Stop Making Sense?

ANY opera or ballet?

C'mon. I have my doubts about the real sonic basis of Simonel's
perceptions, but the desire for high-fidelity soundtracks is quite
legitimate.

bob
As I mentioned in another post, it's easy to see that whizz bang
soundtracks comprise a significant majority of soundtracks on DVD-V.
The High quality soundtracks are out there, but, like Classical music
in CD format, they surely comprise is very small minority of the
total.

CD

[jeffc]

"David Spear" wrote in message
...


I've never understood the whole concept of the "high-end CD
transport". The purpose of such a device is to read a stream of 1's
and 0's off of a spinning disc and output the (relatively
low-frequency) digital bitstream. Period. The bitstream is EXACTLY THE
SAME, BIT FOR BIT, coming out of a

Walmart Electro-Sonic $39 special (assuming it has a digital output)
as out of a Krell.

The problem is how it's read. On a computer, it works with cheap units
because they work differently. On an audio device, it has to read the bits
in real time. Turns out, that's a harder problem. Shouldn't be difficult,
but my understanding is that most audio CD players simply aren't engineered
correctly to do that.

[Simonel]

On Jun 3, 11:30 am, Simonel wrote:

(BTW - although DVD-V's sound better on my Denon, CD's and DVD-A's
sound the same from the old Panasonic. Go figure.)

On Jun 4, 7:12 pm, codifus wrote:

This seems definitely plausible. The DVD-V soundtrack is compressed,
lossy, multi-channel audio data. The Panasonic and Denon probably have
different types of circuits to expand/extract that data to 0s and 1s
before pushing it out to the DAC. The difference in the sound probably
lies there, and the fact that you observe the Denon And Panny sounding
the same when playing DVD-As and CDs makes complete sense. They're
only transporting the 0s and 1s to the DAC, that's it.

Thanks, this is the first time someone has actually offered an
explanation rather than suggest it's in my head and/or pound ex
cathedra pronouncements. I have a large collection of music DVD's,
which are a big market (check for example http://opera_on_dvd.home.att.net/
- and this is just the opera corner of that market.) Sound quality of
DVD-V's matters a great deal and varies significantly between players,
so it makes sense to invest in a quality product - Simonel

[ScottW]

"jeffc" wrote in message
...
"David Spear" wrote in message
...


I've never understood the whole concept of the "high-end CD
transport". The purpose of such a device is to read a stream of 1's
and 0's off of a spinning disc and output the (relatively
low-frequency) digital bitstream. Period. The bitstream is EXACTLY THE
SAME, BIT FOR BIT, coming out of a

Walmart Electro-Sonic $39 special (assuming it has a digital output)
as out of a Krell.

The problem is how it's read. On a computer, it works with cheap units
because they work differently. On an audio device, it has to read the bits
in real time. Turns out, that's a harder problem.

Odd...then why does my PC CD Rom or DVD drive have no problem
reading and playing audio CDs in very nearly the same "real" time as
an audio CD player?
Simple answer is because at the time of reading the bits off the disc,
the process is virtually the same. Then they must go through the
de-interleaving which means at least one spiral of data is
in buffer at all times. Nothing truly real time about it.
Since minor data errors in audio aren't catastrophic the playback
of an audio CD doesn't have near the error correction of a data CD
nor does it have the required data rate so data reading is actually
more complicated, not less.

How Stuff works has a good basic description of the process.
http://www.howstuffworks.com/cd.htm

Shouldn't be difficult,
but my understanding is that most audio CD players simply aren't engineered
correctly to do that.

I cannot agree with that.

ScottW

[Arny Krueger]

"Steven Sullivan" wrote in message


I'd call that bad design then.

True, but mostly on philosophical grounds.

Yet, the '24-bit'
material I play over the Oppo seems to sound fine.

Real world recordings have so much built-in noise that they are
self-dithering in a true 16 bit system.

This
could be because the dynamic range of the source material
or the mastering is already well within 16,

Almost always true.

much less 24 bit, limits, or because the extra 8 bits were just
'padding' (implying a lower-bit stage somewhere in
recording/production), or because I'm not set up to
easily properly compare it to a player that doesn't
truncate.

By definition, quantization error is always = 1 LSB. For a 16 system that
puts it at least 90 dB down. Most forms of distortion are inaudible when
they are = 80 dB down.

[Arny Krueger]

"ScottW" wrote in message

"jeffc" wrote in message
...
"David Spear" wrote in message
...


