Discussion:
DSD vs PCM Explanation & Comparison
(too old to reply)
Audy
2003-10-17 23:25:57 UTC
Permalink
Here is an excellent article written by Barry Henderson, President of
iZ Technology Corp. who is the wizard behind the Radar 24 system. This
is in response to a question about whether iZ Corp. plans to release a
DSD recorder in the future:

Audy
There has been alot of discussion here at iZ lately about putting out a DSD
machine. We even had a visit from Ayataki Nishio of Sony Japan, the man who
invented DSD. We invited him to come to Vancouver a few months ago to meet
with us and discuss the technical aspects of building a DSD machine. We may
build one if we could get around some of the problems he pointed out to us,
namely;
1) Advanced editing is not possible. The Pyramix system first converts the DSD
files to PCM and then does the editing. If there is an advantage to DSD
recording, then you may have lost it in the conversion.
2) Mixing is not possible. According to Sony, it is not ever possible to have
a DSD mixer. AMS just released a digital mixer with DSD I/O. However the DSD
immediately converts the DSD stream to PCM for internal use in the mixer and
then converts it back again. We are concerned with having so many conversions
that you may be introducing many artifacts into the audio.
3) The sample rate is not high enough. DSD is in its infant stage of
development. The current "State of the Art" is 1 bit x 64fs (means 64 x 44.1
kHz sample rate) which is a data rate of 2.822 megabits per second per
channel. RADAR currently has a maximum resolution of 24 bits x 216 kHz sample
rate = 5.184 megabits per second per channel. The 64fs limitation is a Sony
thing cause you need to use special Sony chips to make the recorder work which
brings us to our 4th problem...
4) Development is very Sony dependant.
Like it or not, its a PCM world and until that changes there will be very
little demand for DSD. So it is hard to justify an R&D budget to do alot of
very complex development only to still have significant problems remaining and
a very small market.
That said, we are looking at PCM to DSD file conversion. In other words we
would record in RADAR PCM at 5 megabits per second in order to capture as many
nuances of the audio as possible and then down sample it to the lower 2.822
megabit rate and save the files on another hard disk or even burn a DVD.
Barry Henderson, President
iZ Technology Corp.
Kurt Albershardt
2003-10-18 01:57:16 UTC
Permalink
1) Advanced editing is not possible. The Pyramix system first converts the DSD
files to PCM and then does the editing. If there is an advantage to DSD
recording, then you may have lost it in the conversion.
Open to debate, since Pyramix maintains the 2.8 mbit data rate of the
DSD stream and doesn't mangle its gentle filters.
Arny Krueger
2003-10-18 09:08:54 UTC
Permalink
Post by Kurt Albershardt
1) Advanced editing is not possible. The Pyramix system first
converts the DSD files to PCM and then does the editing. If there
is an advantage to DSD recording, then you may have lost it in the
conversion.
Open to debate, since Pyramix maintains the 2.8 mbit data rate of the
DSD stream and doesn't mangle its gentle filters.
I think we have to consider the mind set that the statement: "If there is an
advantage to DSD recording, then you may have lost it in the conversion."
refers to.

My understanding is that some people believe that there are some
undefinable, immeasurable, currently unquantifiable *something(s)* that make
DSD recordings sound better than traditional PCM with a similar or greater
bitrate.

I've certainly run into these people on Usenet and seen articles in other
media that seem to say to the same basic thing(s).

Since "gentle filters" actually exist, can be analyzed, quantified, and
found to disagree with this belief about the alleged superior sound of DSD,
it *can't* be what these people are talking about.

BTW, one context in which I find these kinds of beliefs about the
superiority of DSD is PCABX. I've suggested to DSD advocates that the output
of a SACD player be digitized at 24/192, downsampled to 16/44, and the two
be compared.

The response is that re-quantizing DSD into any form of PCM destroys a
significant sound quality advantage of DSD.

Ironically, some of these same people go on for zillions of posts on Usenet
about the improved sound quality of DSD recordings that are now known to
have been made from 48 KHz PCM recordings but...
William Sommerwerck
2003-10-18 10:36:07 UTC
Permalink
Post by Arny Krueger
I think we have to consider the mind set that the statement: "If there is an
advantage to DSD recording, then you may have lost it in the conversion."
refers to.
My understanding is that some people believe that there are some
undefinable, immeasurable, currently unquantifiable *something(s)* that
make DSD recordings sound better than traditional PCM with a similar
or greater bitrate.
I reviewed the dbx 700 for Stereophile about 20 years ago. This unit used a
delta-modulation system similar to DSD. I made at least one live recording with
the mics feeding both the dbx 700 and my Nakamichi DMP-100 (a modified Sony
PCM-F1).

There was no question that dbx was superior to the Nakamichi, particularly in
regard to high-frequency distortion. Brass instruments, in particular, were
cleaner.

Of course, this was a comparison between two particular products, both designed
in an era when digital recording was undergoing significant refinement. It
proves nothing about the ultimate subjective quality of either PCM or DSD.
However, I've always had a prejudice toward bit-stream systems, simply because
they're so crude.

I bought a high-end Sony multi-channel SACD player last Christmas, and was
pleasantly surprised to discover that SACDs were noticeably less hard-sounding
and "colored" than CDs. "Unfortunately," my existing CDs showed a similar sonic
improvement. So, if there really is a significant difference between SACD and
CD, it might be due to nothing more than improved conversion.

"One of these days" I'm going to get around to comparing the Sony's CD output to
my Kinergetics DAC. The SACD player has an S/PDIF digital output for CDs, so
it's possible to make instantaneous comparisons.
Rob Adelman
2003-10-18 14:19:38 UTC
Permalink
Post by Arny Krueger
Ironically, some of these same people go on for zillions of posts on Usenet
about the improved sound quality of DSD recordings that are now known to
have been made from 48 KHz PCM recordings but...
Well the first step in forming your own opinion might be to ACTUALLY
LISTEN TO ONE!
Arny Krueger
2003-10-18 14:57:08 UTC
Permalink
Post by Rob Adelman
Post by Arny Krueger
Ironically, some of these same people go on for zillions of posts on
Usenet about the improved sound quality of DSD recordings that are
now known to have been made from 48 KHz PCM recordings but...
Well the first step in forming your own opinion might be to ACTUALLY
LISTEN TO ONE!
Please post a 24/96 or 24/192 .wav file of a track from your favorite SACD
at your earliest convenience.
William Sommerwerck
2003-10-18 15:17:41 UTC
Permalink
Post by Arny Krueger
Post by Rob Adelman
Well the first step in forming your own opinion might be to
ACTUALLY LISTEN TO ONE!
Please post a 24/96 or 24/192 .wav file of a track from your
favorite SACD at your earliest convenience.
I can't (always) read your mind and determine your intent, but...

How can we take _anything_ you say seriously when you offer such a response...?
Rob Adelman
2003-10-18 15:24:35 UTC
Permalink
Post by William Sommerwerck
I can't (always) read your mind and determine your intent, but...
How can we take _anything_ you say seriously when you offer such a response...?
Arny is afraid to listen to the format for fear he might actually agree
that it does sound better. Easier to just avoid the possibility.

b.t.w. a good player can now be had for 150 bucks!
<http://store.acousticsounds.com/store.cfm?Title_ID=12119&do=detail>
Arny Krueger
2003-10-18 15:32:37 UTC
Permalink
Post by Rob Adelman
Post by William Sommerwerck
I can't (always) read your mind and determine your intent, but...
How can we take _anything_ you say seriously when you offer such a response...?
Arny is afraid to listen to the format for fear he might actually
agree that it does sound better. Easier to just avoid the possibility.
b.t.w. a good player can now be had for 150 bucks!
<http://store.acousticsounds.com/store.cfm?Title_ID=12119&do=detail>
Great, send one to me and a few disks of your choosing. If this expense is
actually that little of a hassle, why not do so?
Harry Lavo
2003-10-18 17:48:54 UTC
Permalink
About a year ago on RAHE Arnie was holding forth about the price of SACD
players being prohibitive, when challenged as to why he didn't just buy one
and listen. He then said none were available under $200. When that price
barrier was broken it was "under $150".

All he ever does is propose we convert DSD to PCM and then be surprised when
the signal gives a "no difference" under his abx testing to the same signal
using the same PCM technology. It is an insult to our intelligence,
frankly.
Post by Rob Adelman
Post by William Sommerwerck
I can't (always) read your mind and determine your intent, but...
How can we take _anything_ you say seriously when you offer such a response...?
Arny is afraid to listen to the format for fear he might actually agree
that it does sound better. Easier to just avoid the possibility.
b.t.w. a good player can now be had for 150 bucks!
<http://store.acousticsounds.com/store.cfm?Title_ID=12119&do=detail>
Arny Krueger
2003-10-18 19:08:03 UTC
Permalink
Post by Harry Lavo
About a year ago on RAHE Arnie was holding forth about the price of
SACD players being prohibitive, when challenged as to why he didn't
just buy one and listen. He then said none were available under
$200. When that price barrier was broken it was "under $150".
Harry, given your well-knokwn fascination with audio-related fiction, why
don't you prove that claim with a quote from google?
Post by Harry Lavo
All he ever does is propose we convert DSD to PCM and then be
surprised when the signal gives a "no difference" under his abx
testing to the same signal using the same PCM technology. It is an
insult to our intelligence, frankly.
I don't believe I've ever insulted your intelligence, Harry. Or your
veracity for that part.

;-)
Arny Krueger
2003-10-18 15:31:07 UTC
Permalink
Post by William Sommerwerck
Post by Arny Krueger
Post by Rob Adelman
Well the first step in forming your own opinion might be to
ACTUALLY LISTEN TO ONE!
Please post a 24/96 or 24/192 .wav file of a track from your
favorite SACD at your earliest convenience.
I can't (always) read your mind and determine your intent, but...
How can we take _anything_ you say seriously when you offer such a response...?
Just put me down as a believer in science. I know that in some circles that
makes me an object of hostility and derision, but I'm used to it by now.
William Sommerwerck
2003-10-18 16:46:21 UTC
Permalink
Arny Krueger wrote...
William Sommerwerck wrote...
Post by William Sommerwerck
Post by Arny Krueger
Post by Rob Adelman
Well the first step in forming your own opinion might be to
ACTUALLY LISTEN TO ONE!
Please post a 24/96 or 24/192 .wav file of a track from your
favorite SACD at your earliest convenience.
I can't (always) read your mind and determine your intent, but...
How can we take _anything_ you say seriously when you offer
such a response...?
Just put me down as a believer in science. I know that in some circles that
makes me an object of hostility and derision, but I'm used to it by now.
I am reminded of a demo ReVox used to stage at audio shows. They had a line of
five (I think) A-77s with the output of one feeding the input of the next. They
claimed the output of the fifth was indistinguishable from the input to the
first.

