Jitter

[Jerry Gerber]

I am soon to be using a Big Ben master clock which is syncing a Yamaha
DM2000 and a MOTU 2408. Analog comes into the DM2000 and the A/D conversion
is now clocked via Big Ben. So much the better. My concern is I have
decided to bring in Gigastudio sounds (2 separate computers generating
strings, winds and brass) via ADAT, rather than analog into the DM2000.
Though this will bypass the D/A stage (from Giga computer to analog) and the
A/D stage (at the DM2000's converters) am I possibly going to introduce more
jitter? The ADAT cable length is 20 feet, which may be a contributing
factor. I might be able to move the Giga computers closer to the board,
but for noise reasons, and because of SVGA, mouse and keyboard cable length
considerations, I'd rather not.

Am I worrying about something that probably won't be an issue? Other than
for studio monitoring, there will be no D/A conversion process. All digital
audio in the DM2000 goes into the DAW digitally, where mixing, signal
processing and mastering takes place.

--
Best Wishes,

Jerry Gerber
www.jerrygerber.com
www.cdbaby.com/virtualorchestra

[David Collins]

In article et>,
"Jerry Gerber" > wrote:

> Am I worrying about something that probably won't be an issue? Other than
> for studio monitoring, there will be no D/A conversion process. All digital
> audio in the DM2000 goes into the DAW digitally, where mixing, signal
> processing and mastering takes place.

Jitter is only a problem at a/d and d/a conversions.

DC

--
Dave Collins Entropy just isn't what it used to be!


www.collinsaudio.com

[Jerry Gerber]

The information in Mastering Audio by Bob Katz and also Principles of
Digital Audio by Ken Pohlmann says there are two kinds of jitter, one the
kind you speak about (data jitter during the conversion stage, and the other
known as interface jitter. If I understand it correctly, this is jitter
produced by a poor cable or a cable that is too long rather than by a clock
that is not well-designed and affecting the A/D or D/A process.

I think what I am really wondering about is whether to buy a soundcard that
can sync directly to word clock but has more ADAT than I need, or get one
that is less expensive but only syncs to ADAT or SPDIF. I hear wordclock
cables are more hearty in regard to signal transmission at longer lengths.

--
Jerry Gerber
www.ottavarecords.com



"David Collins" > wrote in message
...
> In article et>,
> "Jerry Gerber" > wrote:
>
> > Am I worrying about something that probably won't be an issue? Other
than
> > for studio monitoring, there will be no D/A conversion process. All
digital
> > audio in the DM2000 goes into the DAW digitally, where mixing, signal
> > processing and mastering takes place.
>
> Jitter is only a problem at a/d and d/a conversions.
>
> DC
>
> --
> Dave Collins Entropy just isn't what it used to be!
>
>
> www.collinsaudio.com

[Neil Gould]

Recently, Jerry Gerber > posted:

> The information in Mastering Audio by Bob Katz and also Principles of
> Digital Audio by Ken Pohlmann says there are two kinds of jitter, one
> the kind you speak about (data jitter during the conversion stage,
> and the other known as interface jitter. If I understand it
> correctly, this is jitter produced by a poor cable or a cable that is
> too long rather than by a clock that is not well-designed and
> affecting the A/D or D/A process.
>
More directly, "data jitter" is a loss of synchronicity due to clocking
errors during data transfer. Whether or not this would ever be a problem
depends on things in addition to your cables.

> I think what I am really wondering about is whether to buy a
> soundcard that can sync directly to word clock but has more ADAT than
> I need, or get one that is less expensive but only syncs to ADAT or
> SPDIF. I hear wordclock cables are more hearty in regard to signal
> transmission at longer lengths.
>
Wordclock is useful for synchronizing *multiple* data sources with a
single input. In a system with a single ADAT or SP/DIF feed, it would be
of little value. For example, you have one ADAT (or source that uses
lightpipe) that you want to feed to your computer for editinig. Since the
sync signal is imbedded in the data stream, any decent ADAT receiver
(audio card) and decent ADAT cable will transfer the data reliably and
without "data jitter". The same is true for SP/DIF. Adding wordclock is
not likely to improve the quality of the transfer.

OTOH, if you have two or more ADATs that you wish to transfer
simultaneously to your computer, you will need to synchronize those units.
That's where wordclock comes in handy, and in fact, some method of sync is
required.

Hope this helps.

