SACD vs. CD - an illustration of differences - Page 2

Last message on previous page:
Archived from groups: rec.audio.high-end (More info?)

uraniumcommittee@yahoo.com wrote:
> Stewart Pinkerton wrote:
>> On 10 Sep 2005 03:18:42 GMT, uraniumcommittee@yahoo.com wrote:
>>
>> >Stewart Pinkerton wrote:
>> >
>> >> Indeed so, and it's fair to point out that the SNR of properly
>> >> dithered CD is more like 93dB, although without the dynamically
>> >> varying noise floor of SR. It's also true that the 15ips tapes will
>> >> likely give better results than 30ips in the bass, although of course
>> >> not comparable with digital in this regard.
>> >
>> >Why is that? Why better bass at 15ips?
>>
>> Because the spedd of the tape past the head gap causes a 'wobble' in
>> the bass response, which is typically around 20Hz for 15ips, but of
>> course shifts to the more sensitve 40Hz at 30ips. This is well-known
>> in the industry, and is why some recording engineers still prefer to
>> worrk at 15ips and trade off some (ultrasonic) treble extension for a
>> smoother bass response.
>>
>> Note that the dynamic range is essentially the same in each case -
>> more speed does *not* buy you better resolution.
>
> Are you going to sit there and tell me that 'dynamic range' is
> equivalent to 'resolution'? I almost fell out of my seat. 'Resolution'
> goes up DIRECTLY and LINEARLY with speed. 'Resolution' is the limit of
> the frequencies recordable.

You are simply wrong here. Resolution is the ability to extract details,
and is intimately related to the noise floor, which is also intimately
related to dynamic range.

Ever wonder what people mean when they say a certain DAC (digital to
analog converter) have 16 bits of resolution? No, they are not talking
about the limit of the frequencies that can be reproduced.

It's OK to not understand these technical terms, but it is rather
foolish of you to display your ignorance with such arrogance and
apparent conviction. It severely weakens your position.

Archived from groups: rec.audio.high-end (More info?)

On 11 Sep 2005 02:06:59 GMT, uraniumcommittee@yahoo.com wrote:

>Stewart Pinkerton wrote:
>> On 10 Sep 2005 03:18:42 GMT, uraniumcommittee@yahoo.com wrote:
>>
>> >Stewart Pinkerton wrote:
>> >
>> >> Indeed so, and it's fair to point out that the SNR of properly
>> >> dithered CD is more like 93dB, although without the dynamically
>> >> varying noise floor of SR. It's also true that the 15ips tapes will
>> >> likely give better results than 30ips in the bass, although of course
>> >> not comparable with digital in this regard.
>> >
>> >Why is that? Why better bass at 15ips?
>>
>> Because the speed of the tape past the head gap causes a 'wobble' in
>> the bass response, which is typically around 20Hz for 15ips, but of
>> course shifts to the more sensitve 40Hz at 30ips. This is well-known
>> in the industry, and is why some recording engineers still prefer to
>> worrk at 15ips and trade off some (ultrasonic) treble extension for a
>> smoother bass response.
>>
>> Note that the dynamic range is essentially the same in each case -
>> more speed does *not* buy you better resolution.
>
>Are you going to sit there and tell me that 'dynamic range' is
>equivalent to 'resolution'?

Yes, that is universally agreed - except apparently at your house.

> I almost fell out of my seat. 'Resolution'
>goes up DIRECTLY and LINEARLY with speed.

You clearly have noi idea of standard definitions. Resolution is a
measure of the smallest *amplitude* which can be uniquely identified,
i.e. just detectable above the noise floor.

> 'Resolution' is the limit of the frequencies recordable.

No, that's just an increase in the information content, which is a
function of resolution and bandwidth. Please learn the basics of
information theory.

>The higher the speed, the higher the
>frequencies recordable. 'Dynamic range' also goes up with speed, but
>for different reasons. 'Dynamic range' goes up because of the increase
>in recording area. 'Resolution' (high-frequency limit) goes up with
>increased speed simply because of mechanical factors. The faster the
>tape is moving against the head, the higher the frequencies that can be
>written and read. That's why VCR tapes are wrapped around a spinning
>head, to enable the very high frequencies of video to be written and
>read.

