Sign in with
Sign up | Sign in
Your question

CD vs. SACD

Last response: in Home Audio
Share
Anonymous
April 13, 2005 7:26:02 PM

Archived from groups: rec.audio.tech (More info?)

What's the story here?

More about : sacd

April 14, 2005 1:42:12 AM

Archived from groups: rec.audio.tech (More info?)

In article <jne7e.2964$O11.2375@fe12.lga>, post news <thanks@for.info>
wrote:

> What's the story here?

SACD has a longer word length, higher sampling rate, higher cost than CD
and in some cases.. surround as opposed to CD's stereo only.

Many will argue about what's better, but thats the jist of it.

hth,

--
Cyrus

*coughcasaucedoprodigynetcough*
Anonymous
April 14, 2005 4:26:17 PM

Archived from groups: rec.audio.tech (More info?)

post news wrote:
> What's the story here?


I guess I'm not buying that SACD player then, thanks, more money for
CD's now.
Related resources
April 14, 2005 5:40:11 PM

Archived from groups: rec.audio.tech (More info?)

On 4/14/2005 11:26 AM, post news wrote:
> post news wrote:
>
>> What's the story here?
>
>
>
> I guess I'm not buying that SACD player then, thanks, more money for
> CD's now.

Isn't SACD obsolete as the new cd players are going to be DVD audio?
Anonymous
April 14, 2005 7:45:39 PM

Archived from groups: rec.audio.tech (More info?)

On Thu, 14 Apr 2005 13:40:11 -0500, Dan <dan@nospam.com> wrote:

>Isn't SACD obsolete as the new cd players are going to be DVD audio?

Or was it the other way around? ;-)

Kal
Anonymous
April 15, 2005 2:59:00 AM

Archived from groups: rec.audio.tech (More info?)

"post news" <thanks@for.info> wrote in message
news:jne7e.2964$O11.2375@fe12.lga...
> What's the story here?

Dunno it anybody has produced a CD and SACD from the exact same master yet.
That owuld be necessary to tell if the CD version is seriously (or at all)
compromised wrt the SACD version.


geoff
Anonymous
April 15, 2005 2:59:01 AM

Archived from groups: rec.audio.tech (More info?)

Geoff Wood wrote:
> "post news" <thanks@for.info> wrote in message
> news:jne7e.2964$O11.2375@fe12.lga...
>> What's the story here?
>
> Dunno it anybody has produced a CD and SACD from the exact same
> master yet. That owuld be necessary to tell if the CD version is
> seriously (or at all) compromised wrt the SACD version.
>
>
> geoff


Such a disc would hardly help sell SACD, now would it?


Mark Z.
Anonymous
April 15, 2005 2:59:01 AM

Archived from groups: rec.audio.tech (More info?)

On Thu, 14 Apr 2005 22:59:00 +1200, "Geoff Wood"
<geoff@nospam-paf.co.nz> wrote:

>
>"post news" <thanks@for.info> wrote in message
>news:jne7e.2964$O11.2375@fe12.lga...
>> What's the story here?
>
>Dunno it anybody has produced a CD and SACD from the exact same master yet.

Problematic. Would that master be in 44.1/16 PCM or DSD format?
Either way, the other would require recoding. Except for that caveat,
there are several that come close to matching your standard.

Kal
Anonymous
April 15, 2005 2:59:02 AM

Archived from groups: rec.audio.tech (More info?)

Mark D. Zacharias wrote:
> Geoff Wood wrote:
>> "post news" <thanks@for.info> wrote in message
>> news:jne7e.2964$O11.2375@fe12.lga...
>>> What's the story here?
>>
>> Dunno it anybody has produced a CD and SACD from the exact same
>> master yet. That owuld be necessary to tell if the CD version is
>> seriously (or at all) compromised wrt the SACD version.

> Such a disc would hardly help sell SACD, now would it?

LOL!
Anonymous
April 15, 2005 2:59:02 AM

Archived from groups: rec.audio.tech (More info?)

Mark D. Zacharias wrote:
> Geoff Wood wrote:
>
>>"post news" <thanks@for.info> wrote in message
>>news:jne7e.2964$O11.2375@fe12.lga...
>>
>>>What's the story here?
>>
>>Dunno it anybody has produced a CD and SACD from the exact same
>>master yet. That owuld be necessary to tell if the CD version is
>>seriously (or at all) compromised wrt the SACD version.
>>
>>
>>geoff
>
>
>
> Such a disc would hardly help sell SACD, now would it?
>
>
> Mark Z.
>
>
Would Sony/Philips sue the creator of such a disc? after all, it only
proves their first invention sounds as good as their second. Talk about
a catch-22.