Neil

[Mike Rivers]

In article .net> writes:

> I think what I am really wondering about is whether to buy a soundcard that
> can sync directly to word clock but has more ADAT than I need, or get one
> that is less expensive but only syncs to ADAT or SPDIF. I hear wordclock
> cables are more hearty in regard to signal transmission at longer lengths.

Jitter isn't going to be your problem. What might be your problem, if
you don't have cards in your Gigastudio computers (didn't we just go
through this?) that sync to some external word clock input, you could
have problems with data losing sync on its way into the console.

If you have some spare Lightpipe outputs from the console, you may be
able to connect them up to the inputs of your computer's sound cards
and set them up so that they sync to incoming ADAT Lightpipe, which
will get the computer in sync with the console, which is in sync with
the Big Ben.

Just one more possibility for you to experiment with.

However the reality is that since you're using synthesized instruments
anyway, any change in the sound resulting from analog outputs going
into analog inputs is either likely to be completely insignificant or
something for which you can compensate with some small tweaks. And
when it comes to making music, simpler is always better.



--
I'm really Mike Rivers - )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

[Jerry Gerber]

Hi Mike,

So, in your opinion, it may not even be worth the time or expense to have
Gigastudio samples going into the digital mixing console via ADAT, by
passing the d/a and a/d conversion processes because I am working with
sampled sounds anyway? You may very well be right. I don't want to buy
equipment for the sake of buying new equipment, my goal is always to improve
the sound. By the way, there is nothing feeding Gigastudio from the board,
the only thing being sent to Giga is midi data, it is what happens to the
sound coming out of Giga that I am contemplating, which is whether to change
from my current setup, analog out to analog in, to a new one, digital out to
digital in...

Jerry Gerber


"Mike Rivers" > wrote in message
news:znr1078920067k@trad...
>
> In article .net>
writes:
>
> > I think what I am really wondering about is whether to buy a soundcard
that
> > can sync directly to word clock but has more ADAT than I need, or get
one
> > that is less expensive but only syncs to ADAT or SPDIF. I hear
wordclock
> > cables are more hearty in regard to signal transmission at longer
lengths.
>
> Jitter isn't going to be your problem. What might be your problem, if
> you don't have cards in your Gigastudio computers (didn't we just go
> through this?) that sync to some external word clock input, you could
> have problems with data losing sync on its way into the console.
>
> If you have some spare Lightpipe outputs from the console, you may be
> able to connect them up to the inputs of your computer's sound cards
> and set them up so that they sync to incoming ADAT Lightpipe, which
> will get the computer in sync with the console, which is in sync with
> the Big Ben.
>
> Just one more possibility for you to experiment with.
>
> However the reality is that since you're using synthesized instruments
> anyway, any change in the sound resulting from analog outputs going
> into analog inputs is either likely to be completely insignificant or
> something for which you can compensate with some small tweaks. And
> when it comes to making music, simpler is always better.
>
>
>
> --
> I'm really Mike Rivers - )
> However, until the spam goes away or Hell freezes over,
> lots of IP addresses are blocked from this system. If
> you e-mail me and it bounces, use your secret decoder ring
> and reach me here: double-m-eleven-double-zero at yahoo

[Mike Rivers]

In article .net> writes:

> So, in your opinion, it may not even be worth the time or expense to have
> Gigastudio samples going into the digital mixing console via ADAT, by
> passing the d/a and a/d conversion processes because I am working with
> sampled sounds anyway? You may very well be right.

Maybe, maybe not. It depends on how fussy you are, and what you're
fussy about. If you live your life by making comparisons between A and
B, you'll always wonder about what C that you don't have might sound
like and you'll never get anything constructive done. What you should
use as your criteria is whether, when you get the sounds into your
mix, they sound like you planned for them to sound. There may be a
perceptible difference between using an analog and digital interface,
but a piano will still sound like a piano, and the difference is
unlikely to make or break your mix.

Some people never get started working because they're always worried
that they have to buy another piece of gear. If you have sounds coming
out of your computers now, and have matching inputs on the console,
just plug it in. You'll have to pay attention to setting levels since
you have controls in that domain that you don't have in the digital
interface domain, but that's not a big deal, and just a normal
operation for an audio engineer.

> By the way, there is nothing feeding Gigastudio from the board,
> the only thing being sent to Giga is midi data, it is what happens to the
> sound coming out of Giga that I am contemplating, which is whether to change
> from my current setup, analog out to analog in, to a new one, digital out to
> digital in...

