Proper Word clock setup with SYNC I/O, Digital 192 and external converters

[doubtful guest]

Hi all, I'm trying to figure out the best setup for word clock/loop sync distribution for one of our systems.
We have an older G4/HD3 running protools 7.1.
The hardware is:
SYNC I/O
Digital 192
Apogee AD16 and DA16 feeding the Digital 192(AES)
Rosetta 800 with X-HD card.

Ignoring the rosetta 800 for the moment, is it ideal to have the Digital 192 locked to loop sync from the SYNC I/O?

Is there an advantage to using loop sync when the converters are clocked from external word clock?

In general we have had the SYNC I/O set as the master and generating loop sync to the Digital 192.
In the protools setup we have had the Digital 192 set to external word clock.(usually 88.2)
We then have word clock distributed from the 192 to the AD16 then DA16 then the rosetta.

Would it be better to run the word clock from the SYNC I/O?

Should the master clock in protools be set to the 192 external clock or the SYNC I/O?

Any information is appreciated, I can't find much discussion about this online. Thanks!

[Craig F]

Ignoring the rosetta 800 for the moment, is it ideal to have the Digital 192 locked to loop sync from the SYNC I/O?

it's mandatory

Is there an advantage to using loop sync when the converters are clocked from external word clock?

Loop is how clock is used inside the Digi Hardware

Would it be better to run the word clock from the SYNC I/O?

depends on your external generator

Should the master clock in protools be set to the 192 external clock or the SYNC I/O?

Sync, the only time I use the 192 as Loop Master is when I need to transfer digitally from a device that does not have a clock input.

[Eric Seaberg]

We have a house wordclock master that feeds all of our rooms via AES, then is distributed throughout the room as wordclock BNC.

The AES clock actually goes to the Sync I/O, the same you could do with your Apogee, either WC or AES audio. Then we take wordclock out of the Sync I/O and feed a Sync DA to distribute around the room, i.e. CDRs with WC, etc. LoopSync still MUST feed the 192 I/Os, yet we can determine what the room 'locks' to, either house at 44.1 or go internal Sync I/O at another SR, and the room will follow.

In your case, you would have the Sync I/O look at the Apogee, setup using Command-2 (numeric pad) for Session info. This is where we select AES or Internal or Video (if we're doing that).

[doubtful guest]

Thanks for the replies!

In general we aren't doing any video or external/house sync. In the past we have locked to 2 inch or ADATs but that hasn't been an issue lately.(luckily)

We are using the SYNC I/0 as the master clock(and loop master), would there be an advantage to using the Digital 192 as the master clock?

Is the word clock generated by the Digital 192 and the SYNC I/O the same if the 192 is locked to the SYNC I/O by loop sync?

thanks!

[doubtful guest]

Loop sync is higher resolution than word clock, right?(256x?)

Is there a sonic advantage to using loop sync over word clock?(or using wordclock over loop sync?)

If a Digital 192 is locked via loop sync to a SYNC I/O is it regenerating word clock? Or is it passing the same word clock that the SYNC I/O is generating?

[doubtful guest]

re:256x... thanks, I guess I was having a flashback to superclock.

The Digi paper on clock jitter(http://akmedia.digidesign.com/suppor...tter_30957.pdf) seems to be just related to the analog input of the regular 192 and the issues of clocking converters internally or externally.

I know in the past Digi has strongly stated that loop sync is better than word clock.
So is that not an issue at all with a Sync I/O master? Can DTS confirm this?
It seems that the SYNC I/O would follow the session sample rate without problems and I can run wordclock to everything else.

Does it matter what the wordclock cable order is if you are using star distribution?

[Eric Seaberg]

Originally Posted by doubtful guest

Does it matter what the wordclock cable order is if you are using star distribution?

You shouldn't use any type of 'T' connector, always loop in/out if possible. Better yet is to use a Sync DA.

I thought you were trying to use this setup using the clock of the Apogee? If you take an AES out of it, or Wordclock into the Sync I/O it should work as long as you select the proper source in your session setup.

BTW, I did some early testing comparing the jitter of the 192 running internal (with Sync I/O) and an external clocksource. There is more jitter in the system when it's trying to lock to an external source. If you think it sounds better, then use it. If you don't care, then have your Apogee reference to the Sync I/O.

[doubtful guest]

You shouldn't use any type of 'T' connector, always loop in/out if possible. Better yet is to use a Sync DA.

I have been under the impression that T connectors(with proper termination) were the best way to distribute low-jitter clock.

I thought you were trying to use this setup using the clock of the Apogee? If you take an AES out of it, or Wordclock into the Sync I/O it should work as long as you select the proper source in your session setup.

I'm really trying to sort out the best way to keep the Digital 192 or SYNC I/O as the master. My experience has also been that converters sound best running on their internal clock, but it's tricky with a Digital 192. I can't run both the ADC and the DAC as the master, I also would prefer to have the AD16 and rosetta getting the same clock signal.

[doubtful guest]

I guess my question is really, how do I get the lowest jitter system wide when using a Digital 192?

Does a Digital 192 have lower jitter running on internal clock instead of the SYNC I/O clock?

Does a Digital 192 have lower jitter running on loop sync or wordclock from a SYNC I/O?

thanks!

[doubtful guest]

you obviously have word clock in and out connections on your hardware as well as loop sync in and out. the settings for each device are accessed in the session window and in the i/o dialogue.

This part of my question, which has lower jitter?
Since I can't run all the interfaces from loop sync, should I get rid of it altogether or run the Digi interfaces on loop sync and the other boxes on word clock? Is there a difference in jitter either way?

not exactly sure what you are trying to say there, but using loop sync will kick everything down to a proprietary digi form of base rate for sync, without regard to whatever your actual sample rate is.

As far as I can tell the SYNC I/O will follow the actual sample rate of the session.

again, not sure what you are saying. if you are using star distribution there is no "cable order". you would have a sync generator with multiple outputs, and each slave device receiving sync directly from one of those outputs.

As I mentioned previously I am using T connectors for the word clock distribution. Someone mentioned this was not the best setup. Digi seems to recommend this because everything gets a parallel 1st generation clock.

if you are recording an analog source, then try to have the device that contains the a to d that you are using for conversion running off its internal clock.

As I mentioned before I have two ADC's so I can't have both running off their internal clock and if I have one as internal clock it will not follow the session files.