CD
Anonymous
April 15, 2005 2:59:03 AM

Archived from groups: rec.audio.tech (More info?)

On Thu, 14 Apr 2005 10:23:47 -0400, Codifus <codifus@optonline.net>
wrote:

>Mark D. Zacharias wrote:
>> Geoff Wood wrote:
>>
>>>"post news" <thanks@for.info> wrote in message
>>>news:jne7e.2964$O11.2375@fe12.lga...
>>>
>>>>What's the story here?
>>>
>>>Dunno it anybody has produced a CD and SACD from the exact same
>>>master yet. That owuld be necessary to tell if the CD version is
>>>seriously (or at all) compromised wrt the SACD version.
>>>
>>>
>>>geoff
>>
>>
>>
>> Such a disc would hardly help sell SACD, now would it?
>>
>>
>> Mark Z.
>>
>>
>Would Sony/Philips sue the creator of such a disc? after all, it only
>proves their first invention sounds as good as their second. Talk about
>a catch-22.

Well, since some lame-brained marketing wag advertised the first one
as 'perfect sound forever', that would kind of preclude *any*
improvement, wouldn't it? :-)
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
April 15, 2005 12:19:59 PM

Archived from groups: rec.audio.tech (More info?)

"Kalman Rubinson" <kr4@nyu.edu> wrote in message
news:e93t511dn2u8bquc20h2o3tb84pkv92ild@4ax.com...
> On Thu, 14 Apr 2005 22:59:00 +1200, "Geoff Wood"
> <geoff@nospam-paf.co.nz> wrote:
>
>>
>>"post news" <thanks@for.info> wrote in message
>>news:jne7e.2964$O11.2375@fe12.lga...
>>> What's the story here?
>>
>>Dunno it anybody has produced a CD and SACD from the exact same master
>>yet.
>
> Problematic. Would that master be in 44.1/16 PCM or DSD format?
> Either way, the other would require recoding. Except for that caveat,
> there are several that come close to matching your standard.

Doubtlessly the master would be at a higher bit-depth or even sample rate,
but if the only difference was what was incurred in teh transcoding, that
would say something about the extent of the differences, no ?

Or how about a 'direct to disc' analogue source ?

geoff
Anonymous
April 15, 2005 12:20:00 PM

Archived from groups: rec.audio.tech (More info?)

On Fri, 15 Apr 2005 08:19:59 +1200, "Geoff Wood"
<geoff@nospam-paf.co.nz> wrote:

> Doubtlessly the master would be at a higher bit-depth or even sample rate,
>but if the only difference was what was incurred in teh transcoding, that
>would say something about the extent of the differences, no ?

Sure. There are a few. WaterLily did one with the Philadelphia Orch
in CD, SACD and DVD-A. There's also a neat disc from Musical Fidelity
with the DSD on the SACD layer and downsampled on the CD layer. It
also has an analog source converted to DSD on the SACD layer and
converted to PCM on the CD layer.

Kal
Anonymous
April 15, 2005 12:20:01 PM

Archived from groups: rec.audio.tech (More info?)

On Fri, 15 Apr 2005 10:21:02 +0930, Kalman Rubinson <kr4@nyu.edu> wrote:

> On Fri, 15 Apr 2005 08:19:59 +1200, "Geoff Wood"
> <geoff@nospam-paf.co.nz> wrote:
>
>> Doubtlessly the master would be at a higher bit-depth or even sample
>> rate,
>> but if the only difference was what was incurred in teh transcoding,
>> that
>> would say something about the extent of the differences, no ?
>
> Sure. There are a few. WaterLily did one with the Philadelphia Orch
> in CD, SACD and DVD-A. There's also a neat disc from Musical Fidelity
> with the DSD on the SACD layer and downsampled on the CD layer. It
> also has an analog source converted to DSD on the SACD layer and
> converted to PCM on the CD layer.
>
> Kal
>
New release from Naxos -CANTELOUBE: Chants d'Auvergne (Selection) 6.1100
SACD

CANTELOUBE: Chants d'Auvergne (Selection) 5.1100 Audio CD
The CD is an excelent recording in my opinion
--
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
Anonymous
April 15, 2005 12:20:01 PM

Archived from groups: rec.audio.tech (More info?)