I understand. The only reason why I suggested connecting a Lightpipe
out of the console to the input of the computer's sound card was to
get the card to use that input as a word clock source so its OUTPUT
(what you're mixing) will be in sync with the console's clock. But of
course if you're using analog interfacing, you don't need to worry
about that.



--
I'm really Mike Rivers - )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

[Jerry Gerber]

Thanks for taking the time to post about this topic Mike,

Jerry Gerber


"Mike Rivers" > wrote in message
news:znr1078940406k@trad...
>
> In article .net>
writes:
>
> > So, in your opinion, it may not even be worth the time or expense to
have
> > Gigastudio samples going into the digital mixing console via ADAT, by
> > passing the d/a and a/d conversion processes because I am working with
> > sampled sounds anyway? You may very well be right.
>
> Maybe, maybe not. It depends on how fussy you are, and what you're
> fussy about. If you live your life by making comparisons between A and
> B, you'll always wonder about what C that you don't have might sound
> like and you'll never get anything constructive done. What you should
> use as your criteria is whether, when you get the sounds into your
> mix, they sound like you planned for them to sound. There may be a
> perceptible difference between using an analog and digital interface,
> but a piano will still sound like a piano, and the difference is
> unlikely to make or break your mix.
>
> Some people never get started working because they're always worried
> that they have to buy another piece of gear. If you have sounds coming
> out of your computers now, and have matching inputs on the console,
> just plug it in. You'll have to pay attention to setting levels since
> you have controls in that domain that you don't have in the digital
> interface domain, but that's not a big deal, and just a normal
> operation for an audio engineer.
>
> > By the way, there is nothing feeding Gigastudio from the board,
> > the only thing being sent to Giga is midi data, it is what happens to
the
> > sound coming out of Giga that I am contemplating, which is whether to
change
> > from my current setup, analog out to analog in, to a new one, digital
out to
> > digital in...
>
> I understand. The only reason why I suggested connecting a Lightpipe
> out of the console to the input of the computer's sound card was to
> get the card to use that input as a word clock source so its OUTPUT
> (what you're mixing) will be in sync with the console's clock. But of
> course if you're using analog interfacing, you don't need to worry
> about that.
>
>
>
> --
> I'm really Mike Rivers - )
> However, until the spam goes away or Hell freezes over,
> lots of IP addresses are blocked from this system. If
> you e-mail me and it bounces, use your secret decoder ring
> and reach me here: double-m-eleven-double-zero at yahoo

[Jerry Gerber]

By the way, I just installed Big Ben a few minutes ago. Music I know well
definitely sounded crisper, and sharper, better to my ears.

Jerry


"Mike Rivers" > wrote in message
news:znr1078940406k@trad...
>
> In article .net>
writes:
>
> > So, in your opinion, it may not even be worth the time or expense to
have
> > Gigastudio samples going into the digital mixing console via ADAT, by
> > passing the d/a and a/d conversion processes because I am working with
> > sampled sounds anyway? You may very well be right.
>
> Maybe, maybe not. It depends on how fussy you are, and what you're
> fussy about. If you live your life by making comparisons between A and
> B, you'll always wonder about what C that you don't have might sound
> like and you'll never get anything constructive done. What you should
> use as your criteria is whether, when you get the sounds into your
> mix, they sound like you planned for them to sound. There may be a
> perceptible difference between using an analog and digital interface,
> but a piano will still sound like a piano, and the difference is
> unlikely to make or break your mix.
>
> Some people never get started working because they're always worried
> that they have to buy another piece of gear. If you have sounds coming
> out of your computers now, and have matching inputs on the console,
> just plug it in. You'll have to pay attention to setting levels since
> you have controls in that domain that you don't have in the digital
> interface domain, but that's not a big deal, and just a normal
> operation for an audio engineer.
>
> > By the way, there is nothing feeding Gigastudio from the board,
> > the only thing being sent to Giga is midi data, it is what happens to
the
> > sound coming out of Giga that I am contemplating, which is whether to
change
> > from my current setup, analog out to analog in, to a new one, digital
out to
> > digital in...
>
> I understand. The only reason why I suggested connecting a Lightpipe
> out of the console to the input of the computer's sound card was to
> get the card to use that input as a word clock source so its OUTPUT
> (what you're mixing) will be in sync with the console's clock. But of
> course if you're using analog interfacing, you don't need to worry
> about that.
>
>
>
> --
> I'm really Mike Rivers - )
> However, until the spam goes away or Hell freezes over,
> lots of IP addresses are blocked from this system. If
> you e-mail me and it bounces, use your secret decoder ring
> and reach me here: double-m-eleven-double-zero at yahoo

[Mike Rivers]

> From: "Jerry Gerber" >

> By the way, I just installed Big Ben a few minutes ago. Music I know well
> definitely sounded crisper, and sharper, better to my ears.