We both know that, given a high-quality program source, a _single_ ReVox would
not pass such a test. I've never heard a piece of electronics that was 100%
transparent (though some come awfully close), and I doubt anyone has ever built
a recorder (analog or digital) that is subjectively perfect. Do you know of any?

Given the differences between PCM and DSD, it would not be surprising if one
were more-accurate than the other -- or that they at least had different
errors/colorations.
Scott Dorsey
2003-10-18 16:55:15 UTC
Permalink
Post by William Sommerwerck
I am reminded of a demo ReVox used to stage at audio shows. They had a line of
five (I think) A-77s with the output of one feeding the input of the next. They
claimed the output of the fifth was indistinguishable from the input to the
first.
I bet through a pair of Auratones, it really was indistinguishable.
Post by William Sommerwerck
We both know that, given a high-quality program source, a _single_ ReVox would
not pass such a test. I've never heard a piece of electronics that was 100%
transparent (though some come awfully close), and I doubt anyone has ever built
a recorder (analog or digital) that is subjectively perfect. Do you know of any?
No, but the thing is that today, the quality of the recorders has got to
the point where it is far higher than the quality of the people behind them
in most cases. Recorders aren't transparent at all, but they have long ago
ceased to be the weak link in the chain.
Post by William Sommerwerck
Given the differences between PCM and DSD, it would not be surprising if one
were more-accurate than the other -- or that they at least had different
errors/colorations.
Absolutely. The question is whether these errors and colorations are really
all that significant in light of the more severe errors present in speakers
and microphones. I can't answer this, either, and I don't think anyone really
can yet.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Arny Krueger
2003-10-18 17:05:48 UTC
Permalink
Post by William Sommerwerck
Arny Krueger wrote...
William Sommerwerck wrote...
Post by William Sommerwerck
Post by Arny Krueger
Post by Rob Adelman
Well the first step in forming your own opinion might be to
ACTUALLY LISTEN TO ONE!
Please post a 24/96 or 24/192 .wav file of a track from your
favorite SACD at your earliest convenience.
I can't (always) read your mind and determine your intent, but...
How can we take _anything_ you say seriously when you offer
such a response...?
Just put me down as a believer in science. I know that in some
circles that makes me an object of hostility and derision, but I'm
used to it by now.
I am reminded of a demo ReVox used to stage at audio shows. They had
a line of five (I think) A-77s with the output of one feeding the
input of the next. They claimed the output of the fifth was
indistinguishable from the input to the first.
This might actually work if you pick the music *right*. A somewhat distant
recording of say a piano... But if one runs the stuff I use, detection at
one generation would probably be a cakewalk. For a more interesting test,
look here: http://www.pcavtech.com/abx/abx_tapg.htm
Post by William Sommerwerck
We both know that, given a high-quality program source, a _single_
ReVox would not pass such a test.
Been there, done that, I had an A77 for years.
Post by William Sommerwerck
I've never heard a piece of
electronics that was 100% transparent (though some come awfully
close),
I've heard a great many of them, especially lately.
Post by William Sommerwerck
and I doubt anyone has ever built a recorder (analog or
digital) that is subjectively perfect. Do you know of any?
http://www.pcabx.com/product/cardd_deluxe/index.htm

Tell me when you start hearing differences in a blind test - 1, 5, 10, 20
cycles?

This one is a little easier:

http://www.pcabx.com/product/santa_cruz/index.htm

One generation is hard, the rest shouldn't be a problem.

This one is easier still:

http://www.pcabx.com/product/ct4830/index.htm

One generation is easy, the rest are almost so easy its an insult.

And then:

http://www.pcabx.com/product/layla24/index.htm

It's a broken card as it was built, so it's very easy.

If you actually want to try these tests out and not just talk about them,
start here and work towards the top. When you hit the top, you'll be glad
you started here.
Post by William Sommerwerck
Given the differences between PCM and DSD, it would not be surprising
if one were more-accurate than the other -- or that they at least had
different errors/colorations.
Given how they measure, not a freakin' chance. There's a reason why all the
record companies are remastering the DSD layers - that's the only way they
can bet on people hearing a difference.
Rob Adelman
2003-10-18 17:45:29 UTC
Permalink
Post by Arny Krueger
Given how they measure, not a freakin' chance. There's a reason why all the
record companies are remastering the DSD layers - that's the only way they
can bet on people hearing a difference.
In fact you are quite wrong. The vast majority of the SACD's I bought
are not remastered at all. The music sounds exactly the same as it has
since I first heard it on vinyl, 10, 20, 30 years ago without the added
noise. I still have most of the vinyl too. The couple of SACD's I have
(one I sent back) that WERE obviously remastered sound like ass. DSD
seems to be a clear reproduction of whatever you give it. So if you have
an old analog master that sounds stunning, and LEAVE IT ALONE, that is
what you hear. If you mess with it, the changes you make become even
more obvious.

The lesser systems, pcm, and especially MP3 kind of muddy the waters a
little, so some of this stuff is less obvious, you can actually get away
with a lot more futzing.
Arny Krueger
2003-10-18 19:19:27 UTC
Permalink
Post by Rob Adelman
Post by Arny Krueger
Given how they measure, not a freakin' chance. There's a reason why
all the record companies are remastering the DSD layers - that's the
only way they can bet on people hearing a difference.
In fact you are quite wrong. The vast majority of the SACD's I bought
are not remastered at all. The music sounds exactly the same as it has
since I first heard it on vinyl, 10, 20, 30 years ago without the
added noise. I still have most of the vinyl too.
I strongly suspect that if I said the same thing about SACD and CD Rob,
you'd probably condemn my hearing, my choice of player, and/or the rest of
my audio system.
Post by Rob Adelman
The couple of SACD's
I have (one I sent back) that WERE obviously remastered sound like
ass.
OK so what are you saying here, there's some magic something that makes
*all* remastered SACDs sound like "ass"? I can't believe that the guys who
remastered these things blew it 100% of the time.
Post by Rob Adelman
DSD seems to be a clear reproduction of whatever you give it.
Of course I'm bound to agree with that. See the paragraph that I wrote and
you quoted, Rob. That's what I meant - DSD can be reasonably be expected to
provide sonically transparent reproduction. However, sonically transparent
reproduction is happily NOT a feature that DSD ALONE can rightfully claim.
Post by Rob Adelman
So if you have an old analog master that sounds stunning, and LEAVE IT
ALONE, that is what you hear. If you mess with it, the changes you
make become even more obvious.
Thing is, no way is LP record/playback *anything* like sonically
transparent. I'm still waiting for *someone* to push my contribution to RAP
5 through the vinyl mill, and bring back an indistinguishable reproduction
of it. Interestingly enough, the world is full of $59 CD Burners, <$200 CD
players and $39 CD ROM drives that can do this *trick* without a bit of
trouble.
Post by Rob Adelman
The lesser systems, pcm, and especially MP3 kind of muddy the waters a
little, so some of this stuff is less obvious, you can actually get
away with a lot more futzing.
How did MP3 creep into this discussion? Isn't the iTunes discussion in
another thread?
Rob Adelman
2003-10-18 19:30:02 UTC
Permalink
Post by Arny Krueger
I strongly suspect that if I said the same thing about SACD and CD Rob,
you'd probably condemn my hearing, my choice of player, and/or the rest of
my audio system.
No i wouldn't. I would say, ok there is someone who disagrees. But
currently, I am still waiting for ONE person who has heard a comparable
recording on both to say the DSD does not sound better.
Post by Arny Krueger
Post by Rob Adelman
The couple of SACD's
I have (one I sent back) that WERE obviously remastered sound like
ass.
OK so what are you saying here, there's some magic something that makes
*all* remastered SACDs sound like "ass"? I can't believe that the guys who
remastered these things blew it 100% of the time.
Well I have only run into two, so in my case 100% of the time was easy..
Brian Takei
2003-10-18 21:49:58 UTC
Permalink
Post by Rob Adelman
Post by Arny Krueger
I strongly suspect that if I said the same thing about SACD and CD Rob,
you'd probably condemn my hearing, my choice of player, and/or the rest of
my audio system.
No i wouldn't. I would say, ok there is someone who disagrees. But
currently, I am still waiting for ONE person who has heard a comparable
recording on both to say the DSD does not sound better.
"comparable" being a key term. Correct me if I'm wrong, but it seems
paramount to ask -- what can be objectively proven by a given listening
test, particularly with regard to a hypothesis that one _format_ is
superior to another?

Elsewhere in this thread, Scott made a point on this issue (that has
been made in the past), but while Rob is going back and forth with Arny,
a response to Scott is conspicuously absent [reference article
bmrm8u$1s1$***@panix2.panix.com]. I'd really like to see this taken head
on.

POINT: If one doesn't _know_ the detailed production/mastering details
of two 'comparables', one is significantly limited in the intelligent
conclusions one can draw from a comparison.

QUESTIONS:
1) Is there any disagreement on this?
2) Has anyone done any significant PCM vs. DSD testing, with complete
and verified knowledge of the production/mastering details of the
subjects used for the testing?


- Brian
Arny Krueger
2003-10-18 22:43:59 UTC
Permalink
"Rob Adelman" <***@mn.rr.com> wrote in message news:bmru7o$qfjjg$***@ID-75267.news.uni-berlin.de

Rob, since you're not responding to about 80% of my posts, I thought that a
little turnabout is fair play. I'm going to bring up a point that you've
been dodging for months, and one that you just dropped like a hot potato,
again.

In the past you've claimed to have a Delta 1010 which is IME entirely
adequate for performing the PCABX listening tests that I've just recommended
in a post you responded to.

Your out-of-pocket cost for PCABX listening tests appears to be $0.00.

In contrast you've been complaining that I 'm unwilling to spend $150 for a
SACD player plus per-unit costs for SACD discs to listen to.