I've never understood the whole concept of the
"high-end CD transport". The purpose of such a device
is to read a stream of 1's and 0's off of a spinning
disc and output the (relatively low-frequency) digital
bitstream. Period. The bitstream is EXACTLY THE SAME,
BIT FOR BIT, coming out of a
Walmart Electro-Sonic $39 special (assuming it has a
digital output) as out of a Krell.

The problem is how it's read. On a computer, it works
with cheap units because they work differently. On an
audio device, it has to read the bits in real time.
Turns out, that's a harder problem.

Odd...then why does my PC CD Rom or DVD drive have no
problem reading and playing audio CDs in very nearly the same
"real" time as an audio CD player?

(1) Neither the audio player nor the PC CD ROM should have a problem, under
ideal conditions.

(2) The means actually used to "play" an audio CD on a PC is up to the
program that is supervising the operation and the CD ROM drive itself. A PC
CDROM, depending on the program supervising the operation, can read an audio
CD using its own firmware for reading audio CDs. The details of how this
firmware works are up to the manufacturer. There are several options that
the frimware can pick from. Again, depending on the program supervising the
operation, a PC CDROM can read the audio CD using the CDROM's firmware for
reading data CDs. This is much more work for the program, but it can
produce superior results.

Simple answer is because at the time of reading the bits
off the disc, the process is virtually the same. Then
they must go through the de-interleaving which means at
least one spiral of data is
in buffer at all times. Nothing truly real time about it.

This is true for programs that read an audio CD using the CDROM's firmware
for reading audio CDs.

Since minor data errors in audio aren't catastrophic the
playback of an audio CD doesn't have near the error correction of
a data CD nor does it have the required data rate so data
reading is actually more complicated, not less.

If the program supervising the process chooses to read the audio CD by
alternative means, then the program can implement more sophisticated
strategies for reading the CD, and can for example handle retrying reads
that fail using strategies of its own devising.

How Stuff works has a good basic description of the
process. http://www.howstuffworks.com/cd.htm

This is a rather basic explanation, and does not get into a lot of relevant
details.

Here is a more detailed description of some of the issues related to the
second method for reading audio CDs:

http://teamcombooks.com/mp3handbook/15.htm

Shouldn't be difficult,
but my understanding is that most audio CD players
simply aren't engineered correctly to do that.

I cannot agree with that.

ScottW

[[email protected]]

On Jun 4, 7:12 pm, Steven Sullivan wrote:
wrote:
Steven Sullivan wrote:
...

I should point out that Minnetonka Software's
excellent, low-cost Discwelder Bronze program does
make it very easy for audiophiles to burn their own
hi-rez DVDs.

From what sources, pray tell? Other than a live recording
the hobbyist has made him/herself at 24 bits?

I have 100s of hours of such recordings that burn on to DVD-A
for friends. Others send me _their_ hi-rez recordings on self
burned DVD-As. Also, several companies are planning downloads
of 2-channel hir-rez recordings, which could also be burned
to DVD for personal convenience if not for distribution.

Burning an LP-to-digital transfer at 24 bits is overkill.

In your opinion. Others disagree. I don't see why such people
should be discourage from experimenting.

the '24-bit' material I play over the Oppo seems to sound fine...

A feeling shared by the Stereophile reviewer. It is possible that
the recorded material has enough analog noise of the right spectrum
to be self-dithering. And some listeners have also preferred
truncation to dithering in listening tests.

As for truncation being subjectively *preferable* to dithering (and
let's assume it was *good* dithering) ...oy vey. Do you have a
reference for this?

See Keith Howard's "Contingent Dither" article in
Stereophile: http://www.stereophile.com/features/705dither/ .

Also, in my case, the 24-bit sources are only being listened to via
the HDMI output, whose handling of 24-bit sources has not been
described.

No, I have no HDMI-equipped playback components with which to
test this aspect of the Oppo's performance.

The argument may be moot, in any case, given the commercial failure
of DVD-A. I am told that the next-generation players' HDMI outputs
will be able to provide a 24-bit/88k2 LPCM datastream from an
SACD's DSD layer, BTW.

HDMI 1.2 , from 2005, already can handle 8-channel DSD . But afaik,
no player has ever implemented that, and we're already up to HDMI 1.3.

I wouldn't hold my breath waiting. Personally, I think the
introduction of
the HDMI standard has been mishandled. But that's just me.

John Atkinson
Editor, Stereophile

[Codifus]

On Jun 6, 12:29 am, Simonel wrote:
On Jun 3, 11:30 am, Simonel wrote:

(BTW - although DVD-V's sound better on my Denon, CD's and DVD-A's
sound the same from the old Panasonic. Go figure.)

