Jump to content

External Clock


larry long

Recommended Posts

Per request I have started a new thread for this topic. I originally posted it under the new Deva announcement as it pertains to Zaxcom users primarily.
 

Let me risk sliding into a ditch on this one.

I have been putting together an updated home rig Lynx Aurora VT16, SPL MixDream , Neve MBP among other pieces. In my rooting around and research Ext Clocking has reared its ugly head again.

Here's the paper that stirred my interest.

http://pinknoisemag.com/pink-papers/pink-paper-002

 

Granted we have very stable recording devices thanks to people like Glenn and other engineers that are highly educated and very aware of our needs. 

When we went from tape to DAT and generally speaking it was assumed "well it's digital" 1's and 0's right? Come to find out there were many ways to come up with 1's and 0's and you could hear the difference.

I'm sure some of you have used word clock on your Deva's and other devices and that has been an option.

I am curious to know what the popular opinions are regarding us having access to this option going forward.

I certainly don't want to doubt Mr. Sanders and assume he didn't think of everything and perhaps I just need a more in depth understanding of why we wouldn't need this or how it may be a moot point.

In my over simplistic mind I imagine that it would be a simple software adjustment to make an AES input available to an external clock.

Perhaps I'm naive and if so please enlighten me with out using a flamethrower.

thanks,

LL

Link to comment
Share on other sites

Larry, explain to me why ANY external word clock should be needed with Deva or Sound Devices or any other of the production sound recorders we use in the basic setups we have for the majority of the projects we do. The Deva has SRC and interfaces with just about everything imaginable in our world, so why the need for external? I don't pretend to be even a minor expert on any of this, my comments only come from real world experience with all the gear we typically use. I do remember having a conversation with Glenn Sanders many years ago (actually when Deva did have the possibility of external word clock) and Glenn said that the real danger with external clock is having to rely on the external sync system --- if anything goes array with the clock sync signal, it has the very real potential of disrupting or stopping the recording --- this would be unacceptable in terms of reliability for a field recorder (and besides, with sample rate conversion buyilt into the recorder, external should be unnecessary, right?).

Link to comment
Share on other sites

And when your only digital device is your recorder then there us no need for an ext clock. 

When there are several digital devices involved, an ext clock might be worthwile, as the src could also impair the signal, but I think it's not required. Digital mics, like the Schoeps SuperCMIT (which I'm still looking for...) would need an extra box to sync to ext digital sync, so you'd have two extra boxes. Although, your recorder could probably sync to the digital mic and the next device downstreamcould sync to the digital signal from your recirder and so on. 

Link to comment
Share on other sites

Jeff,

Built in SRC, that is where I may be misunderstanding.

I would assume (and I know that is dangerous) that the unit relies and syncs to a single crystal oscillator across all of the converters. Given this assumption,  I was just curious about the availability of the choice to sync externally or rely on the built in clock. I know that this is an esoteric and highly subjective topic that will alway be argued just like the difference between analog and digital summing but I would be remiss not to bring it up in light of the test and observations in the music world.

Regarding the reliability of an external clock well it's like any other critical piece of gear on our carts. Keep an eye on it.

Maybe this clocking thing really isn't that important in our case I was just curious after I read that Pink Paper I referred to in my post.

LL

 

 

ps.

"Someone" has a fun story about pulling his Deva 2 off of the cart to jump on an insert car forgetting that he was getting sync from the Cameo. Wow the lessons we learn usually under fire.

 

11 minutes ago, Constantin said:

And when your only digital device is your recorder then there us no need for an ext clock.

Constantin,

Yes sir this has been a commonly held opinion regarding clocks. Did you happen to read the article referenced in my original post?

 

LL

Link to comment
Share on other sites

11 minutes ago, larry long said:

Constantin,

Yes sir this has been a commonly held opinion regarding clocks. Did you happen to read the article referenced in my original post?

 

LL

No I didn't, yet. But I do know that there may be a value in employing an ext clock with even just one device, potentially improving jitter and other things. You'd need a clock that's more stable than the one already in your recorder. Still, I would doubt the impact on our recordings 

Link to comment
Share on other sites

Constantin,

I know I can get wrapped up around the axle sometimes with the smallest of things

We are just recording dialog for sure but I love rich sound as much as any of us do.

I am curious to hear from Glenn as we have texted regarding this subject. I guess it's really for my own edification.

Just putting it out there.

 

LL

Link to comment
Share on other sites

External clock is unnecessary and dangerous. It can cause recording problems if not 100 percent stable and will cause time code accuracy problems because it can come from an uncalibrated source possibly at the wrong sample rate. A non starter from the word go. With sample rate converting inputs it is unnecessary.

Link to comment
Share on other sites

31 minutes ago, larry long said:

Copy that, moving on.

thanks,

 

LL

Not so fast, moving on. I just took the time to read the majority of the "pink paper" you linked and I can now say with absolute certainty that entertaining the idea of taking our production recorders and working out a system of external clocking would be a pointless and unrewarding exercise. It falls in the same category as much of the writing in the magazine that used to be called the Absolute Sound. For our purposes, production sound recording for motion pictures and television, it would be very much like if I went on the quest for XLR interconnect cables that were potentially less "bass shy" as the ones I am using now. The recordings that any recorder that uses high quality A to D converters will not be improved in any sonically meaningful way by employing any sort of external clocking. As pointed out in the article, in one of the sections, it was determined through very careful testing, that many of the converters that were externally clocked actually had their performance suffer as a result. Is this the sort of testing that we want to be doing before going on that big TV series we're about to start? Evaluating 10 mghz external clocks from some supposedly reputable company (that has spent years and years analyzing digital jitter) is not something I want to do. One statement in the article kind of says it all: "Just as drivers now can choose from a variety of suspension profiles, perhaps digital converters will one day include user-accessible control over the jitter profile (and other aspects) of the system.". Right. "Jitter Profile." I can see that coming as the new spec from some Studio Post Supervisor: "I hope you're not using that recorder from Sound Devices --- don't you know it has a reputation of having a lousy jitter profile?"

