IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Jitter, Linearity ... part of orig Red-Book specs?
hollowman
post Sep 23 2011, 11:35
Post #1





Group: Members
Posts: 81
Joined: 24-March 07
Member No.: 41808



The audio community, at least as far as the pop magazines like Stereophile and Audio are concerned, started measuring and tracking (TTBMK) linearity and jitter several years after the consumer-version CDDA format (aka Red Book CD) appeared. I think I first encountered linearity in 1987 and jitter in 1990 (rough estimates on both).

I don't know if both of these parameters were specified in the orig. "Red Book" spec. sheet. Anyone know? If not part of orig. spec, what led to the "discovery" of each? Were they important (being tracked, measured) in the pro (recording) arena before consumer stuff?

Seems like impulse response is the latest of these Johnny-come-latelys -- i.e., wrt the infamous pre-ring and minimum-phase (or apodizing?) filtration. True ... a lot of it (i.e, the spec Zeitgeist) is marketing hype -- certainly, linearity was. Maybe jitter was evolutionary, rather than revolutionary. IAC, if digital indeed has a long sonic way to go before it better simulates the natural acoustic, what is/are the best scientific method(s) that lead to the knowledge-discovery of these new "parameterized" phenomenon ... phenomenon in Nature that importantly affect and improve digital state-of-the-art?
Go to the top of the page
+Quote Post
DVDdoug
post Sep 24 2011, 00:58
Post #2





Group: Members
Posts: 2441
Joined: 24-August 07
From: Silicon Valley
Member No.: 46454



There is no jitter on the disc itself.* The data on the disc is just a sequence of numbers. I'm not an expert on the Redbook, but I doubt there is anything about of the quality of those numbers, or the quality of the playback device.

The playback drive can generate jitter, but it's virtually NEVER A PROBLEM. I've never heard of "linearity" being a audible problem either (in a properly functioning CD player).

There is a bit of information on jitter (and other nonsense) in Ethan Winer's Audiophoolery article.

If you search HydrogenAudio you can find more discussion of jitter and why it doesn't matter. Here is a good thread to start with.... Check-out what Arnold B. Krueger has to say... He's actually built a jigger-generating machine!

QUOTE
...pop magazines like Stereophile and Audio are concerned...
The writers in these publications talk about a lot of things that they can't hear in blind listening tests. wink.gif You can easily measure things that you can't hear. I've got an oscilloscope that measures up to 100MHz, but I can't hear that high! biggrin.gif


* There could be jitter in the data on the disc if the analog-to-digital converter(s) used for recording were jittery. But, that doesn't have anything to do with the physical disc and I don't think it has anything to do with the standards. And, with a music disc you'd never know since there is no non-jittery master to compare to!

This post has been edited by DVDdoug: Sep 24 2011, 01:20
Go to the top of the page
+Quote Post
hollowman
post Sep 24 2011, 02:04
Post #3





Group: Members
Posts: 81
Joined: 24-March 07
Member No.: 41808



I should have been a bit more specific ... e.g., how PCM data is affected by jitter or linearity "problems".

About jitter .... I'm not certain it is "NEVER A PROBLEM". But as I noted, it's concomitantly marketing jargon. Like "low carb", etc.
IIRC, 1-bit DAC (aka MASH or Bitstream) measured v. well WRT "linearity" -- better than than R2R or multibit DACs. IAC, and FWIW, you can search Stereophile for discussion on linearity ... it's hype did not last long as R. Harley himself noted, in March 1995, that ... "Low-level linearity was touted a few years back as the definitive measure of digital-processor performance. While good linearity is a high engineering goal, it doesn't guarantee good sound."

Agreed about optical disc being a source of jitter ... some argue that laser-based data-read process is effectively an analog phenomenon.

This post has been edited by db1989: Sep 25 2011, 11:17
Reason for edit: deleting pointless fullquote of prior post
Go to the top of the page
+Quote Post
Wombat
post Sep 24 2011, 03:03
Post #4





Group: Members
Posts: 950
Joined: 7-October 01
Member No.: 235



QUOTE (hollowman @ Sep 24 2011, 03:04) *
I should have been a bit more specific ... e.g., how PCM data is affected by jitter or linearity "problems".

This "more specific" question has not much more to do with Red-Book. I donīt know of any definition in the specs defining the amount of jitter in any way.

QUOTE (hollowman @ Sep 24 2011, 03:04) *
IAC, and FWIW, you can search Stereophile for discussion on linearity

Since you like reading at Stereophile you may ask over there and bring some evidence with hard facts over here.




Go to the top of the page
+Quote Post
Canar
post Sep 24 2011, 03:36
Post #5





Group: Super Moderator
Posts: 3327
Joined: 26-July 02
From: princegeorge.ca
Member No.: 2796



All the articles I've read on the perceptibility of jitter put the human perception threshold several orders of magnitude (on the order of 10us, unless I'm completely off-base) longer than the jitter specified by most audio devices (on the order of 10ns). This would lead me to conclude that jitter is pretty much a non-issue.