Agreed - but this has nothing to do with resolution, which is *much*
more a function of track width than tape speed.
--

Stewart Pinkerton | Music is Art - Audio is Engineering

Archived from groups: rec.audio.high-end (More info?)

Chung wrote:

>
> You are simply wrong here. Resolution is the ability to extract details,
> and is intimately related to the noise floor, which is also intimately
> related to dynamic range.

'Resolution' is not really an audio term. 'Dynamic range' and
'frequency response' are limited by tape/head speed. 'Signal/noise
ratio' also benefits from higher tape/head speed and wider tape.

> Ever wonder what people mean when they say a certain DAC (digital to
> analog converter) have 16 bits of resolution? No, they are not talking
> about the limit of the frequencies that can be reproduced.
>
> It's OK to not understand these technical terms, but it is rather
> foolish of you to display your ignorance with such arrogance and
> apparent conviction. It severely weakens your position.

See above. You are the one who has got the terminology wrong.

Archived from groups: rec.audio.high-end (More info?)

uraniumcommittee@yahoo.com wrote:
> Chung wrote:
>
>>
>> You are simply wrong here. Resolution is the ability to extract details,
>> and is intimately related to the noise floor, which is also intimately
>> related to dynamic range.
>
> 'Resolution' is not really an audio term.

Really? Wonder why there is a Hi-Rez forum at the Audio Asylum:

http://www.audioasylum.com/forums/hirez/bbs.html

Ever heard "resolution" of a DAC (or ADC)? Or "CD's have 16 bits of
resolution"?

And if indeed it is not really an audio term, why did you say
"'Resolution' is the limit of the frequencies recordable"?

>'Dynamic range' and
> 'frequency response' are limited by tape/head speed. 'Signal/noise
> ratio' also benefits from higher tape/head speed and wider tape.

Irrelvant in the discussion of what resolution means.

>
>> Ever wonder what people mean when they say a certain DAC (digital to
>> analog converter) have 16 bits of resolution? No, they are not talking
>> about the limit of the frequencies that can be reproduced.
>>
>> It's OK to not understand these technical terms, but it is rather
>> foolish of you to display your ignorance with such arrogance and
>> apparent conviction. It severely weakens your position.
>
> See above. You are the one who has got the terminology wrong.

See above, and see posts from others. Could it be that you are the only
one to get the terminology wrong?

It's one thing to not know what a technical term mean, but it is quite
another thing to insist that everyone else is wrong...

Archived from groups: rec.audio.high-end (More info?)

dpierce@cartchunk.org wrote:
> uraniumcommittee@yahoo.com wrote:
> > Chung wrote:
> > > You are simply wrong here. Resolution is the ability to extract details,
> > > and is intimately related to the noise floor, which is also intimately
> > > related to dynamic range.
> >
> > 'Resolution' is not really an audio term.
>
> Really? You are, of course, certain of this?

I thought we were talking about analogue audio tape.
'Resolution' is not used in the measurements for such a recording
system.

Archived from groups: rec.audio.high-end (More info?)

"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in message
newsg1ng30fq8@news2.newsguy.com...
> On 10 Sep 2005 16:47:49 GMT, "Harry Lavo"
> <hlavo@comcast.net> wrote:
>
>>"Stewart Pinkerton" <patent3@dircon.co.uk> wrote in
>>message
>>newsftk280vom@news2.newsguy.com...
>>> On 9 Sep 2005 02:40:52 GMT, uraniumcommittee@yahoo.com
>>> wrote:
>>>
>>>>Freedom from noise is not resolution, Stewart, and you
>>>>know it.
>>>
>>> If you don't understand that they are inextricably
>>> interdependent in a
>>> linear system such as a tape recording, then you have no
>>> business
>>> joining a technical debate.
>>>
>>>>The speed of the tape is the limiting factor. 30ips is
>>>>damned fast.
>>>
>>> The speed of the tape has little relevance to
>>> resolution, and in fact
>>> is notoriously destructive of bass accuracy. I'm sure
>>> Iain will be
>>> happy to enlighten you regarding 'head wobble'.
>>>
>>This is true, but if it were the only factor we'd all be
>>recording at 3 3/4.
>
> No Harry, because tape speed *does* have relevance to
> treble
> extension. The key is finding a balance, so some still
> prefer 15ips to
> 30ips - and not just to economise on tape!
>
>>Professional machines used a lot of physics and
>>engineering in the tape path
>>and head designs to keep this problem under enough control
>>for it to be
>>sonically minor. I have much worse on my old Teac 7030SL
>>at 15ips because
>>it was essentially a souped-up 7 1/2ips design.
>
>>The main advantage of a 30ips machine from a practical
>>standpoint is a much
>>higher saturation point in the high frequencies, as well
>>as extended high
>>frequencies. This allows for recording at higher
>>practical levels when
>>there are loud high-frequency overtones present without
>>the sound becoming
>>in any way constricted or distorted, and without the need
>>for Dolby.