On Jun 4, 7:12 pm, codifus wrote:

This seems definitely plausible. The DVD-V soundtrack is compressed,
lossy, multi-channel audio data. The Panasonic and Denon probably have
different types of circuits to expand/extract that data to 0s and 1s
before pushing it out to the DAC. The difference in the sound probably
lies there, and the fact that you observe the Denon And Panny sounding
the same when playing DVD-As and CDs makes complete sense. They're
only transporting the 0s and 1s to the DAC, that's it.

Thanks, this is the first time someone has actually offered an
explanation rather than suggest it's in my head and/or pound ex
cathedra pronouncements. I have a large collection of music DVD's,
which are a big market (check for examplehttp://opera_on_dvd.home.att.net/
- and this is just the opera corner of that market.) Sound quality of
DVD-V's matters a great deal and varies significantly between players,
so it makes sense to invest in a quality product - Simonel

It's quite odd that the difference you pointed out between the cheap
Panasonic and the more expensive Denon is more apparent in how they
play lossy audio than when they play hi rez lossless audio. You would
think it would be the reverse.

I know I've always like my Panny

CD

[[email protected]]

On Jun 6, 9:56 pm, "Arny Krueger" wrote:
"Steven Sullivan" wrote in message
Yet, the '24-bit'
material I play over the Oppo seems to sound fine.

Real world recordings have so much built-in noise that they are
self-dithering in a true 16 bit system.

If and only if the "built-in-noise" has the appropriate properties.
Room noise that has any component periodicity like motor
hum or the like is perfectly awful as dither.

Depending upon system, environment or other "built-in"
noise for dither is a really BAD idea. Eliminating quantization
error through techniques like dither or noise shaping is
a solved problem, why would anyone want to depend upon
unpredictable sources of noise energy for dither?

much less 24 bit, limits, or because the extra 8 bits were just
'padding' (implying a lower-bit stage somewhere in
recording/production), or because I'm not set up to
easily properly compare it to a player that doesn't
truncate.

By definition, quantization error is always = 1 LSB.

re full scale.

For a 16 system that
puts it at least 90 dB down.

re Full scale.

Most forms of distortion are inaudible when
they are = 80 dB down.

That's right, but high-quality, wide dynamic range recordings
are seldom sitting at full scale. Consider any wide-dynamic
range classical recording that might have much of its stuff
sitting 40 dB below full scale. Now that -90 dB for quantization
artifacts doesn't look so good any more, it's now sitting down
at -50 dB. And while the artifacts are "correlated," they are not
necessarily "harmonic," given that they are aliased all through
the audible spectrum.

[[email protected]]

On Jun 3, 11:32 am, wrote:
A feeling shared by the Stereophile reviewer. It is possible that
the recorded material has enough analog noise of the right spectrum
to be self-dithering.

It might be possible, but it's not guaranteed. In fact, there's
enough opinion among the professional community to suggest
at the sorts of precisions required for end-user delivery, it's not
very likely to be correct.

And some listeners have also preferred
truncation to dithering in listening tests.

Uh, hold it a second. Dithering and truncation are two
separate stages in the requantizing process. They are
NOT substitutes for one another, as the statement seems
to imply. Truncation is simply the discarding of precision,
and dithering is made necessary by truncation in order to
preserve the data lost through truncation.

If you're going from a higher precision (e.g., 18-
or 24-bit) to a lower precision (e.g. 16 bit), you ARE
truncating, no if ands or buts. So whether these listeners
think so or not, they are always listening to truncated
audio. There are different types of truncation, such as
rounding, floor, ceiling, simple length truncation and so
on. The effects of these choices is VERY insignicant,
far less than people seem to want to claim. For example,
the effective difference between straight discarded truncation
and arithmetic rounding is to simply impose a 1/2 LSB
DC offset on the result.

Now, whether the audio is dithered FIRST or
not IS a choice.

If the choice they're really making is dithered vs non-
dithered, then that's exactly the same choice as between
no quantization artifacts vs quantization artifacts. If they
prefer the quantization artifacts, that's their choice. I shan't
comment on how dumb a choice it is, though.

[Arny Krueger]

wrote in message


On Jun 6, 9:56 pm, "Arny Krueger" wrote:
"Steven Sullivan" wrote in message

Yet, the '24-bit'
material I play over the Oppo seems to sound fine.

Real world recordings have so much built-in noise that
they are self-dithering in a true 16 bit system.

If and only if the "built-in-noise" has the appropriate
properties. Room noise that has any component periodicity
like motor hum or the like is perfectly awful as dither.

I can't believe that you're actually saying this. It is simply not true.
Onc there is enough noise to properly dither a quantization step no matter
where the noise comes from, whatever else you have is just signal or noise,
depending on your preferences. There is no combination of signals that can
cause a properly-dithered quantization to become undithered.

