Turtlerock Forum

I haven't seen a thread about this topic - what do you guys think about it?

Is it worth having an external word clock?

Does it affect the sound quality or just the sync?

Are there any cheap but worthwhile options other than using the ADAC's or (e.g.) ProTools clocking?

ty

They....you know....they who know more than I do .....I presume....... say the clock is one of the more significant components in the conversion process in achieving better sound. Something to do with Jitter. Never heard jitter myself......maybe if someone pointed it out while i was listening i might recognise it. The converter chip is of lesser significance. The analog componetry and filter circuitry is also a key factor in the sound. The Universal converter is said to be a great sounding unit because of the analogue stage and the Antelope Audio Atomic clock is meant to be the beez neez of converters for the clock stage. Some bright sparks even make their own using the satelite clock that can be accessed by those who know about such stuff........
What would I know from direct experience? ...nothing much, except that my Appogee Ensemble is plenty good enough for my needs.
Some Info on the Universal
http://www.soundonsound.com/sos/aug10/a ... master.htm

And the Antelope
http://www.antelopeaudio.com/en/products_iso_10m.html

the greatest digital minded people i know and have spoke to about this subject tell me if the
box is designed properly to talk to other boxes external clocking is not only not needed but damaging
(weiss/lavry/crane song/ prism take this line )

but i had a very long and valuable conversation with bruce jackson one time he was the guy that quietly put the wordclock in our studio vocab with the apogee super clock 20 years before you ever heard of an apogee big ben

his attitude is nearly every digital device IS NOT designed correctly to play with others and word clock is of great use
i tend to agree with him from experience

when do i use word clock ..?

when i am getting digital artifacts WITHOUT it
( odd clicks, glitches and weird things )

i dont use it on the stuff i cannot hear the difference on

some of the claims i read about word clocks only leads me to believe

the devices they are "fixing" were not designed well to play with other devices in the first place ( which is lots of stuff) or its a case of the emperors new clothes (most scenerios) and its all placebo effect

I asked bruce what the difference between the apogee word clock
I have been using from the uv1000 cd encoder forever and the newer apogee big ben ?
he just smiled and said
"well the one i invented and designed ..... i designed "

So my feeling is dont pay big bucks for your clock worry about other things but dont forget about it !

and also note i wrote an magazine article on wordclocks 15 years ago and i had a hell of a time getting the editor to understand it.. little lone print it
he said .. "oh well if you think its important ..." but its only for high end super expensive pro stuff right ..?
i said well no all digital stuff - he blanked me

i would suggest if anybody is hearing the "jitter" without the wordclock they are dealing with a box that needs a better clock sometimes its hard to know where to look with this stuff but if your are hearing actual jitter as a freq sound would indicate something was WAY off andthere will be random unexplained clicks and phisses as well

i have never heard "jitter" myself

but got those clicks and pops ..? you need a clock !

dont read me wrong angus
clocks fix things


if your hearing wierd stuff without them and its better with them
thats a fix in my book

As far as jitter goes, I rate it on Sammas' list of "I ain't got a clue". Right up there with dither. Do the different types of dither really sound all that different???

Sometimes they do, but only when you're comparing them.

Word clocks and external clocking especially are a seemingly controversial area within the DAW world. Once you are running a number of digital devices in separate chassis' that need to be synchronized however, a master clock is sometimes necessary and the easiest, hassle-free solution to keeping everything in sync. Clocking off the embedded clock within AES/SPDIF/ADAT streams is possible, but can get tricky and one dodgy cable or setting can bring the whole system down. I use a master clock for this very simple reason.

In addition to this (as Rick mentions via Bruce Jackson) many digital devices are simply not designed correctly when it comes to word-clock interfacing and even basic digital output specs. Early Mackie digital desks notoriously provided under-volted word-clock outputs, and the Presonus 8pre rack mount units are problematic with digital clocking at times. The JetClock PLL scheme used by them and many other manufacturers causes problems when interfacing with other schemes with tighter tolerances, causing dropouts and glitches.