--------------------
∑:<
Go to the top of the page
+Quote Post
hollowman
post Sep 24 2011, 20:07
Post #6





Group: Members
Posts: 81
Joined: 24-March 07
Member No.: 41808



QUOTE (hollowman @ Sep 23 2011, 03:35) *
I don't know if both of these parameters were specified in the orig. "Red Book" spec. sheet. Anyone know? If not part of orig. spec, what led to the "discovery" of each? Were they important (being tracked, measured) in the pro (recording) arena before consumer stuff?
Here's another tracked parameter that came "after-the-fact" ...
"Noise modulation" was suggested by an AES paper given by Dr. Richard Cabot, Principal Engineer at Audio Precision. The measurement looks at how a digital processor's noise floor shifts (modulates) with changing input level. Research by Louis Fielder at Dolby Labs suggests that spectral shifts as small as 2dB are audible.

"Noise Modulation in Digital Audio Devices," presented at the 90th AES Convention, Paris, 1991.
http://www.aes.org/e-lib/browse.cfm?elib=5672
http://www.dolby.com/uploadedFiles/English...37_ac3-flex.pdf
Go to the top of the page
+Quote Post
saratoga
post Sep 24 2011, 20:16
Post #7





Group: Members
Posts: 4718
Joined: 2-September 02
Member No.: 3264



QUOTE (hollowman @ Sep 23 2011, 21:04) *
Agreed about optical disc being a source of jitter ... some argue that laser-based data-read process is effectively an analog phenomenon.


Since digital data can only be stored in real, physical objects, and lasers and disks are objects, this is trivially true. The process of reading any stored data involves taking a series of analog measurements, processing them to extract the 1s and 0s, and then applying error correction and hash checking to make sure that the data was unchanged from storage (or to error out if it has been corrupted).

If someone "argues" to you that reading a digital value involves an analog measurement they're probably just babbling about words they don't understand.
Go to the top of the page
+Quote Post
hollowman
post Sep 24 2011, 22:31
Post #8





Group: Members
Posts: 81
Joined: 24-March 07
Member No.: 41808



QUOTE (saratoga @ Sep 24 2011, 12:16) *
If someone "argues" to you that reading a digital value involves an analog measurement they're probably just babbling about words they don't understand.
And if we take it the next level down -- quantum, or QM -- everything becomes hunky-dory discrete again! But HUP says we can't be exact with our measurements ... so back to close approx. analogs again?? Enough digression.
Speaking of QM, did anyone catch this in Scientific American (June 2011):
QUOTE
Living in a Quantum World ... Quantum mechanics is not just about teeny particles. It applies to things of all sizes: birds, plants, maybe even people
Full-length article not avail. online. Check you libr.
How does this relate to THIS topic or audio? See synopsis.

This post has been edited by hollowman: Sep 24 2011, 22:35
Go to the top of the page
+Quote Post
cliveb
post Sep 25 2011, 08:40
Post #9


WaveRepair developer


Group: Developer
Posts: 824
Joined: 28-July 04
Member No.: 15845



QUOTE (Canar @ Sep 24 2011, 03:36) *
All the articles I've read on the perceptibility of jitter put the human perception threshold several orders of magnitude (on the order of 10us, unless I'm completely off-base) longer than the jitter specified by most audio devices (on the order of 10ns). This would lead me to conclude that jitter is pretty much a non-issue.

I think you'll find that you're about 3 orders of magnitude out. The classic Benjamin & Gannon paper concluded that uncorrelated jitter was inaudible below about 10nS on pure tones, and about 20nS on music programme.
A later study by Ashihara et al concluded that uncorrelated jitter was inaudible on music programme below about 250nS.
Compare this to typical jitter figures of modern digital transports, which are usually a few hundred pS or less.

Note however that uncorrelated jitter is considerably less noticable than correlated, and most mechanisms that introduce jitter in real-world playback devices generate correlated jitter. I'm not aware of any studies that have measured the threshold of audibility for that - such an experiment would be very complex, since the characteristics of the correlation will affect audibility, and the combination of playback device + programme material will have its own jitter signature.