Depending upon system, environment or other "built-in"
noise for dither is a really BAD idea.

Agreed - it is living dangerously.

Eliminating
quantization error through techniques like dither or
noise shaping is
a solved problem, why would anyone want to depend upon
unpredictable sources of noise energy for dither?

Good point, but from time to time people actually do such things, probably
by accident. The Oppo DVD player apparently is a practical example.

Just to be clear I never recommend that others, nor do I myself
intentionally quantize without proper dithering, except as an experiment.

However there's a lot of folklore about dithering, which tends to recede in
importance when put under the microscope of properly-designed listening
tests.

much less 24 bit, limits, or because the extra 8 bits
were just 'padding' (implying a lower-bit stage
somewhere in recording/production), or because I'm not
set up to
easily properly compare it to a player that doesn't
truncate.

By definition, quantization error is always = 1 LSB.

re full scale.

Agreed, and I see where you're headed.

For a 16 system that puts it at least 90 dB down.

re Full scale.

Agreed.

Most forms of distortion are inaudible when
they are = 80 dB down.

That's right, but high-quality, wide dynamic range
recordings are seldom sitting at full scale. Consider any
wide-dynamic range classical recording that might have much of its
stuff sitting 40 dB below full scale. Now that -90 dB for
quantization artifacts doesn't look so good any more,
it's now sitting down
at -50 dB. And while the artifacts are "correlated," they
are not necessarily "harmonic," given that they are
aliased all through the audible spectrum.

Agreed that quantization error can be very nasty-sounding stuff.

The argument that the a -50 dB signal is only 40 dB above the -90 dB
quantization artifacts must be tempered by the fact that the artifacts are
still well below the acoustical noise in the listening room and at or below
the listener's practical threshold of hearing.

We can argue all day about the theoretical reasons why, but the fact is that
it's difficult or impossible to show that people can hear the difference
between a optimally dithered and undithered quantization to 16 bits, as long
as listeners aren't allowed to turn up the gain during quiet passages. This
experiment has been done many times, and the results are pretty consistent
provided proper testing methodologies are used (unlikely for the average
audiophile or a audiophile magazine reviewer).

Note that following the undithered quantification with a fair amount of
dynamics compression can accomplish about the same thing as allowing the
listener to turn up the gain during the quiet passages, and thus possibly
make an unwise choice of quantization strategy more likely to be audible.

[Andrew Haley]

writes:
On Jun 4, 7:12 pm, Steven Sullivan wrote:

As for truncation being subjectively *preferable* to dithering (and
let's assume it was *good* dithering) ...oy vey. Do you have a
reference for this?

See Keith Howard's "Contingent Dither" article in
Stereophile: http://www.stereophile.com/features/705dither/ .

I guess it isn't so very surprising that people might prefer simple
truncation to dithered truncation. I believe that truncation is
sometimes preferred in low-bit ADCs because it results in an audibly
lower noise floor. It's not inconceivable that some people might
prefer undithered quantization in high-bit conversion, especially if
they ride the volume control in quiet passages.

It would be interesting to perform an experiment to find out where the
sensitivity threshold is. Can you tell the difference with 12-bit
truncation? 14-bit? 16-bit? And so on.

Andrew.

[Arny Krueger]

"Andrew Haley" wrote in message

writes:
On Jun 4, 7:12 pm, Steven Sullivan
wrote:

As for truncation being subjectively *preferable* to
dithering (and let's assume it was *good* dithering)
...oy vey. Do you have a reference for this?

See Keith Howard's "Contingent Dither" article in
Stereophile:
http://www.stereophile.com/features/705dither/ .

I guess it isn't so very surprising that people might
prefer simple truncation to dithered truncation.

It's pretty much moot at the 16 bit level in most cases.

I believe that truncation is sometimes preferred in low-bit
ADCs because it results in an audibly lower noise floor.

Usually the noise floor that came with the music is very much higher. The
dynamic range of live performance and orchestral recordings is often in the
65 dB range. This corresponds to about 12 bits.

It's not inconceivable that some people might prefer
undithered quantization in high-bit conversion,
especially if they ride the volume control in quiet
passages.

The gain-riding would have to be pretty extreme.

It would be interesting to perform an experiment to find
out where the sensitivity threshold is. Can you tell the
difference with 12-bit truncation? 14-bit? 16-bit? And
so on.

You can easily do this experiment with files that you can download from

http://www.pcabx.com/technical/bits44/index.htm

Most people find that the audibility of bit truncation is very difficult to
detect at 14 bits, or less bits.