Go Back   Cockos Incorporated Forums > REAPER Forums > Recording Technologies and Techniques

Reply
 
Thread Tools Display Modes
Old 11-25-2012, 08:51 AM   #1
tspring
Human being with feelings
 
tspring's Avatar
 
Join Date: Jan 2007
Location: Eastern shore of Maryland, USA
Posts: 1,484
Default Should I reaaly care about jitter in audio streams

I see a lot of hardware ads that emphasize jitter reduction technology. I understand that jitter does introduce distortion, but it is not clear to me that jitter introduces audible distortion into sound being processed by modern hardware. So, should I care about jitter?

T
tspring is offline   Reply With Quote
Old 11-25-2012, 11:03 AM   #2
JHughes
Banned
 
Join Date: Aug 2007
Location: Too close to Charlotte, NC
Posts: 3,554
Default

Jitter matters a great deal! But, only during A-D and D-A conversions.
JHughes is offline   Reply With Quote
Old 11-25-2012, 11:19 AM   #3
EpicSounds
Human being with feelings
 
EpicSounds's Avatar
 
Join Date: Jul 2009
Posts: 7,570
Default

Jitter is not a concern until you connect additional digital hardware.

The problem comes from when the two clocks are not in sync. I could sound like distortion but more likely it will be progressively worse and worse clicks or gaps in the audio.

Anything you buy now (even Behringer!) can sync just fine if it's set right.
__________________
REAPER Video Tutorials, Tips & Tricks and more at The REAPER Blog
EpicSounds is offline   Reply With Quote
Old 11-25-2012, 01:19 PM   #4
JHughes
Banned
 
Join Date: Aug 2007
Location: Too close to Charlotte, NC
Posts: 3,554
Default

Please don't confuse the issue. Clocks not being in sync is a sync issue, not a jitter issue. If jitter is so bad that it causes a sync issue then something is broken. Besides, you're talking about D-D operations which are totally unaffected by jitter.

Bob Katz dispelled the myths way back in 2007.
http://www.digido.com/articles-and-d...15-jitter.html

Also, external clocking of devices is no guarantee of better quality and can lead to worse jitter results than using the internal clock during the A-D and D-A conversions.
JHughes is offline   Reply With Quote
Old 11-25-2012, 04:49 PM   #5
tspring
Human being with feelings
 
tspring's Avatar
 
Join Date: Jan 2007
Location: Eastern shore of Maryland, USA
Posts: 1,484
Default

Jhughs, thanks for the link to the Bob Katz article. Based on what Katz has to say, it seems that controlling jitter when burning a CD for mastering is probably more critical than any other step. He also says that "Many computer-based digital audio cards have quite high jitter, which makes listening through them a variable experience." So I suppose that I now need to learn a bit more about how to interpret measurements of jitter.

T
tspring is offline   Reply With Quote
Old 11-25-2012, 07:22 PM   #6
JHughes
Banned
 
Join Date: Aug 2007
Location: Too close to Charlotte, NC
Posts: 3,554
Default

Quote:
Originally Posted by tspring View Post
Jhughs, thanks for the link to the Bob Katz article.
You're welcome.

Quote:
Originally Posted by tspring View Post
Based on what Katz has to say, it seems that controlling jitter when burning a CD for mastering is probably more critical than any other step.
That's only when making a GLASS MASTER, and is beyond your control.

The most critical step is when recording because that's when the A-D conversion takes place. Any jitter you get during a recording is permanent and can't be undone, unlike the D-A process which is variable.
JHughes is offline   Reply With Quote
Old 11-26-2012, 08:48 AM   #7
drtedtan
Human being with feelings
 
drtedtan's Avatar
 
Join Date: Jul 2010
Posts: 387
Default

Quote:
Originally Posted by JHughes View Post
Also, external clocking of devices is no guarantee of better quality and can lead to worse jitter results than using the internal clock during the A-D and D-A conversions.
Looks like I'm late to the pary, but I want to say that JHughes is spot on here. Jitter is cumulative, e.g., the jitter of the Master clock is added to the jitter of the Slave clock, resulting in worse jitter than just using the Slave device's clock alone.
drtedtan is offline   Reply With Quote
Old 11-27-2012, 06:38 PM   #8
Audiosyncrasy
Human being with feelings
 
Audiosyncrasy's Avatar
 
Join Date: Aug 2010
Location: Seattle, WA
Posts: 211
Default

Quote:
Originally Posted by drtedtan View Post
Looks like I'm late to the pary, but I want to say that JHughes is spot on here. Jitter is cumulative, e.g., the jitter of the Master clock is added to the jitter of the Slave clock, resulting in worse jitter than just using the Slave device's clock alone.
If I may ask a clarifying question: based on Bob's article, could one conclude that the best way to minimize clock jitter while recording (i.e. A/D conversion) is to sync the converters to their internal clock?