That's just what Apogee says about it too.


--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me here: double-m-eleven-double-zero at yahoo

[Jerry Gerber]

Well, maybe so, but I know what I hear. The thing I noticed very quickly
was the stereo separation was much improved.


Jerry Gerber


"Mike Rivers" > wrote in message
news:znr1078959969k@trad...
>
> > From: "Jerry Gerber" >
>
> > By the way, I just installed Big Ben a few minutes ago. Music I know
well
> > definitely sounded crisper, and sharper, better to my ears.
>
> That's just what Apogee says about it too.
>
>
> --
> I'm really Mike Rivers )
> However, until the spam goes away or Hell freezes over,
> lots of IP addresses are blocked from this system. If
> you e-mail me and it bounces, use your secret decoder ring
> and reach me here: double-m-eleven-double-zero at yahoo

[David Collins]

In article .net>,
"Jerry Gerber" > wrote:

> The information in Mastering Audio by Bob Katz and also Principles of
> Digital Audio by Ken Pohlmann says there are two kinds of jitter, one the
> kind you speak about (data jitter during the conversion stage, and the other
> known as interface jitter. If I understand it correctly, this is jitter
> produced by a poor cable or a cable that is too long rather than by a clock
> that is not well-designed and affecting the A/D or D/A process.

When you are making a digital to digital transfer, the interface jitter
is not a problem.

DC

--
Dave Collins Entropy just isn't what it used to be!


www.collinsaudio.com

[Arny Krueger]

"Jerry Gerber" > wrote in message
hlink.net
> The information in Mastering Audio by Bob Katz and also Principles of
> Digital Audio by Ken Pohlmann says there are two kinds of jitter, one
> the kind you speak about (data jitter during the conversion stage,
> and the other known as interface jitter. If I understand it
> correctly, this is jitter produced by a poor cable or a cable that is
> too long rather than by a clock that is not well-designed and
> affecting the A/D or D/A process.

I see at least four distinct kinds of jitter:

(1) Power line related jitter
(2) Jitter that is related to tracking servoes
(3) Jitter related to the frame rate
(4) Self jitter or interface jitter

[Roger W. Norman]

Assuming the I/O on the computer, there is no reason to bring the Giga
computer sounds into the console and then pass it on. But as long as
external word clock is used, there's no reason not to, either. The matter
of possible jitter would more seriously be a product of inferior cable in
your description. Glass instead of plastic would help eliminate that.

The obvious solution would be to drive all your digital devices by a single
external clock. The idea that ADAT lightpipe carries it's own clocking
signal makes no nevermind once you set up all the devices to slave to word
clock. The problem is whether your Giga computers have that ability and
that depends on the audio interface. Ideally you'd have an audio interfi
that would have ADAT out and Word Clock In. A number of audio cards fit
that description, although I now see that RME has discontinued a lot of them
that fit your requirements. Guess it depends on the number of outs you need
from the Giga computers as to what's reasonable, but having something like
the Digi96/8 with the Word Clock Module would make sense and eliminate your
concerns.

If the Big Ben doesn't do multiple clock outputs, daisy chaining is a
reasonable secondary method of clocking all sources to one clock, assuming
that you know what is and isn't terminated.