Kalman Rubinson wrote:
> On Fri, 15 Apr 2005 08:19:59 +1200, "Geoff Wood"
> <geoff@nospam-paf.co.nz> wrote:
>
>> Doubtlessly the master would be at a higher bit-depth or even
sample
>> rate, but if the only difference was what was incurred in teh
>> transcoding, that would say something about the extent of the
>> differences, no ?
>
> Sure. There are a few. WaterLily did one with the Philadelphia
Orch
> in CD, SACD and DVD-A. There's also a neat disc from Musical
Fidelity
> with the DSD on the SACD layer and downsampled on the CD layer. It
> also has an analog source converted to DSD on the SACD layer and
> converted to PCM on the CD layer.

Do these recordings have related names?
Anonymous
April 15, 2005 4:20:00 PM

Archived from groups: rec.audio.tech (More info?)

On Fri, 15 Apr 2005 03:01:33 GMT, "Kerry Gascoigne"
<Kerry.Gascoigne@flinders.edu.au> wrote:

>>
>New release from Naxos -CANTELOUBE: Chants d'Auvergne (Selection) 6.1100
>SACD
>
>CANTELOUBE: Chants d'Auvergne (Selection) 5.1100 Audio CD
>The CD is an excelent recording in my opinion

It's an OK recording of an OK performance. I've not heard the CD but
I am judging from the DVD-A and the SACD. Typical of Naxos, it is
recorded PCM and the SACD is made from a transfer.

Kal
Anonymous
April 15, 2005 4:21:41 PM

Archived from groups: rec.audio.tech (More info?)

On Fri, 15 Apr 2005 07:52:32 -0400, "Arny Krueger" <arnyk@hotpop.com>
wrote:

>Do these recordings have related names?
Sure but I am not at home so I cannot just pick them up and look.
The WaterLily is the only one they made with the Philadelphia
Orchestra and the Musical Fidelity is Mozart Clarinet Concerto.

Kal
Anonymous
April 15, 2005 10:34:13 PM

Archived from groups: rec.audio.tech (More info?)

It would be better to have the same performance, same mics, but
recorded by two machines (or three), analog, PCM and DSD.

>>Super Audio Center established to meet growing demand for SA-CD
production
October 26, 2004
San Francisco, October 28, 2004 -On the eve of the 117th Audio
Engineering Society (AES) Convention, a new company has been formed to
meet the growing demand for SA-CD music production.

The Super Audio Center, LLC (SAC) will be directed by Gus Skinas,
formerly head of Field Operations for the Super Audio Project at Sony
Corporation of America. During his tenure at Sony, Mr. Skinas launched
the first production center in the U.S. dedicated to providing a
comprehensive array of support services for SA-CD clientele, including
recording and tape transfer, mixing, editing, mastering and authoring
for SA-CD projects.

The new Super Audio Center will be based in Boulder, Colorado and will
work closely with field engineers in both Los Angeles and New York to
support remote recording and mix sessions wherever they may occur.

SA-CD Momentum

"As the Super Audio CD format continues to gather momentum, a number
of new businesses have been spawned," said David Kawakami, Director
of the Super Audio CD Project for Sony Corporation of America. "Under
Gus's guidance, this new SA-CD facility will enable us to better
address the needs of those artists, producers and music labels who want
to issue their music on this exciting, high resolution multi-channel
format."

SA-CD compatible products are currently being offered by more than 26
manufacturers, with nearly 120 different models available. According to
the Recording Industry Association of America (RIAA), more than 1.3
million SA-CD albums were shipped in the U.S. in 2003 alone, bringing
the total installed base of SA-CD software to over 12 million discs
life-to-date. There are now more than 2,500 SA-CD titles in release
worldwide, with almost half of them available in multi-channel
surround.

Sonoma Workstation Support

The new Super Audio Center will also assume the responsibility for the
management and support of the Sonoma Workstation-a standard tool used
for recording and editing SA-CD projects in their native Direct Stream
Digital (DSD) domain.