Don't you think that your complaints ring mighty darn hollow, all things
considered?

<Sorry folks for the pile of crap that will inevitably fall on RAP quite
shortly!>
Rob Adelman
2003-10-18 23:12:05 UTC
Permalink
Post by Arny Krueger
Rob, since you're not responding to about 80% of my posts, I thought that a
little turnabout is fair play. I'm going to bring up a point that you've
been dodging for months, and one that you just dropped like a hot potato,
again.
I am suppose to learn something from some crazy test you devised?
Post by Arny Krueger
In the past you've claimed to have a Delta 1010 which is IME entirely
adequate for performing the PCABX listening tests that I've just recommended
in a post you responded to.
Haven't had the Delta 1010 for years. Guess you must have missed that.
Post by Arny Krueger
In contrast you've been complaining that I 'm unwilling to spend $150 for a
SACD player plus per-unit costs for SACD discs to listen to.
What don't you go to a store? Or perhaps you know somebody with one?
Arny Krueger
2003-10-18 23:55:39 UTC
Permalink
Post by Rob Adelman
Post by Arny Krueger
Rob, since you're not responding to about 80% of my posts, I thought
that a little turnabout is fair play. I'm going to bring up a point
that you've been dodging for months, and one that you just dropped
like a hot potato, again.
I am suppose to learn something from some crazy test you devised?
Due to your total absence of good faith, I won't bother with your test,
Adelman.
Rob Adelman
2003-10-19 00:03:20 UTC
Permalink
Post by Arny Krueger
Due to your total absence of good faith, I won't bother with your test,
Adelman.
Well geez, I am heartbroken...

Only problem is mine is not a "test". Your loss my friend.
Arny Krueger
2003-10-19 09:46:02 UTC
Permalink
Post by Rob Adelman
Post by Arny Krueger
Due to your total absence of good faith, I won't bother with your
test, Adelman.
Well geez, I am heartbroken...
I'd worry more about the damage you're doing to your own credibility, Rob.

I was trying to be nice so I didn't mention you specifically in my post that
said:

"Ironically, some of these same people go on for zillions of posts on Usenet
about the improved sound quality of DSD recordings that are now known to
have been made from 48 KHz PCM recordings but..."

But somehow you jumped right up and started screaming quite defensively.

Then you made the wild claim that these SACD discs weren't remastered.
You're supposed to know about stuff like this, Rob. Of course the discs were
remastered. Different format, different masters! In many cases the
manufacturers advertise that these SACDs were remastered!

BTW, I'm not making this stuff up about SACDs being made from
bandpass-limited recordings. It has been reported by no less than David
Greisinger of Lexicon:

http://world.std.com/~griesngr/intermod.ppt

This presentation was made to the AES International Convention in Baniff
last July. Dismiss at your own risk!
Post by Rob Adelman
Only problem is mine is not a "test".
Good example of the kind of etymological hair-splitting that ruins one's
reputation on a forum like this.
Post by Rob Adelman
Your loss my friend.
It seems Rob like you don't really think I'm your friend. Or, is this how
you treat your friends in real life?

But Rob you DID do me a big favor. I said that:

"My understanding is that some people believe that there are some
undefinable, immeasurable, currently unquantifiable *something(s)* that make
DSD recordings sound better than traditional PCM with a similar or greater
bitrate."

You jumped right up and gave me a living, breathing example, complete with
self-contradictions, denial of obvious well-known facts, and general
bad-faith behavior.

Thank you. It made my day...

;-)
Rob Adelman
2003-10-19 15:19:51 UTC
Permalink
Post by Arny Krueger
I'd worry more about the damage you're doing to your own credibility, Rob.
My credibility? Whatever. It is what it is. All I ever said is that the
only important "test" is to use your ears and decide what sounds good.
For anybody interested, all I have asked is to check out the format and
let us know what you think. As you convienently keep avoiding, the fact
is that anybody who has heard it seems to conclude that it does sound
better. You can come up with a million reasons as to why they might have
concluded this, but the fact is that they have.
Arny Krueger
2003-10-20 09:13:39 UTC
Permalink
Post by Rob Adelman
All I ever said is that
the only important "test" is to use your ears and decide what sounds
good.
On that we agree, except that in practice you don't use just your ears and I
do.
Post by Rob Adelman
For anybody interested, all I have asked is to check out the
format and let us know what you think.
For the record it will cost me several $100's to do that, because none of my
audio friends in the area (100's of people) have made the prerequisite
expenditures, either. In contrast Rob, you won't listen to the stuff I asked
you to check out for free. How much bad faith Rob, does it take to convince
anybody that you aren't credible?
Post by Rob Adelman
As you convienently keep avoiding, the fact is that anybody who has heard
it seems to conclude
Post by Rob Adelman
that it does sound better.
I've provided evidence from an AES conference that says otherwise and you
keep deleting the URL and ignoring its contents. How much bad faith Rob,
does it take to convince anybody that you aren't credible?
Post by Rob Adelman
You can come up with a million reasons as
to why they might have concluded this, but the fact is that they have.
It's a fact that millions of people have bought Monster Cable and at least
initially concluded that it sounds better. Thus, this argument lends zero to
your credibility.
Rob Adelman
2003-10-20 13:25:47 UTC
Permalink
Post by Arny Krueger
On that we agree, except that in practice you don't use just your ears and I
do.
And people wonder why you always seem to be the center of so many
flamefests. Statements like that make it pretty obvious.
LeBaron & Alrich
2003-10-20 14:42:11 UTC
Permalink
Post by Rob Adelman
Post by Arny Krueger
On that we agree, except that in practice you don't use just your ears and I
do.
And people wonder why you always seem to be the center of so many
flamefests. Statements like that make it pretty obvious.
Arny might stir his share of shit here, but he's pointing out that your
listening "tests" lack the essential procedures that would make them
_tests_, and that you can put your own ears to a test for free.

That is not flamefest material except to someone who has made grand
claims about the sounds of things and _why_ things sound the way they do
without bothering to remove obvious sources of bias. If you want to wear
a fitting shoe why blame it on Arny?

--
ha
Rob Adelman
2003-10-20 14:56:19 UTC
Permalink
Some reason you need to keep sticking your nose in here?

Arny claims I don't use my ears to make audio decisions which could not
possibly be further from the truth.
Post by LeBaron & Alrich
Post by Rob Adelman
Post by Arny Krueger
On that we agree, except that in practice you don't use just your ears and I
do.
And people wonder why you always seem to be the center of so many
flamefests. Statements like that make it pretty obvious.
Arny might stir his share of shit here, but he's pointing out that your
listening "tests" lack the essential procedures that would make them
_tests_, and that you can put your own ears to a test for free.
That is not flamefest material except to someone who has made grand
claims about the sounds of things and _why_ things sound the way they do
without bothering to remove obvious sources of bias. If you want to wear
a fitting shoe why blame it on Arny?
--
ha
Wayne
2003-10-20 16:15:11 UTC
Permalink
Come on guys. Is this another my dog's bigger than you're dog thingy? If you
were as sensitive sexually, you'd have everything in hand all the time.
Post by Rob Adelman
Some reason you need to keep sticking your nose in here?
Arny claims I don't use my ears to make audio decisions which could not
possibly be further from the truth.
Post by LeBaron & Alrich
Post by Rob Adelman
Post by Arny Krueger
On that we agree, except that in practice you don't use just your ears and
I
Post by LeBaron & Alrich
Post by Rob Adelman
Post by Arny Krueger
do.
And people wonder why you always seem to be the center of so many
flamefests. Statements like that make it pretty obvious.
Arny might stir his share of shit here, but he's pointing out that your
listening "tests" lack the essential procedures that would make them
_tests_, and that you can put your own ears to a test for free.
That is not flamefest material except to someone who has made grand
claims about the sounds of things and _why_ things sound the way they do
without bothering to remove obvious sources of bias. If you want to wear
a fitting shoe why blame it on Arny?
--
ha
--Wayne

-"sounded good to me"-
Arny Krueger
2003-10-20 17:32:49 UTC
Permalink
Post by Rob Adelman
Some reason you need to keep sticking your nose in here?
I suspect he's trying to express his opinion. Imagine that!

;-)
Post by Rob Adelman
Arny claims I don't use my ears to make audio decisions which could
not possibly be further from the truth.
Well Rob, just like you *lost* the entire Greisinger AES paper two, maybe
three times, you also managed to *lose* the word "just" when you spewed out
the previous paragraph.

Just for the record, here's what was *really* said:

"Rob Adelman" <***@mn.rr.com> wrote in message news:bmua2m$r4lb1$***@ID-75267.news.uni-berlin.de

"All I ever said is that the only important "test" is to use your ears and
decide what sounds good."

"Arny Krueger" <***@hotpop.com> wrote in message news:vOSdnSWHVMXaOg6iRVn-***@comcast.com

"On that we agree, except that in practice you don't use just your ears and
I do."

Notice the phrase: "...*just* your ears...".

Cutting to the chase Rob, you run like hell every time I point you at
bias-controlled listening tests. Or as Hank put it more nicely:

"LeBaron & Alrich" <***@thegrid.net> wrote in message news:1g3485t.bdz63rnck7ggN%***@thegrid.net

"Arny might stir his share of shit here, but he's pointing out that your
listening "tests" lack the essential procedures that would make them
_tests_, and that you can put your own ears to a test for free."

"That is not flamefest material except to someone who has made grand claims
about the sounds of things and _why_ things sound the way they do without
bothering to remove obvious sources of bias. If you want to wear a fitting
shoe why blame it on Arny?"

FWIW, I think that just about every SACD layer of a legacy recording would
sound different from any previously-released CD in a level-matched,
time-synched, bias-controlled listening test. That's because both common
engineering sense and the people who are selling this stuff say it was
remastered.

Which is the gist of what Scott was trying to gently but quite clearly
convey with this post:

"Scott Dorsey" <***@panix.com> wrote in message news:bmrm8u$1s1$***@panix2.panix.com

"I have LPs that sound better than the CD, and CDs that sound better than
the LPs. The guys in the mastering room have a lot more to do with the
sound quality than the release format. This isn't going to change."

Rob, you've already denied, deleted and dissembled your way around each of
these critical points several times in this thread, but I thought it would
be neat to assemble them all in one place so you could give us one more
memorable performance.