In other words, an internal clock (crystal oscillator) will be less jittery than if the unit were synched to an external clock (PLL). Correct? Perhaps, I'm oversimplifying the issue . . .
__________________
How can you tell the stage is level? . . . The drummer drools out of both sides of his mouth.
Audiosyncrasy.com - Custom drum tracks for your project.
Audiosyncrasy is offline   Reply With Quote
Old 11-27-2012, 07:25 PM   #9
JHughes
Banned
 
Join Date: Aug 2007
Location: Too close to Charlotte, NC
Posts: 3,554
Default

Quote:
Originally Posted by Audiosyncrasy View Post
could one conclude that the best way to minimize clock jitter while recording (i.e. A/D conversion) is to sync the converters to their internal clock?
Yes. Of course, testing is the only way to be certain.

An analogy: Much info on the net says that a CD will be less jittery if burned at 2X instead of a higher speed. Well, that is not true in my experience. Maybe it was in Windows 95 days when burners maxed out at 8X or whatever, but based on my tests it's not true. But I don't guess, I test, mostly with Plextools.
JHughes is offline   Reply With Quote
Old 11-28-2012, 09:14 AM   #10
drtedtan
Human being with feelings
 
drtedtan's Avatar
 
Join Date: Jul 2010
Posts: 387
Default

JHughes beat me to the punch again! And once again he is spot on.

My understanding is that, even though it would produce less jitter to sync to the internal clock, it may actually sound better syncing to the external clock (or not), so try it both ways and see which way sounds better to you.
drtedtan is offline   Reply With Quote
Old 11-29-2012, 08:36 AM   #11
tspring
Human being with feelings
 
tspring's Avatar
 
Join Date: Jan 2007
Location: Eastern shore of Maryland, USA
Posts: 1,484
Default Related question re wordclock cables

Lets say that you are using a setup where you have a Focusrite Octapre hooked up to a Focusrite Safire Liquid 56 using ADAT to transmit the digitized audio from the Octapre to the Liquid 56. Setup suggests using a coaxial cable to send the wordclock timing from one unit to the other.

From what I have read the wordclock cable is the same 75 ohm coax used for video. Yet if I look on the Sweetwater web site I see wordclock cables that cost $40 - $50. Is there any reason to plunk down $50 instead of the less than $10 for the wordclock cables I see on Amazon? The only way I can see that the wordclock cable would affect jitter would be if stray capacitance or inductance in cable or connectors distorted edges of clock pulses - and that somehow resulted in impaired clocking of the ADC at one end or the other. Are wordclock cables actually designed differently from standard BNC connected 75 ohm coax cables?

T
tspring is offline   Reply With Quote
Old 11-29-2012, 05:52 PM   #12
Audiosyncrasy
Human being with feelings
 
Audiosyncrasy's Avatar
 
Join Date: Aug 2010
Location: Seattle, WA
Posts: 211
Default

Quote:
Originally Posted by tspring View Post
Lets say that you are using a setup where you have a Focusrite Octapre hooked up to a Focusrite Safire Liquid 56 using ADAT to transmit the digitized audio from the Octapre to the Liquid 56. Setup suggests using a coaxial cable to send the wordclock timing from one unit to the other.

From what I have read the wordclock cable is the same 75 ohm coax used for video. Yet if I look on the Sweetwater web site I see wordclock cables that cost $40 - $50. Is there any reason to plunk down $50 instead of the less than $10 for the wordclock cables I see on Amazon? The only way I can see that the wordclock cable would affect jitter would be if stray capacitance or inductance in cable or connectors distorted edges of clock pulses - and that somehow resulted in impaired clocking of the ADC at one end or the other. Are wordclock cables actually designed differently from standard BNC connected 75 ohm coax cables?

T
As far as I understand it, the question should be: am I getting $30 to $40 additional VALUE when I buy the over-marketed WC cables?