Utilizing converters designed by Ed Meitner of EMM Labs, the Sonoma is
widely regarded as one of the most transparent recording systems
marketed today. The new SAC organization will be chartered to not only
lease Sonoma products, but distribute them as well, making it a
critical factor to ensure SA-CD's continued growth.

According to Tom Jung, president and chief engineer of DMP Records,
"The Sonoma System, when optically interfaced with Ed Meitner's DSD
converters, is the most musically accurate recording and editing system
available today at any price, period."

New Sonoma Model

During the 117th AES Convention, SAC will be demonstrating a Sonoma
Workstation capable of recording and editing 24 tracks of DSD audio.

"Many of our clients have expressed a need to record and edit more
DSD channels in order to handle more complex pop music projects,"
said SAC director Gus Skinas. "The new Sonoma's added capacity
makes it possible for us to offer up to 24 channels of high quality DSD
audio in a package that behaves much like a standard multi-track
recorder but delivers powerful editing as well." "With DSD, we are
much closer to achieving the true capability of digital recording,"
said producer Michael Beinhorn. "The Sonoma machines are, in my
opinion, sonically unsurpassed and have become a staple for my
mixdowns. I can only imagine how remarkable the new 24-track Sonoma
is."

The new Sonoma Workstation will be shown in the Philips-sponsored Super
Audio CD Booth #536 and in demonstrations sponsored by ATC Loudspeaker
Technology in room 224. <<
Anonymous
April 16, 2005 11:36:17 AM

Archived from groups: rec.audio.tech (More info?)

calcerise@hotmail.com wrote:

> It would be better to have the same performance, same mics, but
> recorded by two machines (or three), analog, PCM and DSD.

Only if you believe in the universal existance of audible differences
and have the logical abilites of a turnip.

SACD and DVD-A are supposed to be superset formats as compared to
CD-A. They's supposed to always sound better when given access to
program material that is unfettered by the shackles and chains of the
earlier "obviously-flawed" 16/44 CD-A digital format.

If you believe hi-rez recording enthusiasts, then its even reasonable
to expect old 48 KHz digital masters to sound better when transcribed
to SACD and/or DVD-A.

If any of that is true, then simply transcribing any randomly-chosen
SACD to the CD format should result in a clearly audible loss of sound
quality.

You shouldn't need carefully chosen program material, a megabuck
high-end audio system, or a listening panel composed of the best
reviewers from the three leading high end ragazines. Is SACD or DVD-A
is effective for mid-fi systems, then the audible advantages of these
formats should be obvious on mid-fi systems. Its just a matter of
listening, right?

As soon as the hi-rex format clique start retrenching and saying that
SACD and DVD-A can't possibly show an audible advantage without an
endless list of careful mouth-holding exercises, then they are
basically falling in behind the critics of these formats.
Anonymous
April 16, 2005 2:41:29 PM

Archived from groups: rec.audio.tech (More info?)

The Krouchebagg's logical flaw on this one is so obvious that a turnip
like me will let everyone else figure it out for themselves.

That said, it's just obvious that highbit digital will more closely
approximate an analog signal asymptotically as its bitrate increases.

SACD and DVD-A have never impressed me as much as a few occasions of
listening to either first rate vinyl or 1/2" 30 ips tape on a well set
up Ampex. However the variety of source material on highbit hasn't been
as great yet, and naysayers aren't helping the situation.
Anonymous
April 16, 2005 6:58:12 PM

Archived from groups: rec.audio.tech (More info?)

calcerise@hotmail.com wrote:

The prerequisite perenial dumasses personal attack:

> The Krouchebagg's logical flaw on this one is so obvious that a
turnip
> like me will let everyone else figure it out for themselves.

> That said, it's just obvious that highbit digital will more closely
> approximate an analog signal asymptotically as its bitrate
increases.

Why just highbit digital signals, Cal?

Isn't it true that increasing the bitrate doesn't necessarily increase
precision, particularly if the increased bitrate comes from increased
sample rate and the signal being digitized is band-limted?

> SACD and DVD-A have never impressed me as much as a few occasions of
> listening to either first rate vinyl or 1/2" 30 ips tape on a well
set
> up Ampex.

So Cal, how old are you anyway? Offhand I would say that only an
octogenarian would fail to hear the way that vinyl trashes the audio
off a good high speed analog tape.

> However the variety of source material on highbit hasn't
> been as great yet, and naysayers aren't helping the situation.