;-)
Rob Adelman
2003-10-20 18:17:53 UTC
Permalink
Post by Arny Krueger
Rob, you've already denied, deleted and dissembled your way around each of
these critical points several times in this thread, but I thought it would
be neat to assemble them all in one place so you could give us one more
memorable performance.
;-)
Nah, you can just remember that I think your an asshole.
hollywood_steve
2003-10-20 19:16:05 UTC
Permalink
Post by Rob Adelman
For anybody interested, all I have asked is to check out the
format and let us know what you think.
For the record it will cost me several $100's to do that, because none
of my
audio friends in the area (100's of people) have made the prerequisite
expenditures, either.
***********************************

Somebody in your corner of the world must have a Genex Gx8500, GX9000
or GX9048 recorder. (the local mega-studio, or a rental house?)Ask to
hear a recording made in DSD format (not an old PCM recording that has
been converted). Then you can decide for yourself.

***@pacbell.net
Jay Kadis
2003-10-20 19:22:00 UTC
Permalink
Post by Arny Krueger
Post by Rob Adelman
For anybody interested, all I have asked is to check out the
format and let us know what you think.
For the record it will cost me several $100's to do that, because none
of my
audio friends in the area (100's of people) have made the prerequisite
expenditures, either.
***********************************
Somebody in your corner of the world must have a Genex Gx8500, GX9000
or GX9048 recorder. (the local mega-studio, or a rental house?)Ask to
hear a recording made in DSD format (not an old PCM recording that has
been converted). Then you can decide for yourself.
There's a Telarc 1812 Overture recording done completely in DSD. Telarc #
CD-80541

-Jay
--
x------- Jay Kadis ------- x---- Jay's Attic Studio ----x
x Lecturer, Audio Engineer x Dexter Records x
x CCRMA, Stanford University x http://www.offbeats.com/ x
x-------- http://ccrma-www.stanford.edu/~jay/ ----------x
LeBaron & Alrich
2003-10-19 00:43:31 UTC
Permalink
Post by Rob Adelman
I am suppose to learn something from some crazy test you devised?
What if?

--
ha
Rob Adelman
2003-10-19 01:01:10 UTC
Permalink
Post by LeBaron & Alrich
Post by Rob Adelman
I am suppose to learn something from some crazy test you devised?
What if?
I do have his "test tones" on the rap CD. Call me stupid, but to be
honest I don't have a clue what it all means.
Chris Johnson
2003-10-19 19:18:34 UTC
Permalink
Post by Rob Adelman
Post by Arny Krueger
Rob, since you're not responding to about 80% of my posts, I thought that a
little turnabout is fair play. I'm going to bring up a point that you've
been dodging for months, and one that you just dropped like a hot potato,
again.
I am suppose to learn something from some crazy test you devised?
You can learn what sorts of differences are so obvious you can hear
them every time no matter what the conditions.

That's not so uninteresting. For instance, doing that, I learned that
I could hear the difference between PCM and 320K mp3 every time no
matter what the conditions. I listened for a particular sort of
'personality' to the sound, and I have good speakers. Some people would
say that you can't hear that difference, but I sure can.

On the other hand, I suck at hearing background noise levels- my
brain ignores them, and I routinely listen at lowish levels with the
world's background noise relatively high. I can't hear much about noise
levels but I can tell 320K from PCM under those conditions- go fig.

You can't tell shit about momentary audio epiphanies by such a test,
but the guys hyping it don't know shit about momentary audio epiphanies
anyhow, so don't blame them.

Truly magical TONE can catch your ear and your mind for a fugitive
moment of surpassing beauty. Fuck that up and the moments aren't there,
but you'll still probably pass all the ABX tests because moments of
beauty are no part of double-blind testing.

Simple as that, really.


Chris Johnson
Scott Dorsey
2003-10-19 00:20:36 UTC
Permalink
Post by Rob Adelman
In fact you are quite wrong. The vast majority of the SACD's I bought
are not remastered at all.
Wow... how do they make those things from the original stampers, then?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
EggHd
2003-10-19 00:35:39 UTC
Permalink
<< Wow... how do they make those things from the original stampers, then? >>

Think of it like a cider press?



---------------------------------------
"I know enough to know I don't know enough"
Rob Adelman
2003-10-19 00:58:03 UTC
Permalink
Post by Scott Dorsey
Post by Rob Adelman
In fact you are quite wrong. The vast majority of the SACD's I bought
are not remastered at all.
Wow... how do they make those things from the original stampers, then?
Maybe your right, I guess it depends on how you are using the term
mastered. I guess I should say transfering to DSD encoding with the
least possible amount of changes made, eq, compression, etc.
EggHd
2003-10-19 01:02:10 UTC
Permalink
<< I guess I should say transfering to DSD encoding with the
least possible amount of changes made, eq, compression, etc. >>

But if your not at mastering and you have never heard the original masters (hey
haven't we been here before?) how do you know what's being done or not done?



---------------------------------------
"I know enough to know I don't know enough"
Rob Adelman
2003-10-19 01:12:52 UTC
Permalink
Post by EggHd
But if your not at mastering and you have never heard the original masters (hey
haven't we been here before?) how do you know what's being done or not done?
I guess we can say similar to how the original album was mastered?

I have a hard time accepting everyone saying that if the SACD sounds
better than CD that it isn't because of the format. Why would it be in
every case it is because they did a better job in mastering or somewhere
in the conversion process? They have had plenty of time to do the best
possible with CD.
EggHd
2003-10-19 01:16:08 UTC
Permalink
<< I guess we can say similar to how the original album was mastered? >>

Playing back on what medium?

<< Why would it be in
every case it is because they did a better job in mastering or somewhere
in the conversion process? >>

I found remixer's post very interesting. Did you read it?






---------------------------------------
"I know enough to know I don't know enough"
Les Cargill
2003-10-19 01:37:21 UTC
Permalink
Post by Rob Adelman
Post by Scott Dorsey
Post by Rob Adelman
In fact you are quite wrong. The vast majority of the SACD's I bought
are not remastered at all.
Wow... how do they make those things from the original stampers, then?
Maybe your right, I guess it depends on how you are using the term
mastered. I guess I should say transfering to DSD encoding with the
least possible amount of changes made, eq, compression, etc.
That's pretty much a complete description of the mastering
process, Rob.

--
Les Cargill
Rob Adelman
2003-10-19 01:48:59 UTC
Permalink
Post by Les Cargill
That's pretty much a complete description of the mastering
process, Rob.
But anywhere from a lot to almost nothing can be done to the song in the
mastering process. When you get three different (remastered) CDs of the
same record, and they all sound quite different, something more than a
basic transfer must be happening.
Les Cargill
2003-10-19 04:46:36 UTC
Permalink
Post by Rob Adelman
Post by Les Cargill
That's pretty much a complete description of the mastering
process, Rob.
But anywhere from a lot to almost nothing can be done to the song in the
mastering process. When you get three different (remastered) CDs of the
same record, and they all sound quite different, something more than a
basic transfer must be happening.
Absolutely.

--
Les Cargill
Scott Dorsey
2003-10-19 12:45:47 UTC
Permalink
Post by Rob Adelman
Post by Scott Dorsey
Post by Rob Adelman
In fact you are quite wrong. The vast majority of the SACD's I bought
are not remastered at all.
Wow... how do they make those things from the original stampers, then?
Maybe your right, I guess it depends on how you are using the term
mastered. I guess I should say transfering to DSD encoding with the
least possible amount of changes made, eq, compression, etc.
How can you know this? How can you know the original LP was done this way?
If you haven't heard the original master tapes, all you can say is that
the SACD issue is more or less like the LP or the CD issue that you have
heard. That's where the problem lies.

We can only make guesses as to what happened in the mastering room. Sometimes
we can make pretty good ones, but not always.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
LeBaron & Alrich
2003-10-19 17:12:27 UTC
Permalink
Post by Rob Adelman
Maybe your right, I guess it depends on how you are using the term
mastered. I guess I should say transfering to DSD encoding with the
least possible amount of changes made, eq, compression, etc.
I think the crux of the matter is that you should say that if you wre
there for the mastering sessions.

--
ha
Jay - atldigi
2003-10-19 21:58:46 UTC
Permalink
Post by LeBaron & Alrich
Post by Rob Adelman
Maybe your right, I guess it depends on how you are using the term
mastered. I guess I should say transfering to DSD encoding with the
least possible amount of changes made, eq, compression, etc.
I think the crux of the matter is that you should say that if you wre
there for the mastering sessions.
--
ha
Talking about mastering sessions, he should go get the new Yellowjackets
album, Time Squared, and comment about the CD vs. SACD layer. I was
there for those mastering sessions and at least for that one example I
could tell you the differences. I'm actually quite curious as to what
differences will be heard, and how they compare with what we all heard
in the mastering studio. Unfortunately, there is no DVD (A or V) version
available, so only I (and Dave and Rich) can hear the high sample rate
PCM version for comparison.
--
Jay Frigoletto
Mastersuite
Los Angeles
promastering.com
Les Cargill
2003-10-19 01:36:14 UTC
Permalink
Post by Scott Dorsey
Post by Rob Adelman
In fact you are quite wrong. The vast majority of the SACD's I bought
are not remastered at all.
Wow... how do they make those things from the original stampers, then?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Sort of like they press the reel to reel tapes to make
the records. Y'know?

--
Les Cargill
LeBaron & Alrich
2003-10-19 17:12:25 UTC
Permalink
Post by Scott Dorsey
Post by Rob Adelman
In fact you are quite wrong. The vast majority of the SACD's I bought
are not remastered at all.
Wow... how do they make those things from the original stampers, then?
They squeeze 'em really tightly between their knees. I know it's not
intuitive at all, but SACD is better because the grooves are deeper
because of the extra interknee pressure pressure applied to the original
masters.

--
ha
EggHd
2003-10-19 17:20:21 UTC
Permalink
<< I know it's not
intuitive at all, but SACD is better because the grooves are deeper
because of the extra interknee pressure pressure applied to the original
masters. >>

And now we have the true meaning of hard knee compression.