These three articles have some good basic info on digital clocking:

http://www.soundonsound.com/sos/jun10/a ... clocks.htm
http://www.soundonsound.com/sos/apr03/a ... ocking.asp
http://www.drawmer.com/uploads/File/dow ... k-sync.pdf

If you really want to go down the rabbit hole and get into the gory details, take a look at these PDF's from the Lavry Engineering and Grimm Audio websites. These cover things in a lot more detail, and the Grimm documents go into the way two clocks interact in a master-slave setup with different PLL clocking schemes, filters and tolerances.

http://www.grimmaudio.com/whitepapers/P ... basics.pdf
http://www.grimmaudio.com/whitepapers/p ... ocking.pdf
http://www.grimmaudio.com/whitepapers/P ... %20ppm.pdf

http://www.lavryengineering.com/documen ... Theory.pdf
http://www.lavryengineering.com/forum_i ... _Audio.pdf
http://www.lavryengineering.com/white_papers/jitter.pdf

The basic idea in an ideal setup is to run your word-clock synchronization from the internal clock of your front end AD, as it is at this point that a permanent snapshot your analogue data is first taken. Interface jitter after this point is less of an issue, as the data is just being shipped around from point to point and is reconstructed without error. It is only at the crucial AD and DA points that word-clock is of most importance.

I personally use a Mytek 8x192 AD/DA converter for this purpose as it has six word-clock outputs which I use to clock everything else in my studio. Never had a problem with clocking or the sounds I get with this setup. That being said, I have also used RME converters extensively in another setup, and their higher end boxes are of similar quality. They use a digital synthesis re-clocking scheme (Steadyclock) which has also been completely glitch and trouble free.

I am never commenting on this subject ever again, other than to say...

If you're not sure you need to worry about it then you don't need to worry about it.

If you are worried about how to know if you need to worry about it, then you don't need to worry about it. you will know when you need to worry about it.

If you know enough to know when to worry about it, you will stop worrying.

If you're not worried about it, then congratulations - you have reached nirvana.

^ aaaaaaaaaaaaahahahaha

best ever.

Now I'm really worried!!!!

Wez... that is pure brilliance.

Is it going too far to say post of the year?

Brilliant.

I listened. I worried. I tried everything. I worried. I got a BLA MCII. No worries.

stupid question maybe, sorry to resurrect an old fred, but when hooking up an AD, like a rosetta 800, into a PCI AES card, am i to understand that you'd be using the Rosetta as the master clock device in that kind of set up?

For a single ADAC and DAW setup I would use the ADAC but that's only worth the paper it's written on....

I understand from an entertaining debate between Mr Katz and the gent from Apogee that they disagree on the external clock issue.

I loved the chance to reread Wez's post from above.

lol yes, that was a top post.

I personally dont care which i use, i was just wondering what the standard would be...if there isnt, i'll just doodle about

no standard that I'm aware of but probably use the adac internal.

Stick with your AD as the master clock, as this is when the most important conversion process occurs (ie you are aiming to reduce jitter at this original conversion stage). After this AD process has occurred, what is known as interface jitter does not affect the original digital time-stamping of your audio. Transferring your converted sequence of ones & zeros through your interface vie the digital AES input does not alter their time-stamped positions. These are now set in 'digital stone'. These samples will then line-up as they were originally converted within the timeline of your DAW.

The alternative setup, where you slave your Rosetta off your PCI AES card (with the PCI card now acting as the master clock) is quite different. If the phase locked loop and re-clocking circuitry in your Rosetta are not awesome, the timing of the original analog to digital conversion may be compromised (whether this is perceptible or not is up for debate). This effect is apparently very subtle, and may even be preferred by some people; however slaving your AD to an external clock source is objectively less 'accurate' in the great majority of cases. For this not to be the case, the designers of the Rosetta would have to intentionally go out of their way to make the very stable internal crystal oscillator perform worse than the convoluted process of deriving a clock from an external source, which has multiple sources of potential interference. (dodgy word-clock cabling & impedance issues, RFI, clocking being serially embedded within AES or ADAT streams, etc...)

In summary, stick with the AD as the master unless this causes problems.

Cheers,
George

thanks for the info George!

Jitter (somewhat dumbed down...)

master clock = accurate metronome.
A/D card's internal clock = a human drummer.

You have a choice: you make the rest of the band speed up and slow down to match the crappy clock provided by the drummer or you make the whole band play along to the metronome. Which one do you expect will provide a more measurably consistent performance?

I'm sure many a/d cards/boxes have very accurate clocks. More to the point, as long as all the clocks are in sync, I would contend that any inaccuracies would be totally insignificant.

So really:

Master clock=possibly accurate metronome (at least that's what the ad says, and no manufacturer would just make shit up!)
A/D card's internal clock=still more accurate than any human drummer could ever hope to be.

davek wrote:Jitter (somewhat dumbed down...)

master clock = accurate metronome.
A/D card's internal clock = a human drummer.

You have a choice: you make the rest of the band speed up and slow down to match the crappy clock provided by the drummer or you make the whole band play along to the metronome. Which one do you expect will provide a more measurably consistent performance?

It will help your overall sound quality as it gives each sample less of a chance to shift out of its place.

But whether it merits the amount of money you're going to spend?...