I see this from a totally different perspective.
The use of Dolby SR allows one to *reduce*
the recording level, and thus tape distortion,
while maintaining a good SNR with improved
headroom.

Your point about "loud high-frequency overtones"
is an interesting one. You are probably aware that
it is not possible to record HF signals at peak level
to analogue tape at any tape speed (even 30 ips)
But, as such high level HF signals do not occur in
music anyway, this is not a limiting factor.

The highest note on a concert grand piano is C8
at a frequency of 4 168Hz. So the fifth harmonic
of this note is 20 840Hz. Compared with the
fundamental, this harmonic (musicians use the
term "partial" ) is about -50dB


>
> Of course, when you switch to the ubiquitous 24/96
> digital, all these
> delicate setup problems simply disappear! :-)

Surely you have not failed to notice remarkable similarities
in the transport of a professional analogue tape recorder
and a professional DASH multitrack. If setting up the latter
is not "delicate" then pray tell me what is:-)

Do you have experience of this, Stewart?

Cordially,
Iain

Archived from groups: rec.audio.high-end (More info?)

On 13 Sep 2005 03:38:56 GMT, uraniumcommittee@yahoo.com wrote:

>dpierce@cartchunk.org wrote:
>> uraniumcommittee@yahoo.com wrote:
>> > Chung wrote:
>> > > You are simply wrong here. Resolution is the ability to extract details,
>> > > and is intimately related to the noise floor, which is also intimately
>> > > related to dynamic range.
>> >
>> > 'Resolution' is not really an audio term.
>>
>> Really? You are, of course, certain of this?
>
>I thought we were talking about analogue audio tape.
>'Resolution' is not used in the measurements for such a recording
>system.

Yes, it is, but it's commonly referred to as the noise floor in this
case. Same thing, different terminology.

--

Stewart Pinkerton | Music is Art - Audio is Engineering

Archived from groups: rec.audio.high-end (More info?)

<uraniumcommittee@yahoo.com> wrote in message
newsg83km02te0@news1.newsguy.com...
> Stewart Pinkerton wrote:
>> On 11 Sep 2005 22:19:40 GMT, uraniumcommittee@yahoo.com
>> wrote:
>>
>> >Chung wrote:
>> >
>> >>
>> >> You are simply wrong here. Resolution is the ability
>> >> to extract details,
>> >> and is intimately related to the noise floor, which is
>> >> also intimately
>> >> related to dynamic range.
>> >
>> >'Resolution' is not really an audio term.
>>
>> You mean, as in 'high resolution digital formats like
>> DVD-A and SACD',
>> that kind of thing? And let's not forget that old
>> classic - 'vinyl has
>> infinite resolution'..........
>>
>> > 'Dynamic range' and
>> >'frequency response' are limited by tape/head speed.
>> >'Signal/noise
>> >ratio' also benefits from higher tape/head speed and
>> >wider tape.
>>
>> Shame that you don't understand that dynamic range and
>> SNR are the
>> same thing - and that they are mostly dependant on track
>> width.
>
> No, they're not. S/N ratio and dynamic range are two
> different things.
> Both are dependedent upon track width and tape/head
> (writing) speed.

Please explain how you think they differ.

In professional recording one tends to reserve the term
SNR for the specification of a machine, and the
term "dynamic range" for a performance, or recording.
Maybe this is confusing you. They are one and the
same thing.


With regard to your comments about analogue 30 ips.
It is fortunate that you have had a chance to listen to
this,
and appreciate how good it can sound. Next try 15 ips SR.
Then, if you want the ultimate analogue thrill, listen to a
35mm magnetic film track. High speed, superwide tracks.
Dolby SR is still used though.


Iain

Archived from groups: rec.audio.high-end (More info?)