---------------------------------------
"I know enough to know I don't know enough"
LeBaron & Alrich
2003-10-20 01:54:25 UTC
Permalink
Post by EggHd
<< I know it's not
intuitive at all, but SACD is better because the grooves are deeper
because of the extra interknee pressure pressure applied to the original
masters. >>
And now we have the true meaning of hard knee compression.
Just another route to getting a leg up on yer competitors.

--
ha
Philip Perkins
2003-10-18 18:44:49 UTC
Permalink
In response to the statement that you can't edit or mix DSD audio
files: Sadie has two DSD DAWs out now that purport to do exactly that.
They are pretty straightforward guys, and there is no mention of them
converting the DSD to PCM for editing or mixing. "The integrity of
the 64fs DSD signal is fully maintained throughout the editing and
signal processing operation." www.sadie.com

Philip Perkins
hollywood_steve
2003-10-18 21:41:52 UTC
Permalink
Post by Philip Perkins
In response to the statement that you can't edit or mix DSD audio
files: Sadie has two DSD DAWs out now that purport to do exactly that.
They are pretty straightforward guys, and there is no mention of them
converting the DSD to PCM for editing or mixing. "The integrity of
the 64fs DSD signal is fully maintained throughout the editing and
signal processing operation." www.sadie.com
There are several editing systems that do NOT convert to PCM and they
have been available for quite a while. The latest comes from Genex,
who make some very nice DSD recorders. I purchased the GX9000 earlier
this year and I'm thrilled with the DSD converters. Until you have
recorded directly to DSD and listened to the playback without any PCM
steps involved, you haven't heard a real DSD recording.

And the best way I know of to compare DSD to PCM is to use a recorder
like the Genex that has high quality DSD and PCM converters. Record
the same signal to both PCM and DSD and compare. I've done it several
times and I'm sold.

steve
***@pacbell.net
www.lexington125.com
Rob Adelman
2003-10-18 21:46:04 UTC
Permalink
Post by hollywood_steve
There are several editing systems that do NOT convert to PCM and they
have been available for quite a while. The latest comes from Genex,
who make some very nice DSD recorders. I purchased the GX9000 earlier
this year and I'm thrilled with the DSD converters. Until you have
recorded directly to DSD and listened to the playback without any PCM
steps involved, you haven't heard a real DSD recording.
Are you doing multi-track recording or 2 track? I am very excited to
hear the results of multi-tracking into DSD.
Post by hollywood_steve
And the best way I know of to compare DSD to PCM is to use a recorder
like the Genex that has high quality DSD and PCM converters. Record
the same signal to both PCM and DSD and compare. I've done it several
times and I'm sold.
Maybe you can put both on an MP3 and send to Arny to compare ;)
Kurt Albershardt
2003-10-19 00:49:55 UTC
Permalink
Post by hollywood_steve
And the best way I know of to compare DSD to PCM is to use a recorder
like the Genex that has high quality DSD and PCM converters. Record
the same signal to both PCM and DSD and compare. I've done it several
times and I'm sold.
What sort of filter coefficients do they use on the 96k and 192k
decimations?
Remixer
2003-10-18 23:45:49 UTC
Permalink
By definition, it is not possible to process DSD without transcoding to some
form of PCM. Pure DSD is 1-bit and as soon as you process (add, or
multiply) you have a multi-bit result which is no longer DSD - i.e. it
cannot be played as a 1-bit stream. The Sonoma (Sony), Sonic, and Sadie
DSD DAWs all do their work by transcoding to what is called "DSD-wide" aka
"PCM-narrow" which is 8-bits at 64Fs. (2.8MHz) All processing on these
platforms is done with Sony Oxford "D-Map" modules which are OEM circuit
boards supporting mix, EQ, and dynamics processing in PCM-narrow (DSD-wide)
format.

The Pyramix system transcodes pure DSD to 32-bit float at 8Fs (352.4KHz) for
processing. Tests by Philips indicate that LPCM at 8Fs is sufficiently
transparent for DSD. There are many respected professional ears who
believe that LPCM at 4Fs (176.4KHz) is sufficiently transparent for DSD, or
any audio for that matter, so 8Fs or DSD-wide should be sufficient for
interim production processes.

IMO, the value of DSD is as a recording and distribution medium. For one
thing, DSD uses approximately the same bit-budget for storage as
88.2k/24-bit but (IMO) it sounds much better than 88.2k LPCM. For another,
an inexpensive DSD DAC can sound a lot better than all but the most
expensive PCM DACs because of the relaxed filtering in the audio band. So,
its not that DSD makes a magical transformation of 1Fs sources, it's that
the d-to-a process can sound better. Think of up-sampling 44.1k to DSD and
playing back on decent SACD player as roughly equivalent to playing the
original 44.1 through a really top-of-the line DAC, except that with DSD you
can get that level of quality with a DAC of 1/10th the price of a comparable
PCM converter.
xy
2003-10-19 19:08:11 UTC
Permalink
so if you had a pcm source file that you were going to mix/master
through analog outboard, it would make sense to off-line
process-convert the file into dsd and then run that signal out into
the analog?
Scott Dorsey
2003-10-19 20:01:30 UTC
Permalink
Post by xy
so if you had a pcm source file that you were going to mix/master
through analog outboard, it would make sense to off-line
process-convert the file into dsd and then run that signal out into
the analog?
Only if this allows you to avoid some inherent problem with your PCM
DAC.

And it's entirely possible that all DSD buys us is better conversion, and
transcoding from DSD to PCM and back again will do everything that storing
the DSD bitstream will. It's equally possible that it buys us more. I
don't know yet and I am suspicious of anyone who claims to.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Remixer
2003-10-20 04:47:09 UTC
Permalink
It would depend on what gear is available to you. With a top-tier PCM dac
there might not be much of an advantage to upsample to DSD. Such DACs are
no longer restricted to the $5000 and up category, for example, the Mytek
Stereo for $995 upsamples to 192k and reduces jitter. The Benchmark DAC-1
($795)upsamples to 96k, also with jitter reduction. On the other hand,
there are phenomenal new (off-line) PCM-to-DSD modulators such as the new
Trellis algorithm from Philips, but that's an $8k piece of software.
Post by xy
so if you had a pcm source file that you were going to mix/master
through analog outboard, it would make sense to off-line
process-convert the file into dsd and then run that signal out into
the analog?
Chris Hornbeck
2003-10-20 01:25:17 UTC
Permalink
Post by Remixer
By definition, it is not possible to process DSD without transcoding to some
form of PCM.
Thanks for your very clear explanation. Can you or anyone tell us
if there is a one to one mapping between a 1-bit stream and a
multibit PCM stream? If so, do both have the same bit storage
rate?

Do we need one of the heavy math guys like Bob Cain to answer
this, or is it more in the category of stuff I should already
know?

Thanks,

Chris Hornbeck
new email address
Bob Cain
2003-10-20 02:28:30 UTC
Permalink
Post by Chris Hornbeck
Post by Remixer
By definition, it is not possible to process DSD without transcoding to some
form of PCM.
Thanks for your very clear explanation. Can you or anyone tell us
if there is a one to one mapping between a 1-bit stream and a
multibit PCM stream? If so, do both have the same bit storage
rate?
All you have to do to convert 1-bit delta modulation to PCM
is run it into the Up/Down control of a counter and clock it
on every bit. Where it ceases to be one to one is how often
you would chose to sample the counter for output. That can
be any integral divisor of the 1-bit clock rate.
Post by Chris Hornbeck
Do we need one of the heavy math guys like Bob Cain to answer
this, or is it more in the category of stuff I should already
know?
I can't be of much help on this one as I've never studied
the characteristics of delta modulated codecs. It is
essentially a slew rate limited process, however, and I've
never been terribly comfortable with that.


Bob
--
"Things should be described as simply as possible, but no
simpler."

A. Einstein
Chris Hornbeck
2003-10-20 04:32:10 UTC
Permalink
Post by Bob Cain
Post by Chris Hornbeck
or is it more in the category of stuff I should already
know?
On Sun, 19 Oct 2003 19:28:30 -0700, Bob Cain
Post by Bob Cain
I can't be of much help on this one as I've never studied
the characteristics of delta modulated codecs. It is
essentially a slew rate limited process, however, and I've
never been terribly comfortable with that.
Hi Bob,
I think that I've framed the question so poorly that it
can't be answered in language simple enough for me to
understand. I've made the old mistake of thinking of the
A/D and D/A in isolation, without the anti-alias filtering.

Guess I need to do more research first.
Thanks, as always,

Chris Hornbeck
new email address
Bob Cain
2003-10-20 22:39:40 UTC
Permalink
Remixer, do you happen to know the base clock rate of 1-bit
DSD and what differential in output voltage each bit
represents?


Thanks,

Bob
--
"Things should be described as simply as possible, but no
simpler."

A. Einstein
Harry Lavo
2003-10-19 03:08:22 UTC
Permalink
Post by Philip Perkins
In response to the statement that you can't edit or mix DSD audio
files: Sadie has two DSD DAWs out now that purport to do exactly that.
They are pretty straightforward guys, and there is no mention of them
converting the DSD to PCM for editing or mixing. "The integrity of
the 64fs DSD signal is fully maintained throughout the editing and
signal processing operation." www.sadie.com
Philip Perkins
Moreover, SONY was showing hardware components and Phillips was showing
software modules to allow many manufacturers to build all the processing
tools they can imagine...all in pure dsd or dsd-wide. So the mixers, input
boxes, effectsmachines, software packages etc will soon start showing up
in quantity, I would guess.
Scott Dorsey
2003-10-19 12:52:03 UTC
Permalink
Post by Harry Lavo
Moreover, SONY was showing hardware components and Phillips was showing
software modules to allow many manufacturers to build all the processing
tools they can imagine...all in pure dsd or dsd-wide. So the mixers, input
boxes, effectsmachines, software packages etc will soon start showing up
in quantity, I would guess.
Maybe, but they weren't showing what was inside them. DSD is kind of
funny, in that scaling and gain control are very hard to do without
some serious computational horsepower. How are the Sony board implementing
the EQ and gain control functions? The guys I talked to at the booth
weren't sure, and the only paper I saw from them was on the new DSD/PCM
interface design.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Remixer
2003-10-19 15:17:51 UTC
Permalink
Post by Scott Dorsey
How are the Sony board implementing
the EQ and gain control functions? The guys I talked to at the booth
weren't sure, and the only paper I saw from them was on the new DSD/PCM
interface design.
--scott
Sorry for the re-post, but it seems pertinent again:

By definition, it is not possible to process DSD without transcoding to some
form of PCM. Pure DSD is 1-bit and as soon as you process (add, or
multiply) you have a multi-bit result which is no longer DSD - i.e. it
cannot be played as a 1-bit stream. The Sonoma (Sony), Sonic, and Sadie
DSD DAWs all do their work by transcoding to what is called "DSD-wide" aka
"PCM-narrow" which is 8-bits at 64Fs. (2.8MHz) All processing on these
platforms is done with Sony Oxford "D-Map" modules which are OEM circuits
supporting mix, EQ, and dynamics processing in PCM-narrow (DSD-wide)
format.