That has something to do with the fact that 16/44 can be sonically
transparent in practical use.
Anonymous
April 17, 2005 12:47:26 AM

Archived from groups: rec.audio.tech (More info?)

"post news" <thanks@for.info> wrote in message
news:jne7e.2964$O11.2375@fe12.lga...
> What's the story here?

One of the big differences is that the CD was based on sound mathematical
and engineering principles for sampled data systems, and SACD is based on a
rather bizarre internal data format and "marketecture". But marketing will
probably win out over good engineering, which is sad. If you're interested
in a highly technical discussion, try this: http://sjeng.org/ftp/SACD.pdf
Anonymous
April 17, 2005 12:47:27 AM

Archived from groups: rec.audio.tech (More info?)

Karl Uppiano wrote:
> "post news" <thanks@for.info> wrote in message
> news:jne7e.2964$O11.2375@fe12.lga...
>> What's the story here?
>
> One of the big differences is that the CD was based on sound
> mathematical and engineering principles for sampled data systems,
and
> SACD is based on a rather bizarre internal data format and
> "marketecture". But marketing will probably win out over good
> engineering, which is sad. If you're interested in a highly
technical
> discussion, try this: http://sjeng.org/ftp/SACD.pdf

Which is a posting of Audio Engineering Society Convention Paper 5395:

Why 1-Bit Sigma-Delta Conversion is Unsuitable for High-Quality
Applications by Stanley P. Lipshitz and John Vanderkooy Audio Research
Group, University of Waterloo Waterloo, Ontario N2L 3G1, Canada.

As I recall, AES discussions of this topic caused Sony to eventually
admit that their professional SACD encoders don't actually use DSD to
directly digitize analog signals for recording purposes.
Anonymous
April 17, 2005 12:47:27 AM

Archived from groups: rec.audio.tech (More info?)

>>Why just highbit digital signals, Cal?

Isn't it true that increasing the bitrate doesn't necessarily increase
precision, particularly if the increased bitrate comes from increased
sample rate and the signal being digitized is band-limted?<<

If the data were synchronous that would be so, but by definition
analog is async. Bicycle spokes and popsicle sticks and all that.

>>||> SACD and DVD-A have never impressed me as much as a few occasions
of
> listening to either first rate vinyl or 1/2" 30 ips tape on a well
set
> up Ampex.

So Cal, how old are you anyway? Offhand I would say that only an
octogenarian would fail to hear the way that vinyl trashes the audio
off a good high speed analog tape.<<

Younger than you, Arny.

Good analog tape is the gold standard, is it, Arny? Well, yes. But
really good vinyl comes moderately close. Maybe we should be arguing
for analog optical-sprocketless film on Ampex transports with
photocells instead of mag heads. But like turbine cars and Marilyn
singing anything Lennon-McCartney wrote, it wasn't to be and can't now.
Anonymous
April 17, 2005 12:47:28 AM

Archived from groups: rec.audio.tech (More info?)

calcerise@hotmail.com wrote:
>>> Why just highbit digital signals, Cal?
>
> Isn't it true that increasing the bitrate doesn't necessarily
increase
> precision, particularly if the increased bitrate comes from
increased
> sample rate and the signal being digitized is band-limted?<<

> If the data were synchronous that would be so, but by definition
> analog is async. Bicycle spokes and popsicle sticks and all that.

So you never heard of antialaising, eh Cal?

Thats what bicycle spokes and popsicle sticks and all that is -
aliasing. Since you've brought it up as an objection to digital,
you've again exposed how poorly-informed you are, Cal.

>>>>>> SACD and DVD-A have never impressed me as much as a few
occasions
> of
>> listening to either first rate vinyl or 1/2" 30 ips tape on a well
>> set up Ampex.

> So Cal, how old are you anyway? Offhand I would say that only an
> octogenarian would fail to hear the way that vinyl trashes the audio
> off a good high speed analog tape.<<

> Younger than you, Arny.

> Good analog tape is the gold standard, is it, Arny?

It is for vinyl that was made from it, which was my point.

> Well, yes.

Well no its not the gold standard for the here-and-now or even the
past 20 or more years. Analog tape isn't sonically transparent for
even just one generation. 44/16 digital can be sonically transparent
for lots of generations.

> But really good vinyl comes moderately close.

Like I suggested before, what's broke about your hearing? Good vinyl
doesn't even come close to good analog tape. It's ever further from
good digital.