If you're getting a performance increase of a few percent, it wouldn't be worth it to me. I find that with drum gear all the time. "Buy these crazy-expensive drums made out of wood only found next to a waterfall deep in the rain forests of Brazil!" At that point, they may sound great, but I'm paying more for the rarity of the wood than for the sound quality increase.
__________________
How can you tell the stage is level? . . . The drummer drools out of both sides of his mouth.
Audiosyncrasy.com - Custom drum tracks for your project.
Audiosyncrasy is offline   Reply With Quote
Old 11-29-2012, 08:10 PM   #13
tspring
Human being with feelings
 
tspring's Avatar
 
Join Date: Jan 2007
Location: Eastern shore of Maryland, USA
Posts: 1,484
Default

Quote:
Originally Posted by Audiosyncrasy View Post
As far as I understand it, the question should be: am I getting $30 to $40 additional VALUE when I buy the over-marketed WC cables?
No doubt you are correct. But I still would like to know why spending more might give better results. Otherwise I cannot judge anything. And with all the sales hype, sorting value from non-value is tricky.
tspring is offline   Reply With Quote
Old 11-30-2012, 08:17 AM   #14
Audiosyncrasy
Human being with feelings
 
Audiosyncrasy's Avatar
 
Join Date: Aug 2010
Location: Seattle, WA
Posts: 211
Default

Quote:
Originally Posted by tspring View Post
No doubt you are correct. But I still would like to know why spending more might give better results. Otherwise I cannot judge anything. And with all the sales hype, sorting value from non-value is tricky.
I don't know enough about the technical details of WC cables to give you a firm answer, but if it were me, I would just get the regular 75 Ohm-rated cable and be done with it.

Perhaps someone else can give you a better - more technical - answer
__________________
How can you tell the stage is level? . . . The drummer drools out of both sides of his mouth.
Audiosyncrasy.com - Custom drum tracks for your project.
Audiosyncrasy is offline   Reply With Quote
Old 11-30-2012, 08:33 AM   #15
drtedtan
Human being with feelings
 
drtedtan's Avatar
 
Join Date: Jul 2010
Posts: 387
Default

Quote:
Originally Posted by tspring View Post
Lets say that you are using a setup where you have a Focusrite Octapre hooked up to a Focusrite Safire Liquid 56 using ADAT to transmit the digitized audio from the Octapre to the Liquid 56. Setup suggests using a coaxial cable to send the wordclock timing from one unit to the other.
The two units can sync over the ADAT lightpipe cable just fine because the ADAT lightpipe protocol sends clock info along with the audio data stream. I doubt you will see a significant difference using word clock at all. If this were me, I would try the ADAT first. Then, if I didn't like what I was hearing, I would look into the word clock cables.

As for the different priced cables, I generally find that more expensive cables (within reason, e.g., not the cheap $10 ones) tend to last longer than cheaper cables, especially if you end up plugging and unplugging them quite a bit. If you do end up going the word clock cable route, keep in mind that you will also need a word clock terminator plug.
drtedtan is offline   Reply With Quote
Old 12-01-2012, 12:10 AM   #16
tspring
Human being with feelings
 
tspring's Avatar
 
Join Date: Jan 2007
Location: Eastern shore of Maryland, USA
Posts: 1,484
Default

Thanks for the info drtedtan.I'll take your advice. Much appreciated.
tspring is offline   Reply With Quote
Old 12-01-2012, 01:24 PM   #17
drtedtan
Human being with feelings
 
drtedtan's Avatar
 
Join Date: Jul 2010
Posts: 387
Default

You are welcome, tspring. I hope I helped.
drtedtan is offline   Reply With Quote
Old 12-02-2012, 06:39 AM   #18
uebertone
Human being with feelings
 
Join Date: Oct 2012
Posts: 21
Default

Stop worrying too much - especially about stuff you don't know whether you should worry about or not. In the end that all leads down the worry spiral where ultimately you start to worry about worrying about worrying whether you should start worrying about starting to worry about worrying about worrying.
If you have to ask whether you should worry about something you clearly worry too much.