The Pyramix system transcodes pure DSD to 32-bit float at 8Fs (352.4KHz) for
processing. Tests by Philips indicate that LPCM at 8Fs is sufficiently
transparent for DSD. There are many respected professional ears who
believe that LPCM at 4Fs (176.4KHz) is sufficiently transparent for DSD, or
any audio for that matter, so 8Fs or DSD-wide should be sufficient for
interim production processes.

IMO, the value of DSD is as a recording and distribution medium. For one
thing, DSD uses approximately the same bit-budget for storage as
88.2k/24-bit but (IMO) it sounds much better than 88.2k LPCM. For another,
an inexpensive DSD DAC can sound a lot better than all but the most
expensive PCM DACs because of the relaxed filtering in the audio band. So,
its not that DSD makes a magical transformation of 1Fs sources, it's that
the d-to-a process can sound better. Think of up-sampling 44.1k to DSD and
playing back on decent SACD player as roughly equivalent to playing the
original 44.1 through a really top-of-the line DAC, except that with DSD you
can get that level of quality with a DAC of 1/10th the price of a comparable
PCM converter.

A.
Harry Lavo
2003-10-19 16:09:03 UTC
Permalink
Post by Scott Dorsey
Post by Harry Lavo
Moreover, SONY was showing hardware components and Phillips was showing
software modules to allow many manufacturers to build all the processing
tools they can imagine...all in pure dsd or dsd-wide. So the mixers, input
boxes, effectsmachines, software packages etc will soon start showing up
in quantity, I would guess.
Maybe, but they weren't showing what was inside them. DSD is kind of
funny, in that scaling and gain control are very hard to do without
some serious computational horsepower. How are the Sony board
implementing
Post by Scott Dorsey
the EQ and gain control functions? The guys I talked to at the booth
weren't sure, and the only paper I saw from them was on the new DSD/PCM
interface design.
Sorry, Scott, I don't have the engineering background to understand how they
work so it was not a point of inquiry for me. It was enough for me (a
non-ee) to know they were available to any manufacturer that wanted to use
them. Obviously the engineering is tricky, which is why SONY seems to be
positioning themselves as consultant engineers providing toolkits and ready
to help in development of custom modules as well as providing ones already
prefabbed. To me, the most important thing is that the days of limited
hardware availability for recording and production seem about to end. And
then as night follows the day, competition will heat up and eventually
prices will come down as knowledge spreads and the learning curves are
scaled.
Scott Dorsey
2003-10-18 15:29:34 UTC
Permalink
Post by Rob Adelman
Post by Arny Krueger
Ironically, some of these same people go on for zillions of posts on Usenet
about the improved sound quality of DSD recordings that are now known to
have been made from 48 KHz PCM recordings but...
Well the first step in forming your own opinion might be to ACTUALLY
LISTEN TO ONE!
The problem is that this isn't enough when you don't know the processing
being done.

The fact that DSD makes flattopping harder to do is a big deal, and will
probably result in greatly improved sound quality. Not for any technical
reasons, though, just for social ones.

I have LPs that sound better than the CD, and CDs that sound better than
the LPs. The guys in the mastering room have a lot more to do with the
sound quality than the release format. This isn't going to change.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
LeBaron & Alrich
2003-10-18 22:21:09 UTC
Permalink
Post by Rob Adelman
Post by Arny Krueger
Ironically, some of these same people go on for zillions of posts on Usenet
about the improved sound quality of DSD recordings that are now known to
have been made from 48 KHz PCM recordings but...
Well the first step in forming your own opinion might be to ACTUALLY
LISTEN TO ONE!
You type that in all caps after admitting you used a 1" 24 track deck
for a good long while without ever bothering to align and calibrate it,
and that as far as your own ears could tell, what came out of it sounded
just like what went into it? I am being quite serious, Rob. How is
anyone supposed to hang credibility on these kinds of claims? Please
just stop and think about it for a few minutes...

DSD magically converts 48 KHz PCM recordings into something that the DSD
priests want me to believe can't exist in PCM? Is there anything
potentially erroneous with that "logic"?

So PCM is chickenshit until it gets DSD'd and then all of a sudden it's
chicken soup?

--
ha
Rob Adelman
2003-10-18 23:18:49 UTC
Permalink
Post by LeBaron & Alrich
You type that in all caps after admitting you used a 1" 24 track deck
for a good long while without ever bothering to align and calibrate it,
and that as far as your own ears could tell, what came out of it sounded
just like what went into it? I am being quite serious, Rob. How is
anyone supposed to hang credibility on these kinds of claims? Please
just stop and think about it for a few minutes...
You going to go on and on about for the next 10 years? Did you hear the
deck? No. So you really have no basis for your belief or not.
Post by LeBaron & Alrich
DSD magically converts 48 KHz PCM recordings into something that the DSD
priests want me to believe can't exist in PCM? Is there anything
potentially erroneous with that "logic"?
What are you talking about? I am talking about the release of old tape
masters encoded in DSD. Where does the PCM come in?
Post by LeBaron & Alrich
So PCM is chickenshit until it gets DSD'd and then all of a sudden it's
chicken soup?
No, I am talking about comparing 20 years of the best attempts of
releasing old recordings on CD vs the same masters being released on
SACD. That's why I don't understand Scott's argument either, which I
have in fact responded to before. Same master tapes being released on
the best CD has to offer vs the best SACD has to offer. Seems like a
valid comparison to me.
Tommi
2003-10-19 00:15:49 UTC
Permalink
DVD-A is better than SACD is better than CD.

The SACD presents an inferior signal compared to a _normal cd_ on the lowest
two octaves or so in our hearing range, though. A couple of objective,
independent tests have been made about this. Basically it's because the big,
long-lasting wavelenghts are somehow harder to capture in DSD than PCM. This
has been proven in theory as well as in practice.

All this raving about SACD is just one of those Hi-Fi -things. The fact that
sony is behind it makes it even more difficult to make people understand
that there is nothing mysteriously good about the DSD sound. Sony has been
an influential company, brought us many good things. But, when minidisc came
out they said it was cd-quality. It wasn't. In televisions they make
everything "seem natural, so that the picture represents the truth". This is
done by adding a bit of blue into everything on the screen what's white. Our
eyes perceive this as a natural "white"; that means that it isn't the truth,
it's a psychological thing. Sony have a great marketing background(came out
of nowhere to the game console market without any games but a big marketing
campaign and now they're the leader), and lots of guys who have studied
their psychoacoustics etc. very well. They make things that are easy to
perceive as "good".

And then they advice you to record in PCM even if the final product would be
DSD.
Remixer
2003-10-19 00:15:03 UTC
Permalink
Post by Tommi
independent tests have been made about this. Basically it's because the big,
long-lasting wavelenghts are somehow harder to capture in DSD than PCM. This
has been proven in theory as well as in practice.
Where did you pick up this myth? DSD is exceptionally linear and has
vanishingly low IM distortion.
Tommi
2003-10-19 00:33:36 UTC
Permalink
I just realised I remembered it the wrong way: it's the high frequencies the
DSD's having problems with. Here's one, and it's from a very respectable
person.


http://sound.westhost.com/cd-sacd-dvda.htm
Chris Johnson
2003-10-19 19:54:16 UTC
Permalink
Post by Tommi
I just realised I remembered it the wrong way: it's the high frequencies the
DSD's having problems with. Here's one, and it's from a very respectable
person.
Ooo, and you sounded so convinced, too ;)

...which is pretty much moot- how exactly are you going to percieve
even 10% harmonic distortion on a 20K tone when you can't hardly hear
the 20K tone in the first place? DSD's wise to dump the error into the
extreme highs like that. You can't hear it.

Also, if the noise shaping spazzes out when hit with edgy content,
this produces a sort of correlated supersonic 'tizz' along with the
content, and you could just as well think of that as an 'aural
enhancer'. Noise is at X volume, and when you nail a cymbal the
supersonics, which you can't really hear, do a big distorted splat.
That's not all that different from the splat you get from a real cymbal,
it's just a DIFFERENT splat, and it's sensitive to the hotness of the
signal. Call it a DSD Sonic Maximizer and just enjoy it...


Chris Johnson
Rob Adelman
2003-10-19 00:54:12 UTC
Permalink
Post by Tommi
DVD-A is better than SACD is better than CD.
The SACD presents an inferior signal compared to a _normal cd_ on the lowest
two octaves or so in our hearing range, though. A couple of objective,
independent tests have been made about this. Basically it's because the big,
long-lasting wavelenghts are somehow harder to capture in DSD than PCM. This
has been proven in theory as well as in practice.
Interesting. I find the bass noticeably better on SACD. Along with every
other range.
Scott Dorsey
2003-10-19 01:56:20 UTC
Permalink
Post by Tommi
The SACD presents an inferior signal compared to a _normal cd_ on the lowest
two octaves or so in our hearing range, though. A couple of objective,
independent tests have been made about this. Basically it's because the big,
long-lasting wavelenghts are somehow harder to capture in DSD than PCM. This
has been proven in theory as well as in practice.
I've never heard this, and I am not sure it makes any sense to me. Have
you got a citation for this?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
William Sommerwerck
2003-10-19 12:42:37 UTC
Permalink
Post by Tommi
The SACD presents an inferior signal compared to a _normal cd_ on the
lowest two octaves or so in our hearing range, though. A couple of objective,
independent tests have been made about this. Basically it's because the big,
long-lasting wavelenghts are somehow harder to capture in DSD than PCM.
This has been proven in theory as well as in practice.
Ummm... Long-lasting wavelenghts... Ummm...

Where do you get this crap? If such a difference existed, it would be at the
higher frequencies.