> Maybe we should be arguing
> for analog optical-sprocketless film on Ampex transports with
> photocells instead of mag heads.

Maybe you should try to get back on topic, Cal.

> But like turbine cars and Marilyn
> singing anything Lennon-McCartney wrote, it wasn't to be and can't
now.

This trip through irrelevance-land was brought to you by Cal the
Luddite. Tune in next week for his next insult to your intelligence
and common sense.
Anonymous
April 17, 2005 1:49:59 AM

Archived from groups: rec.audio.tech (More info?)

"Arny Krueger" <arnyk@hotpop.com> wrote in message
news:kOudnZ1iua7rGfzfRVn-pQ@comcast.com...
> Karl Uppiano wrote:
>> "post news" <thanks@for.info> wrote in message
>> news:jne7e.2964$O11.2375@fe12.lga...
>>> What's the story here?
>>
>> One of the big differences is that the CD was based on sound
>> mathematical and engineering principles for sampled data systems,
> and
>> SACD is based on a rather bizarre internal data format and
>> "marketecture". But marketing will probably win out over good
>> engineering, which is sad. If you're interested in a highly
> technical
>> discussion, try this: http://sjeng.org/ftp/SACD.pdf
>
> Which is a posting of Audio Engineering Society Convention Paper 5395:
>
> Why 1-Bit Sigma-Delta Conversion is Unsuitable for High-Quality
> Applications by Stanley P. Lipshitz and John Vanderkooy Audio Research
> Group, University of Waterloo Waterloo, Ontario N2L 3G1, Canada.
>
> As I recall, AES discussions of this topic caused Sony to eventually
> admit that their professional SACD encoders don't actually use DSD to
> directly digitize analog signals for recording purposes.

Well, I can see why Sony would prefer a consumer format to be "in principle
imperfectible". It would make it impossible for consumers to make studio
quality duplicates of SACD recordings. Still, I see it as a big step in the
wrong direction. If I could see an elegant engineering design, I'd probably
be an early adopter. As it is, I'll be a grudging, reluctant adopter if and
when I can't find what I want to hear in any other format.
Anonymous
April 17, 2005 4:59:32 AM

Archived from groups: rec.audio.tech (More info?)

On Sat, 16 Apr 2005 17:32:42 -0400, "Arny Krueger" <arnyk@hotpop.com> wrote:

>Which is a posting of Audio Engineering Society Convention Paper 5395:

It's a preprint, not a paper.

BTW, it's a deeply revised version, correcting the numerous errors of
preprint 5188 presented at AES 109 (August 2000), which should never have
been accepted IMO.

>As I recall, AES discussions of this topic caused Sony to eventually
>admit that their professional SACD encoders don't actually use DSD to
>directly digitize analog signals for recording purposes.

You don't recall correctly: DSD Wide was already widely deployed and well
known in pro circles since 1999 or so. DSD Wide *is* DSD.

That's one of the reasons why Lipshitz and Vanderkooy rewrote their
presentation. They had stated in 5188 that: "In contrast [to DSD] multibit
sigma-delta converters, which output linear PCM code (here, multibit refers
to five or so bits in the converter), are in principle infinitely
perfectible". DSD Wide is an 8-bit form of sigma-delta.

BTW, Thorpe, Bental et al. presented "DSD-Wide. A Practical Implementation
for Professional Audio" at the same AES 110, preprint 5377. While Nuijten
and Reefman refuted most of Lipshitz and Vanderkooy's remaining arguments in
"Why Direct Stream Digital (DSD) is the best choice as a digital audio
format", 5396, where they showed that "1-bit DSD signals can be dithered
properly, so the resulting dithered DSD stream does not contain audible
artifacts in a band from 0-100~kHz".

OK, proper dither isn't perfect dither, but DSD works.
Anonymous
April 17, 2005 12:28:14 PM

Archived from groups: rec.audio.tech (More info?)

On Sun, 17 Apr 2005 00:59:32 +0200, Fran├žois Yves Le Gal
<flegal@aingeal.com> wrote:

>On Sat, 16 Apr 2005 17:32:42 -0400, "Arny Krueger" <arnyk@hotpop.com> wrote:
>
>>Which is a posting of Audio Engineering Society Convention Paper 5395:
>
>It's a preprint, not a paper.
>
>BTW, it's a deeply revised version, correcting the numerous errors of
>preprint 5188 presented at AES 109 (August 2000), which should never have
>been accepted IMO.
>
>>As I recall, AES discussions of this topic caused Sony to eventually
>>admit that their professional SACD encoders don't actually use DSD to
>>directly digitize analog signals for recording purposes.
>
>You don't recall correctly: DSD Wide was already widely deployed and well
>known in pro circles since 1999 or so. DSD Wide *is* DSD.