To the OP: I can't see how the redbook could specify anything to do with jitter. The eventual effect of jitter is an increase in noise and distortion. Therefore specifying jitter limits here is akin to requiring that the playback device must have noise+distortion below a certain level - and what's more, the noise+distortion of pretty much any subsequent analogue circuitry will swamp that from any reasonable level of jitter at the DAC.
Go to the top of the page
+Quote Post
hollowman
post Sep 25 2011, 21:27
Post #10





Group: Members
Posts: 81
Joined: 24-March 07
Member No.: 41808



QUOTE (cliveb @ Sep 25 2011, 00:40) *
To the OP: I can't see how the redbook could specify anything to do with jitter.
I can't seem to locate any definitive refs to Sony/Philips Red-Book (either orig. or subsequent revisions). If anyone can either point to a source or directly send me copies (PM me if you have them), I could try to re/deconstruct the orig. sci/eng. methodology ... how the orig/subsequent parameters were chosen.
Another way to approach the general issue is to investigate similar and/or related technologies ... so one could re-ask the OP question as:

Jitter, Linearity ... part of orig. PCM specs?
Jitter, Linearity ... part of orig DSD/SACD specs?
Jitter, Linearity ... part of orig. DVD specs?
Jitter, Linearity ... part of orig BluRay specs?

An example of an acceptable document may be something similar to NISTs:
Stability Comparison of Recordable Optical Discs—A Study of Error Rates in Harsh Conditions
http://nvl.nist.gov/pub/nistpubs/jres/109/5/j95sla.pdf

Which states:
QUOTE
2.1 Key Measured Parameters:
Jitter:
Jitter is the temporal variation or imprecision in a signal compared to an ideal reference clock. It is a measure of how well defined the pits and lands of a disc are. For CD discs, jitter is defined in nanometers (nm), and the CD specification states that jitter should not exceed 35 nm. For DVD recordable discs, jitter is defined in percentage points, and should not exceed 9%.
Note that NIST seems to be wrong here in defining jitter by optical wavelength (nm).
Go to the top of the page
+Quote Post
greynol
post Sep 25 2011, 21:51
Post #11





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



Seeing that a CD is played in CLV, I don't see much of an issue, except to say that jitter at the CD level is different than jitter at the DAC level due to EFM and CIRC.


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
saratoga
post Sep 25 2011, 22:59
Post #12





Group: Members
Posts: 4718
Joined: 2-September 02
Member No.: 3264



QUOTE (hollowman @ Sep 24 2011, 17:31) *
QUOTE (saratoga @ Sep 24 2011, 12:16) *
If someone "argues" to you that reading a digital value involves an analog measurement they're probably just babbling about words they don't understand.
And if we take it the next level down -- quantum, or QM -- everything becomes hunky-dory discrete again!


I have no idea what you're trying to say.

QUOTE (hollowman @ Sep 24 2011, 17:31) *
But HUP says we can't be exact with our measurements ... so back to close approx. analogs again?? Enough digression.


I can assure you as an optical engineer that the HUP is incorporated into the design of any optical system as it is one of the fundamental tenets of optical design. Seeing as optical read out from a CD does not rely on spectral information nor the position of individual particles, there is no real connection to the HUP. Again, I think if someone brings this up in this context they're just throwing words out to try and sound smart. You should be ignoring these people.

QUOTE (hollowman @ Sep 24 2011, 17:31) *
How does this relate to THIS topic or audio? See synopsis.


Without clicking that link I'm going to guess that it does not relate to this topic. If it did relate, you probably would have just said so.
Go to the top of the page
+Quote Post
Rotareneg
post Sep 26 2011, 01:01
Post #13





Group: Members
Posts: 188
Joined: 18-March 05
From: Non-Euclidean
Member No.: 20701



QUOTE (hollowman @ Sep 25 2011, 15:27) *
QUOTE
2.1 Key Measured Parameters:
Jitter:
Jitter is the temporal variation or imprecision in a signal compared to an ideal reference clock. It is a measure of how well defined the pits and lands of a disc are. For CD discs, jitter is defined in nanometers (nm), and the CD specification states that jitter should not exceed 35 nm. For DVD recordable discs, jitter is defined in percentage points, and should not exceed 9%.
Note that NIST seems to be wrong here in defining jitter by optical wavelength (nm).


They're just saying the ends of the pits on the disc shouldn't vary more than 35 nanometers from their "correct" location, nothing strange about that.
Go to the top of the page
+Quote Post
Canar
post Sep 26 2011, 04:36
Post #14





Group: Super Moderator
Posts: 3327
Joined: 26-July 02
From: princegeorge.ca
Member No.: 2796



QUOTE (cliveb @ Sep 25 2011, 00:40) *
I think you'll find that you're about 3 orders of magnitude out.
More importantly though, I was close enough to the distance between the audible/usual split. Thanks for interceding with hard numbers!


--------------------
∑:<
Go to the top of the page
+Quote Post
Northpack
post Sep 26 2011, 08:31
Post #15





Group: Members
Posts: 455
Joined: 16-December 01
Member No.: 664



Just for clarification: the jitter that is actually contained on the disc by small variations of the pits' distance from each other, is not the kind of jitter which goes through the dac. The signal the laser reads from the disc is buffered so that it won't contain any kind of jitter from the medium itself.

From the user's perspective jitter in digital media in generally a non-issue.

This post has been edited by Northpack: Sep 26 2011, 08:43
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 21st April 2014 - 16:33