I won't comment on the generally ignorant remarks that follow...
Tommi
2003-10-19 23:27:30 UTC
Permalink
Post by William Sommerwerck
Ummm... Long-lasting wavelenghts... Ummm...
Where do you get this crap? If such a difference existed, it would be at the
higher frequencies.
I won't comment on the generally ignorant remarks that follow...
Yeah, I didn't sound very convincing because I remembered the DSD's crap
resolution the wrong way! Personally, I have long ago dropped the DSD (or
SACD to be more specific) as a serious contender for DVD-Audio, which is why
I didn't remember how DSD technique exactly works, or where its problems
are.
The point is, does it matter if SACD sounds better than a cd? DVD-Audio
sounds better than either of them.
A decent 1-bit DSD system is just easier to build than a decent 24/192 PCM
system. There's no "conspiracy theory" aspect about the fact that sony learn
most of their stuff from consumer-based technologies. They've made 1-bit
converters for standard cd players for a long time already, and now they
market sacd as the next-generation option.

Obviously, SACD sounds very good, but DVD-Audio is just much more reliable
and more accurate. SACD seems to smoothen out some transients which is why
some people think it sounds very "natural", but it isn't the truth. DSD
makes assumptions based on the last sample that came through the system.
Then if the following sample is larger in voltage, it gives it a value of 1.
This is it's fundamental flaw: theoretically, it never actually plays back
the same signal material twice identically, because it rounds the sample
intervals in such a simple "louder 1, quieter 0" manner. Therefore, it's
system is based on shaky voltage relationships and not discrete, independent
values like PCM, which is why I think PCM as "rock-solid".
.
DSD trusts in the incapabilities of human hearing, and assumes that no-one
cares about frequencies over 16kHz or so. To those people sacd may actually
sound like the best medium, but in reality it's not as accurate as dvd-a. A
lot of people hear stuff over 23 kHz, and in the real world there are
frequencies over 100kHz. A crash cymbal has 40% of its energy in the
"ultrasonic" range, and wheteher or not that fact matters, the bottom line
is that sacd is incapable of bringing you exact information. The closest
thing you can get to "truth" currently is, without a question, the DVD-A.
I wouldn't make an opinion about which sounds better (sacd or dvd-a) based
purely on my hearing, because both formats are so good that most of their
differences are just beyond the limits of our hearing; if I'd trust my
instincts, I'd say that digital sounds bad because it sounds too cold
compared to tape. I can just say that digitized samples represent the truth
more accurately than an analog tape.
Rob Adelman
2003-10-19 23:42:06 UTC
Permalink
Post by Tommi
Yeah, I didn't sound very convincing because I remembered the DSD's crap
resolution the wrong way!
And you still don't. You go from "knowing" absolutely that DVD-A sounds
better, to technically it has to be better in the high range and you
can't use your ears to verify this, to if you can't hear that it is
better then you can't hear high end.

Sounds to me like you decided DSD must suck before you ever heard it and
having been busy convincing yourself you are right ever since.
Tommi
2003-10-20 00:28:27 UTC
Permalink
Post by Rob Adelman
And you still don't. You go from "knowing" absolutely that DVD-A sounds
better, to technically it has to be better in the high range and you
can't use your ears to verify this, to if you can't hear that it is
better then you can't hear high end.
Sounds to me like you decided DSD must suck before you ever heard it and
having been busy convincing yourself you are right ever since.
These things are seldom simple. What sounds good but isn't theoretically
accurate may be just enough, but why settle for that if there's another
system which also sounds good and on top of that is theoretically more
accurate? (Now I hear someone pointing out PCM LSB bad res)
It's like in visual editing in screens so small our eyes can't see the
difference; if you're looking at a still image with 2000 vertical lines from
a 4"x3" monitor, you say: "wow, that's clear". Then you look at the same
image from the same screen, only this time with 4000 vertical lines(but
no-one's told you the resolution's changed), you go: "geez, that's as
accurate as the previous picture!" In these kind of situations you have to
make decisions based not only on what your senses tell you, but also on
knowing how the computer's handling the data.
Actually I was pretty enthusiastic about DSD and SACD a couple of years ago,
but when the same record's played back on dvd-a and sacd, clearly audible
differences are usually mastering eq or compression etc, and barely audible
differences could be the colorations of any component inside the player. I
didn't say sacd "sucks", I just think that both formats are so good that I
can only say PCM technology is more reliable than DSD.
Kurt Albershardt
2003-10-19 23:47:04 UTC
Permalink
Post by Tommi
The point is, does it matter if SACD sounds better than a cd? DVD-Audio
sounds better than either of them.
I don't know if I'd agree there since I haven't found a way to fairly
compare the two hires formats. How did you come to your conclusion?
Post by Tommi
There's no "conspiracy theory" aspect about the fact that sony learn
most of their stuff from consumer-based technologies.
Sony 'learn most of their stuff' from the fruits of their (rather deep)
research budget. Some of it ends up in consumer gear, and they often
take advantage of the economies of scale it brings to enhance their
industrial and broadcast hardware. More often, the exotic stuff starts
out in broadcast and works its way down.
Tommi
2003-10-20 00:46:06 UTC
Permalink
Post by Kurt Albershardt
Post by Tommi
The point is, does it matter if SACD sounds better than a cd? DVD-Audio
sounds better than either of them.
I don't know if I'd agree there since I haven't found a way to fairly
compare the two hires formats. How did you come to your conclusion?
That's one of the problems, it's difficult to do fair tests to both hi-res
formats yourself at home, but I can say that dvd-a sounds better more than
half of the time. If you add to this the DSD's flaws(which are not really
that big but you have to decide based on something)compared to PCM, I can at
least say that dvd-a seldom sounds worse than sacd and on top of that it's
theoretically more accurate.
William Sommerwerck
2003-10-20 01:18:51 UTC
Permalink
I don't have time to go into detail, but I think your understanding of how
digital audio works, and the various assumptions/compromises, is poor. That's
not a very courteous thing to say, I know, but I simply don't have time to
discuss it at length.

The most-basic point you're missing is that bit-depth and sample rate can traded
off against each other.
Tommi
2003-10-20 03:23:00 UTC
Permalink
Hey, that might even be true. If it is so, what books/websites would you
recommend for me to read?
Since I know only what I've learned , I've been under the impression that
the basics of dsd and pcm differ greatly. The biggest point is that you
can't just trade bit depth for sample rate if the case we're talking about
is a dsd system which basically _compares_ the last signal to the next.
Therefore it is not a linear system anymore like pcm where every consecutive
sample has its own value regardless of the sample preceding it. Even though
bit depth and sample rate are the cornerstones of any digital system, I
personally wouldn't say that it's just a simple trade-off.
Jay - atldigi
2003-10-20 08:19:37 UTC
Permalink
Post by Tommi
Hey, that might even be true. If it is so, what books/websites would you
recommend for me to read?
Since I know only what I've learned , I've been under the impression that
the basics of dsd and pcm differ greatly. The biggest point is that you
can't just trade bit depth for sample rate if the case we're talking
You can think of DSD like PCM that is severely noise shaped. It's 1 bit
and so has a dynamic range of about 6dB, but you can shape a tremendous
amount of that noise which is only 6dB below full scale and push it into
very high (thus inaudible) parts of the spectrum, especially if the
sample rate is great enough and you can spread it out over several
octaves. This essentially gives you an equivalent performance near 120dB
within the limited bandwidth of the range of human hearing.

The other commonly used example for DSD is that it's very much like an
oversampling converter where you directly record the output of the
oversampled ADC without quantizing it into PCM and converting to the
target Fs.

The example of recording only whether the signal is rising or falling as
opposed to a discrete value for each sample is perhaps better thought of
with the initial sampling and not so much the storage. If it were that
simple, you could never start playing a track in the middle!
--
Jay Frigoletto
Mastersuite
Los Angeles
promastering.com
Mike Caffrey
2003-10-20 17:16:58 UTC
Permalink
Post by William Sommerwerck
I don't have time to go into detail, but I think your understanding of how
digital audio works, and the various assumptions/compromises, is poor. That's
not a very courteous thing to say, I know, but I simply don't have time to
discuss it at length.
The most-basic point you're missing is that bit-depth and sample rate can traded
off against each other.
That's news to me.

My understanding is that smaple rate is the number of "picutres" per
second and bit rate is the resolution of the pictures.

I've noticed bigger jumps in fidelity when increasing bit rate than
smaple rate. There a penty of factors that could explain that though.



www.monsterisland.com
Tommi
2003-10-20 22:42:07 UTC
Permalink
Post by William Sommerwerck
Post by William Sommerwerck
The most-basic point you're missing is that bit-depth and sample rate
can traded
Post by William Sommerwerck
off against each other.
That's news to me.
My understanding is that smaple rate is the number of "picutres" per
second and bit rate is the resolution of the pictures.
I've noticed bigger jumps in fidelity when increasing bit rate than
smaple rate. There a penty of factors that could explain that though.
That's what I thought too. If this isn't the case, I guess I don't know
anything. btw thanks to Jay F. for trying to explain dsd in such a pcm way
that my pcm head
understood it. I hope William soon explains how and what basic points I'm
missing, and are they dsd/pcm -specific things or all-round basics which
I've never come across.
Mike Caffrey
2003-10-20 23:35:05 UTC
Permalink
Post by Tommi
Post by William Sommerwerck
Post by William Sommerwerck
The most-basic point you're missing is that bit-depth and sample rate
can traded
Post by William Sommerwerck
off against each other.
That's news to me.
My understanding is that smaple rate is the number of "picutres" per
second and bit rate is the resolution of the pictures.
I've noticed bigger jumps in fidelity when increasing bit rate than
smaple rate. There a penty of factors that could explain that though.
That's what I thought too. If this isn't the case, I guess I don't know
anything. btw thanks to Jay F. for trying to explain dsd in such a pcm way
that my pcm head
understood it. I hope William soon explains how and what basic points I'm
missing, and are they dsd/pcm -specific things or all-round basics which
I've never come across.
I save Jay's explainaiton, but I still don't undstand how DSD works well
enough to explain it to someone like I can with PCM.



www.monsterisland.com
Arny Krueger
2003-10-21 00:30:30 UTC
Permalink
Post by William Sommerwerck
Post by William Sommerwerck
I don't have time to go into detail, but I think your understanding
of how digital audio works, and the various assumptions/compromises,
is poor. That's not a very courteous thing to say, I know, but I
simply don't have time to discuss it at length.
The most-basic point you're missing is that bit-depth and sample rate
can traded
Post by William Sommerwerck
off against each other.
That's news to me.
My understanding is that smaple rate is the number of "picutres" per
second and bit rate is the resolution of the pictures.
Pretty good metaphor.
Post by William Sommerwerck
I've noticed bigger jumps in fidelity when increasing bit rate than
sample rate. There a plenty of factors that could explain that though.
It has a lot to do with where you are starting from. The higher your
starting point, the more subtle the benefits.