Please stop this continued *lying* about DSD Wide.

DSD Wide is a 64x oversampled 8-bit system, functionally identical to
PCM as implemented by dCS and others. The *whole* of DSD's claimed
technical advantage was rooted in it being *single bit*. Since DSD
Wide is *not* single-bit, then it shows *no* technical advantage over
conventional PCM. Indeed, it's more accurately described as PCM
Narrow, not DSD Wide.

>That's one of the reasons why Lipshitz and Vanderkooy rewrote their
>presentation. They had stated in 5188 that: "In contrast [to DSD] multibit
>sigma-delta converters, which output linear PCM code (here, multibit refers
>to five or so bits in the converter), are in principle infinitely
>perfectible". DSD Wide is an 8-bit form of sigma-delta.

Exactly - and therefore *nothing* to do with DSD proper, whatever the
marketing jerks decided to call it. The adoption of DSD Wide in the
recording chain completely destroys Sony's claims for SACD.

>BTW, Thorpe, Bental et al. presented "DSD-Wide. A Practical Implementation
>for Professional Audio" at the same AES 110, preprint 5377. While Nuijten
>and Reefman refuted most of Lipshitz and Vanderkooy's remaining arguments in
>"Why Direct Stream Digital (DSD) is the best choice as a digital audio
>format", 5396, where they showed that "1-bit DSD signals can be dithered
>properly, so the resulting dithered DSD stream does not contain audible
>artifacts in a band from 0-100~kHz".
>
>OK, proper dither isn't perfect dither, but DSD works.

Which is why Sony stopped using it? BWAHAHAHAHA!
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
April 17, 2005 12:28:15 PM

Archived from groups: rec.audio.tech (More info?)

On 16 Apr 2005 16:47:58 -0700, calcerise@hotmail.com wrote:

>
>>>Why just highbit digital signals, Cal?
>
>Isn't it true that increasing the bitrate doesn't necessarily increase
>precision, particularly if the increased bitrate comes from increased
>sample rate and the signal being digitized is band-limted?<<
>
> If the data were synchronous that would be so, but by definition
>analog is async. Bicycle spokes and popsicle sticks and all that.

What the hell are you drivelling about? Synchronicity has nothing to
do with it. Increasing the bitrate simply does *not* increase the
precision, given that the input bandwidth is limited to less than half
the base sampling rate.

If OTOH you were arguing for 'highbit' in the sense of bit depth, i.e.
24 rather than 16, then the same argument applies. Increasing the bit
depth does *not* inherently increase precision, given that the input
dynamic range does not exceed that of the base bit depth. In the case
of audio, we know that there are not *master* tapes with more than
80-85dB dynamic range, so 24-bit offers *no* advantage in precision or
'low level detail' over 16-bit.
--

Stewart Pinkerton | Music is Art - Audio is Engineering
Anonymous
April 17, 2005 5:57:45 PM

Archived from groups: rec.audio.tech (More info?)

On Sun, 17 Apr 2005 08:28:14 +0000 (UTC), Stewart Pinkerton
<patent3@dircon.co.uk> wrote:

>Please stop this continued *lying* about DSD Wide.
>
>DSD Wide is a 64x oversampled 8-bit system

What did I write? That "DSD Wide is an 8-bit form of sigma-delta."

Thank you for confirming that:
i. I was right,
ii. you're already drunk.

>The *whole* of DSD's claimed
>technical advantage was rooted in it being *single bit*.

No. As anyone familiar with DSD can attest, the single bit part is only
relevant for consumer applications.

>Since DSD
>Wide is *not* single-bit, then it shows *no* technical advantage over
>conventional PCM.

Of course it does: DSD Wide decimates very gracefully to either PCM or DSD.
High resolution PCM decimates very gracefully to PCM.

>>OK, proper dither isn't perfect dither, but DSD works.
>
>Which is why Sony stopped using it? BWAHAHAHAHA!