If something should pose a problem you will eventually have the moment where you go "OK this is a problem". Now you are allowed to worry a bit (but not too much) about how to fix the problem, but usually I would first calmly consider whether it really is an actual problem and the problem isn't cause by you knowing that there is a problem.

tl;dr: video
uebertone is offline   Reply With Quote
Old 12-02-2012, 10:34 AM   #19
EnkelMagnus
Human being with feelings
 
Join Date: Feb 2010
Posts: 96
Default

Quote:
Originally Posted by JHughes View Post
An analogy: Much info on the net says that a CD will be less jittery if burned at 2X instead of a higher speed.
How is that even possible ?
EnkelMagnus is offline   Reply With Quote
Old 12-02-2012, 10:47 AM   #20
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 29,260
Default

Quote:
If something should pose a problem you will eventually have the moment where you go "OK this is a problem".
You mean like after its mastered and released, then you notice the problem and its too late to fix? The whole point is to catch issues before its too late to fix them. I don't disagree with your overall premise but it doesn't necessarily fit all situations.
karbomusic is offline   Reply With Quote
Old 12-02-2012, 11:00 AM   #21
JHughes
Banned
 
Join Date: Aug 2007
Location: Too close to Charlotte, NC
Posts: 3,554
Default

Quote:
Originally Posted by EnkelMagnus View Post
How is that even possible ?
If the CD is spinning faster then it's more of a challenge to burn the pits in the exact proper locations. But the problem has been sorted and CD mechanisms are optimized for the higher burn speeds. In the old days where it was a big deal going from 2X to 4X, this wasn't necessarily the case.
JHughes is offline   Reply With Quote
Old 12-02-2012, 11:11 AM   #22
JHughes
Banned
 
Join Date: Aug 2007
Location: Too close to Charlotte, NC
Posts: 3,554
Default

Quote:
Originally Posted by drtedtan View Post
The two units can sync over the ADAT lightpipe cable just fine because the ADAT lightpipe protocol sends clock info along with the audio data stream. I doubt you will see a significant difference using word clock at all. If this were me, I would try the ADAT first. Then, if I didn't like what I was hearing, I would look into the word clock cables.

As for the different priced cables, I generally find that more expensive cables (within reason, e.g., not the cheap $10 ones) tend to last longer than cheaper cables, especially if you end up plugging and unplugging them quite a bit. If you do end up going the word clock cable route, keep in mind that you will also need a word clock terminator plug.
I agree.
Lightpipe is more jittery than dedicated word clock, but the signal is reclocked at the receving end so it shouldn't matter.

About 75 ohm cables: A better cable will have better shielding and last longer too. It's also likely to have lower capacitance. The connectors are important too, because not only is the cable 75 ohm, but the connector must also be. The capacitance/inductance should also be constant along the length of the cable which can only be achieved with tight tolerances. So a $20 cable is likely to be better than a $5 cable, but not necessarily worse than a $50 cable.

Best you can do in the absence of esoteric test gear is look for a known good brand like Belden.
JHughes is offline   Reply With Quote
Old 12-02-2012, 11:48 AM   #23
EnkelMagnus
Human being with feelings
 
Join Date: Feb 2010
Posts: 96
Default

Quote:
Originally Posted by JHughes View Post
If the CD is spinning faster then it's more of a challenge to burn the pits in the exact proper locations.
As i understand it a CD isn't clocked from the actual disc but from the DAC thus making a jittery CD an impossibility. You could possibly get clicks and pops but absolutely not jitter,that is physically impossible if i understand it correctly.
EnkelMagnus is offline   Reply With Quote
Old 12-02-2012, 01:19 PM   #24
JHughes
Banned
 
Join Date: Aug 2007
Location: Too close to Charlotte, NC
Posts: 3,554
Default

A CD is burned and read using an analog process, it is jittery, and jitter causes problems.
http://web.archive.org/web/200903240...Q_2000_02.html

After that, the DAC in most CD players stinks and is also jittery.

Jitter will ruin your sound, assuming your monitoring system and listening skill are up to snuff, well before error correction will.
JHughes is offline   Reply With Quote
Old 12-02-2012, 01:29 PM   #25
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 29,260
Default

Just an FYI... I burned then ripped multiple CDs the last time this came up and they all nulled with the source WAV perfectly. That includes a burn and a rip FYI. The first time I tried I had an issue due to user error but all subsequent tests nulled everytime. Not that its relevant here, just noting.
karbomusic is offline   Reply With Quote
Old 12-02-2012, 01:46 PM   #26
JHughes
Banned
 
Join Date: Aug 2007
Location: Too close to Charlotte, NC
Posts: 3,554
Default

You're right, jitter has nothing to do with data nulling.
JHughes is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 06:29 PM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.