The beginning of the article has this quote: "If it sounds good, it is good" which it portrays as a persistent myth, and that reminds me of the quote I often use from Ray Charles: "I don't care if you've got ninety tracks, what does it sound like, baby!" which is not a myth but actually words to live by.

Link to comment
Share on other sites

1 hour ago, glenn said:

External clock is unnecessary and dangerous. It can cause recording problems if not 100 percent stable and will cause time code accuracy problems because it can come from an uncalibrated source possibly at the wrong sample rate. A non starter from the word go. With sample rate converting inputs it is in necessary.

That would imply serious user error, which we should ignore for the moment, as that is not the main aspect when comparing two systems. At lesst, it shouldn't be. 

Up until this day, I was under the impression that external clocking, and timecode had nothing to do with each other. Was I wrong?

Link to comment
Share on other sites

1 hour ago, Constantin said:

That would imply serious user error, which we should ignore for the moment, as that is not the main aspect when comparing two systems. At lesst, it shouldn't be. 

Up until this day, I was under the impression that external clocking, and timecode had nothing to do with each other. Was I wrong?

Yes external clock makes the system subject to more user errors. 

Link to comment
Share on other sites

I call it a trade off.  If you start with the sort of recorders most of us use external clocking might be an unnecessary risk, since the internal clocks are TXCO and very stable.  However, it you want to keep sync together over a long roll (like a concert), then it starts to be a better, although not totally necessary idea (still using the SD/Zax etc type recorders with good clocks).  The clock thing starts to become vital when you begin to add in other devices with less stable onboard clocks.  This could include pro-type hardware multitracks (JoeCo etc), computer-based rigs using audio interfaces  and even some digital consoles, depending on how the rig is configured.  Thus, in some real-world scenarios, I would be fine with parking one of my SD recorders for a sound-board feed and then running off with another to do doco shooting, secure in the knowledge that the two will be in good sync for up to 8 hrs or so.  If I want to record on my JoeCos for long rolls, however, I will probably clock them both from external (prob from an SD recorder), and this is how I've gotten dead sync over a long show (or shooting day).  The same can be done with computer rigs.   For really short takes that will be individually slated and synced all this is less important, but I can assure you that I learned of the importance of having a coherent and accurate clocking scheme for longer takes (esp with devices with non-accurate clocks) the HARD way.

Link to comment
Share on other sites

Philip, no one here who understands word clock and external sync and when it should be implemented, is disputing the value and necessity in setups such as you describe and Tom Duffy has posted. But that's not what the article (or pink paper) that Larry Long linked is really talking about. That's why I jumped in with my comments so that sound people here on JWSOUND do not worry about using their SD 688, Aaton Cantar, Zaxcom Nomad or Deva without implementing an external word clock. The article is talking about the incredible differences, sonically, that can be measured amongst any one of the "popular" external clock devices, many of which cost as much if not more than the recorder most people here own and use every day! Doing a little bit more research, I discovered that a few of the people mentioned in the article as having done all this testing (to determine such things as a converter's jitter profile) are working as consultants for the companies that produce these overpriced clocks. It's like the newsletters you get from doctors warning you about the dangers of imbalances in blood sugar levels and at the end of the article there is a recommendation for a totally natural supplement that could help you in this regard.

Link to comment
Share on other sites

As always, there are levels to the question:

0. Is synchronization needed for the system to work  (yes in my example, no for bag work)

1. Should two pieces of digital equipment be synchronized, or use SRC?

  -> SRC makes it foolproof, direct synchronization improves audio quality. (SRC built into a typical AES/EBU receiver chip has ~120dB range)

2. Does the short term stability of an external clock make an audible difference?

  -> most likely not, only for badly designed / legacy gear.

3. Does the long term or temperature stability of an external clock make a difference?

 -> possibly, depends on usage.  Knowing that a recording made yesterday is within 1ppm of a recording made today may be important.

4. Does the jitter spectrum of the external clock, and its resulting bias on the sound of the converters matter?

 -> experts disagree. In this market, the difference is most likely unmeasurable / not significant.   In the studio, differences have been heard.

Personally, I'm more interested in being able to reference a GPS 1pps signal as the master clock, than a 10MHz vapor cell; GPS would mean your 48kHz is always the same as someone elses 48kHz to within a hair, and keeps you wire free.

Link to comment
Share on other sites

17 minutes ago, larry long said:

BTW Jeff, 

I did contact the guy that wrote the paper and he said he would be willing to talk to me for $100 an hour with a 2 hour minimum.

Needless to say that was the end of that.

LL

 

 

Exactly! Thanks, Larry, and through all of this I hope you did not feel that I was casting any aspersions your way --- if I hadn't gotten so wrapped up in all of this I was planning on prefacing everything by pointing out that the original poster (you) Larry Long, is a well respected veteran production sound mixer --- that is why it was important to clarify that you found the article interesting but it was not a warning to start distrusting your gear.

Link to comment
Share on other sites

My comment was in response to another post above that indicated that external clocking was "unnecessary and dangerous" without specifying the situation.  As I said, with pro mixer-recorders commonly used by folks around here it isn't needed except in a few sorts of circumstances.  But with other digital audio gear in use around here, particularly computer/interface systems, not using ext clock is dangerous and will lead to trouble in post.

 

Link to comment
Share on other sites

  • 6 years later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...