If you add 4 bits more resolution to 4 bits (giving 8 bits) the benefits are
exceedingly obvious. If you add 4 more bits of resolution to 8 bits, the
results (12 bits) sound obviously better. Add 4 bits to 12 bits (giving 16
bits) and the improvement is far more subtle. Add 4 bits to 16 bits and the
benefits are exceedingly subtle to inaudible. You can listen to some of
these situations as applied to musical and other sounds at
http://www.pcabx.com/technical/bits44/index.htm .

It's much the same way with sample rates. Increases in sample rates have
pretty obvious benefits up to about 22 KHz, but above 32 KHz the benefits
become very much more difficult to hear. You can listen to these situations
as applied to musical sounds at
http://www.pcabx.com/technical/sample_rates/index.htm .

Vladan
2003-10-20 23:37:34 UTC
Permalink
On Sun, 19 Oct 2003 18:18:51 -0700, "William Sommerwerck"
Post by William Sommerwerck
The most-basic point you're missing is that bit-depth and sample rate can traded
off against each other.
It would be nice if we knew who you are talking to, but nevertheless,
I'll bite and ask you: How? How can sample rate be traded with bit
depth?! At what proportion are they traded? 6dB per doubling, or 3dB,
or what? No way!

Hope you have some recording gear to try it at "home". It's as easy as
recording same source twice, hopefuly via digital, than try upsampling
one, downsapling another... No way you can gain anything that way.
Vladan
www.geocities.com/vla_dan_l
www.mp3.com/lesly , www.mp3.com/shook , www.mp3.com/lesly2
www.kunsttick.com/artists/vuskovic/indexdat.htm
Chris Johnson
2003-10-19 19:50:26 UTC
Permalink
Post by Tommi
DVD-A is better than SACD is better than CD.
The SACD presents an inferior signal compared to a _normal cd_ on the lowest
two octaves or so in our hearing range, though. A couple of objective,
independent tests have been made about this. Basically it's because the big,
long-lasting wavelenghts are somehow harder to capture in DSD than PCM. This
has been proven in theory as well as in practice.
News to me. The error distribution of DSD _favors_ accuracy on mid
and low frequencies. Define 'somehow', it seems to me like you're
talking absolute backward nonsense. When you have a continually reducing
noise/error factor as frequency drops, it is BETTER accuracy at the
extreme lows. DSD's compromised on the extreme highs instead- which is
arguably moot, as people can't really hear high resolution up there,
only the existence of sound in a crude way.


Chris Johnson
EggHd
2003-10-19 00:41:56 UTC
Permalink
<< You going to go on and on about for the next 10 years? Did you hear the
deck? No. So you really have no basis for your belief or not. >>

I haven't heard your deck (I've used others) but the 1" 24 track Tascam
machines with NR are OK but nothing like great or anything. Very good is a
stretch. Add that to moving it across the country a couple of times and yikes.





---------------------------------------
"I know enough to know I don't know enough"
Jay - atldigi
2003-10-19 03:37:07 UTC
Permalink
Post by Rob Adelman
No, I am talking about comparing 20 years of the best attempts of
releasing old recordings on CD vs the same masters being released on
SACD.
Then that's SACD vs. CD, not DSD vs. PCM. Generalizations about all PCM
can't be made only from CD.
--
Jay Frigoletto
Mastersuite
Los Angeles
promastering.com
LeBaron & Alrich
2003-10-19 17:12:29 UTC
Permalink
Post by Rob Adelman
You going to go on and on about for the next 10 years? Did you hear the
deck? No. So you really have no basis for your belief or not.
No, but I have heard other of those decks. I've even made a decent
sounding jazz product using an uncalibrated 1/2" 8 track. It wasn't my
machine and the folks owning it weren't interested in paying to get it
maintained, until it quit mid-session and had to go to a shop for a
couple weeks.

But I've heard some other tape decks, Studers a lot, and others more
than a few times. What I get out doesn't exactly match what goes in but
it's a good deal closer than what I've gotten out of narrow guage
machines, and some of that is due to the capabilities of the decks and
some of it has to do with basic analog machine maintenance.

My "belief" is based upon practical firsthand experience with this
equipment.

--
ha
Mike Caffrey
2003-10-19 23:42:23 UTC
Permalink
Post by Rob Adelman
Post by LeBaron & Alrich
You type that in all caps after admitting you used a 1" 24 track deck
for a good long while without ever bothering to align and calibrate it,
and that as far as your own ears could tell, what came out of it sounded
just like what went into it? I am being quite serious, Rob. How is
anyone supposed to hang credibility on these kinds of claims? Please
just stop and think about it for a few minutes...
You going to go on and on about for the next 10 years? Did you hear the
deck? No. So you really have no basis for your belief or not.
Post by LeBaron & Alrich
DSD magically converts 48 KHz PCM recordings into something that the DSD
priests want me to believe can't exist in PCM? Is there anything
potentially erroneous with that "logic"?
What are you talking about? I am talking about the release of old tape
masters encoded in DSD. Where does the PCM come in?
Post by LeBaron & Alrich
So PCM is chickenshit until it gets DSD'd and then all of a sudden it's
chicken soup?
No, I am talking about comparing 20 years of the best attempts of
releasing old recordings on CD vs the same masters being released on
SACD. That's why I don't understand Scott's argument either, which I
have in fact responded to before. Same master tapes being released on
the best CD has to offer vs the best SACD has to offer. Seems like a
valid comparison to me.
The Dylan SACD re-release panel at AES was very interesting. The first
part they talked abou was finding the acutal masters which in many cases
were unlabeld. They had very knowledgable people decide whether or not the
beleived the tapes were that acutal masters. In some cases they knew they
weren't the masters. In some cases they found earlier remasters, say a
multi track copied to another for some tpye of processing, such as dolby.
They did their best to avoid usings these tapes.

They made if very clear that getting back to the original masters that
were used to make the original vinyl was very difficult and that they may
not have a way to know for 100% cetrainty that they did. My sense is that
they were very thorough and probably know the rare cases in which they
didn't find the originals.

My point is, that if you weren't involved in the re-issue process and
don't know whethere your A/Bing a SACD made from original masters to the
original vinyl, your test may not be valid. They also said that many of
the earlier remixes/remasters were not done faithfully to the originals,
so they are not necessarily good samples for comparison tests.

You really have to make your test materials from scratch.



www.monsterisland.com
Rob Adelman
2003-10-19 23:48:09 UTC
Permalink
Post by Mike Caffrey
My point is, that if you weren't involved in the re-issue process and
don't know whethere your A/Bing a SACD made from original masters to the
original vinyl, your test may not be valid. They also said that many of
the earlier remixes/remasters were not done faithfully to the originals,
so they are not necessarily good samples for comparison tests.
The bottom line is the end result. However it got there, my experience
is that the SACD sounds better than the CD in almost every case.
Mike Caffrey
2003-10-19 23:33:27 UTC
Permalink
Post by LeBaron & Alrich
Post by Rob Adelman
Post by Arny Krueger
Ironically, some of these same people go on for zillions of posts on Usenet
about the improved sound quality of DSD recordings that are now known to
have been made from 48 KHz PCM recordings but...
Well the first step in forming your own opinion might be to ACTUALLY
LISTEN TO ONE!
You type that in all caps after admitting you used a 1" 24 track deck
for a good long while without ever bothering to align and calibrate it,
and that as far as your own ears could tell, what came out of it sounded
just like what went into it? I am being quite serious, Rob. How is
anyone supposed to hang credibility on these kinds of claims? Please
just stop and think about it for a few minutes...
DSD magically converts 48 KHz PCM recordings into something that the DSD
priests want me to believe can't exist in PCM? Is there anything
potentially erroneous with that "logic"?
So PCM is chickenshit until it gets DSD'd and then all of a sudden it's
chicken soup?
I htink it's the other way around. DSD is chicken soup until it gets PCM'd.



www.monsterisland.com
Mike Caffrey
2003-10-19 23:28:42 UTC
Permalink
Post by Arny Krueger
Post by Kurt Albershardt
1) Advanced editing is not possible. The Pyramix system first
converts the DSD files to PCM and then does the editing. If there
is an advantage to DSD recording, then you may have lost it in the
conversion.
Open to debate, since Pyramix maintains the 2.8 mbit data rate of the
DSD stream and doesn't mangle its gentle filters.
I think we have to consider the mind set that the statement: "If there is an
advantage to DSD recording, then you may have lost it in the conversion."
refers to.
My understanding is that some people believe that there are some
undefinable, immeasurable, currently unquantifiable *something(s)* that make
DSD recordings sound better than traditional PCM with a similar or greater
bitrate.
I've certainly run into these people on Usenet and seen articles in other
media that seem to say to the same basic thing(s).
Since "gentle filters" actually exist, can be analyzed, quantified, and
found to disagree with this belief about the alleged superior sound of DSD,
it *can't* be what these people are talking about.
BTW, one context in which I find these kinds of beliefs about the
superiority of DSD is PCABX. I've suggested to DSD advocates that the output
of a SACD player be digitized at 24/192, downsampled to 16/44, and the two
be compared.
The response is that re-quantizing DSD into any form of PCM destroys a
significant sound quality advantage of DSD.
Ironically, some of these same people go on for zillions of posts on Usenet
about the improved sound quality of DSD recordings that are now known to
have been made from 48 KHz PCM recordings but...
Where should I go to read some of these posts?

Would that last part depend on how is was mixed? 48k PCM track mixed in
the analog domain to a 2 tracks DSD recorder could easily sound better
than mixed to 2 tracks of PCM.



www.monsterisland.com
Continue reading on narkive:
Loading...