--


Roger W. Norman
SirMusic Studio

"Jerry Gerber" > wrote in message
hlink.net...
> Hi Mike,
>
> So, in your opinion, it may not even be worth the time or expense to have
> Gigastudio samples going into the digital mixing console via ADAT, by
> passing the d/a and a/d conversion processes because I am working with
> sampled sounds anyway? You may very well be right. I don't want to buy
> equipment for the sake of buying new equipment, my goal is always to
improve
> the sound. By the way, there is nothing feeding Gigastudio from the
board,
> the only thing being sent to Giga is midi data, it is what happens to the
> sound coming out of Giga that I am contemplating, which is whether to
change
> from my current setup, analog out to analog in, to a new one, digital out
to
> digital in...
>
> Jerry Gerber
>
>
> "Mike Rivers" > wrote in message
> news:znr1078920067k@trad...
> >
> > In article .net>
> writes:
> >
> > > I think what I am really wondering about is whether to buy a soundcard
> that
> > > can sync directly to word clock but has more ADAT than I need, or get
> one
> > > that is less expensive but only syncs to ADAT or SPDIF. I hear
> wordclock
> > > cables are more hearty in regard to signal transmission at longer
> lengths.
> >
> > Jitter isn't going to be your problem. What might be your problem, if
> > you don't have cards in your Gigastudio computers (didn't we just go
> > through this?) that sync to some external word clock input, you could
> > have problems with data losing sync on its way into the console.
> >
> > If you have some spare Lightpipe outputs from the console, you may be
> > able to connect them up to the inputs of your computer's sound cards
> > and set them up so that they sync to incoming ADAT Lightpipe, which
> > will get the computer in sync with the console, which is in sync with
> > the Big Ben.
> >
> > Just one more possibility for you to experiment with.
> >
> > However the reality is that since you're using synthesized instruments
> > anyway, any change in the sound resulting from analog outputs going
> > into analog inputs is either likely to be completely insignificant or
> > something for which you can compensate with some small tweaks. And
> > when it comes to making music, simpler is always better.
> >
> >
> >
> > --
> > I'm really Mike Rivers - )
> > However, until the spam goes away or Hell freezes over,
> > lots of IP addresses are blocked from this system. If
> > you e-mail me and it bounces, use your secret decoder ring
> > and reach me here: double-m-eleven-double-zero at yahoo
>
>

[Arny Krueger]

"Roger W. Norman" > wrote in message


> Assuming the I/O on the computer, there is no reason to bring the Giga
> computer sounds into the console and then pass it on. But as long as
> external word clock is used, there's no reason not to, either. The
> matter of possible jitter would more seriously be a product of
> inferior cable in your description. Glass instead of plastic would
> help eliminate that.

IME the most common sources of audible jitter are either ground loops in the
digital lines that puts a great deal of power line into the signal being
clocked from, or marginal clock recovery that is not really that well
synched all of the time for whatever reason.

> The obvious solution would be to drive all your digital devices by a
> single external clock.

In the end, all concurrently-operating digital sources you mix down either
have to trace back to a common clock by hook or by crook, or there has to be
resampling someplace in the works.

>The idea that ADAT lightpipe carries it's own
> clocking signal makes no nevermind once you set up all the devices to
> slave to word clock. The problem is whether your Giga computers have
> that ability and that depends on the audio interface. Ideally you'd
> have an audio interfi that would have ADAT out and Word Clock In. A
> number of audio cards fit that description, although I now see that
> RME has discontinued a lot of them that fit your requirements. Guess
> it depends on the number of outs you need from the Giga computers as
> to what's reasonable, but having something like the Digi96/8 with the
> Word Clock Module would make sense and eliminate your concerns.

> If the Big Ben doesn't do multiple clock outputs, daisy chaining is a
> reasonable secondary method of clocking all sources to one clock,
> assuming that you know what is and isn't terminated.

Plan "B": a good distribution amp for composite video. The levels are right,
the impedance is right, the connectors are right, and the bandwidth is
right.

[Scott Dorsey]

In article >,
David Collins > wrote:
>In article .net>,
> "Jerry Gerber" > wrote:
>
>> The information in Mastering Audio by Bob Katz and also Principles of
>> Digital Audio by Ken Pohlmann says there are two kinds of jitter, one the
>> kind you speak about (data jitter during the conversion stage, and the other
>> known as interface jitter. If I understand it correctly, this is jitter
>> produced by a poor cable or a cable that is too long rather than by a clock
>> that is not well-designed and affecting the A/D or D/A process.
>
>When you are making a digital to digital transfer, the interface jitter
>is not a problem.

Unless, of course, it's so outrageously bad that errors start occurring
and the thing screeches and loses synch. It has to be awfully bad for
that to happen, though.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Roger W. Norman]

"Arny Krueger" > wrote in message
...
> Plan "B": a good distribution amp for composite video. The levels are
right,
> the impedance is right, the connectors are right, and the bandwidth is
> right.

Yeah, that works, just as long as "good" is the operative word. And I
really didn't mean to say daisy chain. I meant using a word clock bus like
networking, making certain that something along the bus didn't self
terminate. And that's usable on the cheap. A good video distribution amp
could go high enough in price as to make one wonder why they didn't get a
word clock with distribution amp built in.

--


Roger W. Norman
SirMusic Studio