Christopher Key <cjk32@cam.ac.uk> wrote:
> What sort of results does one get in a DBT comparing an SACD with a CD
> created from the analogue output of the SACD player using a suitably
> competent ADC?

You mean, comparing the analog output of both the SACD and CD layers,
after they've been converted back to.wavs? I haven't done that, but
I have done a relatively quick'n'dirty ABX with SACD output vs CD rip
of a Rolling Stones SACD track and got a 'no difference' result. I
should probably try again with more trials someday.
However, it's notable that the two layers
*measure* very close to each other in terms of RMS average level,
after they have been normalized to the same peak level. I should
probably also run a frequency analysis and see if they have been
mastered differently in any significant way...

I predict that if I were to try it with the Dark Side of the Moon
SACD, I'd have a better shot at telling them apart, since the two
layers there are *measurably* quite different -- they've obviously been
mastered differently.



--

-S

Archived from groups: rec.audio.high-end (More info?)

On 14 Sep 2005 02:58:30 GMT, uraniumcommittee@yahoo.com wrote:

>Stewart Pinkerton wrote:

>> Shame that you don't understand that dynamic range and SNR are the
>> same thing - and that they are mostly dependant on track width.
>
>No, they're not. S/N ratio and dynamic range are two different things.
>Both are dependedent upon track width and tape/head (writing) speed.

Shame that you don't understand that they are the *same* thing. The
difference betwen peak signal and noise floor (i.e. the SNR) *is* the
dynamic range of the system.

>> >> Ever wonder what people mean when they say a certain DAC (digital to
>> >> analog converter) have 16 bits of resolution? No, they are not talking
>> >> about the limit of the frequencies that can be reproduced.
>
>We are discussing analogue recording, not digital recording. There is
>no such term as 'resolution' in analogue recording.

Of course there is, normally expressed in dB down from the peak level.

Did you not read even the titles of the articles cited by Dick Pierce,
one of which referred to *high resolution* tape recording?

> In photography,
>resolution governs the smallest line-pair that is discernible, usually
>from a test target containing lines at progressively smaller distances
>from each other. It is expressed as lp/mm, which is essentially a
>frequency.

Actually, it's a distance. You are thinking of wavelength - which is
also a distance, and is only *related* to a frequency by the speed of
light.

In electronics, resolution is generally taken as the smallest
identifiable value, i.e. that which can just be reliably distinguished
from random noise.

--

Stewart Pinkerton | Music is Art - Audio is Engineering

Archived from groups: rec.audio.high-end (More info?)

"Steven Sullivan" <ssully@panix.com> wrote in message
newsgaork0295p@news3.newsguy.com...
> Christopher Key <cjk32@cam.ac.uk> wrote:
>> What sort of results does one get in a DBT comparing an SACD with a CD
>> created from the analogue output of the SACD player using a suitably
>> competent ADC?
>
> You mean, comparing the analog output of both the SACD and CD layers,
> after they've been converted back to.wavs? I haven't done that, but
> I have done a relatively quick'n'dirty ABX with SACD output vs CD rip
> of a Rolling Stones SACD track and got a 'no difference' result. I
> should probably try again with more trials someday.
> However, it's notable that the two layers
> *measure* very close to each other in terms of RMS average level,
> after they have been normalized to the same peak level. I should
> probably also run a frequency analysis and see if they have been
> mastered differently in any significant way...

Try again with some recordings done in DSD. Hiatt's "Master of Disaster".
Koukanen's "Blue Country Heart". Ivan Fisher's Dvorak Symphanies 8 & 9. And
particularly Alrfred Brendel and Sir Charles mzackerras's Mozart Piano
Concertos #9 and #25.

>
> I predict that if I were to try it with the Dark Side of the Moon
> SACD, I'd have a better shot at telling them apart, since the two
> layers there are *measurably* quite different -- they've obviously been
> mastered differently.

When you are done, make copies of the disk and distribute to others to
confirm your view, in order to make sure a single system doesn't obscure
results. And while you are at it, since you are an objectivist, describe
you equipment, test setup, test techniques, scoring outcome, and confidence
level, plus describe what differences you should be listening for (as
proclaimed by we subjectivists). You'll do the test with a proctor present
recording the scores, of course. And with a "xxx" detector test to make
sure you have given it an honest effort. Let us know when you have turned
this anecdote into a "real" objectivist test via this simple little home
trial.