Sony hasn't stopped using DSD and has just launched new consumer players as
well as some very nice pro stuff thru their Oxford subsidiary... Check the
new Hypermac DSD transport, with 384 channels (!) thru a single Gigabit
Ethernet link!
Anonymous
April 17, 2005 8:50:02 PM

Archived from groups: rec.audio.tech (More info?)

PMFJI but the elephant in the middle of the room that everyone seems to be
dancing around is this: (1) There is the way you do analog-to-digital
conversion, and (2) there is the way you STORE what you converted. Now,
there are lots of 1-bit converters out there. The idea is that doing a huge
number of 1-bit conversions at some huge multiple of the sampling frequency
avoids certain electronic problems. I have no opinions on that, and if I
did, it wouldn't matter as I'm sure this has been discussing more times than
global warming and 9/11 put together. But it's all irrelevant to (2). Your
storage method need not be bound by how you did the conversion. Once you've
got a certain digital stream, then your problem is how to preserve its
information completely, in the most efficient possible format.

And what I don't understand is how any delta-modulation scheme, which is
basically what SACD is, wherever it came from, could store information more
efficiently than a PCM scheme having the same bit rate. We covered delta
modulation, and ADPCM, and all that in engineering school. They have some
uses, mainly I think as a quick-and-dirty scheme for certain
telephony-related, lo-fi purposes. In about 20 years of watching the digital
audio scene, I never heard anyone propose such a thing to replace PCM, until
now, when the CD is a mature technology and somebody desperately needed an
"improvement" to milk more money from consumers.

That said, no doubt I'll eventually own some SACD discs, and a player that
can handle them. That's just because there'll be remastered versions of some
favorite artists, and I'll want the new version just because, I will hope,
the mastering will have been done better. This kind of dilemma occurs now
and then, and certainly not just in audio.
Anonymous
April 17, 2005 8:50:03 PM

Archived from groups: rec.audio.tech (More info?)

RWG wrote:

> And what I don't understand is how any delta-modulation scheme,
which
> is basically what SACD is, wherever it came from, could store
> information more efficiently than a PCM scheme having the same bit
> rate.

Agreed, there is this slight matter of information theory.

SACD decreases the information content of the data it transfers by
means of frequency-dependent dynamic range. The dynamic range drops
precipitously above 20 KHz, which dramatically decreases the
information content.

> In about 20 years of watching the digital audio scene, I never heard
> anyone propose such a thing to replace PCM, until now, when the CD
is
> a mature technology and somebody desperately needed an "improvement"
> to milk more money from consumers.

The same kind of frequency-dependent dynamic range management works
with PCM, with similar effects on required bitrates.

> That said, no doubt I'll eventually own some SACD discs, and a
player
> that can handle them. That's just because there'll be remastered
> versions of some favorite artists, and I'll want the new version
just
> because, I will hope, the mastering will have been done better. This
> kind of dilemma occurs now and then, and certainly not just in
audio.

I was looking for a better-than-average optical player earleir this
year, and ended up with a universal model that handles both SACD and
DVD-A discs. I think it cost me about $30 more than the cheapest
players from the same manufacturer.
Anonymous
April 18, 2005 5:53:15 PM

Archived from groups: rec.audio.tech (More info?)

"Karl Uppiano" <karl.uppiano@verizon.net> wrote in message
news:yRe8e.13896$Zn3.4537@trnddc02...
> and SACD is based on a
> rather bizarre internal data format and "marketecture".

Nice terminology :-)

>But marketing will
> probably win out over good engineering, which is sad.

It often does, but fortunately in this case I doubt it.

MrT.
Anonymous
April 18, 2005 5:53:16 PM

Archived from groups: rec.audio.tech (More info?)

>>But marketing will
>> probably win out over good engineering, which is sad.
>
> It often does, but fortunately in this case I doubt it.

I certainly hope you're right.
Anonymous
April 18, 2005 6:06:53 PM

Archived from groups: rec.audio.tech (More info?)

"Arny Krueger" <arnyk@hotpop.com> wrote in message
news:guqdnawMIbakIv_fRVn-2A@comcast.com...
> I was looking for a better-than-average optical player earleir this
> year, and ended up with a universal model that handles both SACD and
> DVD-A discs. I think it cost me about $30 more than the cheapest
> players from the same manufacturer.

Which is as it should be.

MrT.
!