Basic Sync question

[rjt]

Okay, I finally got some (for me) good equipment and am trying to get it to work together. I have a stand alone DAW (I need portability) and a computer with a pretty decent sound card. My first puzzle has to do with sync. There are (at least) 3 ways to sync this system. I read about Midi Time Code and Midi Clock, which are both over Midi Cables and "Optical Sync" over the ADAT light pipe, which I assume syncs the audio. Here are the questions:

1) What is the difference between Midi Time Code and Midi Clock? (which is "better" if I can do both)

2) Do I have to use both an audio sync (like the optical sync which I presume is for "sample accuracy") and some kind of midi sync.

I have tried to read/learn about this but have not found a good basic article which goes through these issues. Any help is appreciated. Thanks

 

[sonusman]

Different babies...

There is two kinds of sync's you are going to deal with in audio and midi.

One is word clock sync. This is where you have to somehow sync two digital devices together so that any interchange between them has all the sampling rate clocks in sync.

The other is "position" sync. This is where you sync two devices together to play at the same time, and keeping the relative time of the two devices together.

They are two different things all together.

One is for getting digital audio flowing back and forth between units and have each unit sending/recieving data on the same digital clock. The other is to have a tape and maybe a sequencer stay together.

There are numerous different protocols for sync. AES/EBU, S/PDIF, Alesis ADAT sync, T/PDIF, word clock, etc....There are basically the sync's used for aligning digital clocks between units so the samples match up.

For position sync, there is MTC, Midi Clock, and SMPTE. I don't know of any others. SMPTE is by far the most stable. Although, MTC "derives" from SMPTE, so it may be usable. Midi Clock seems to have "drift" depending upon how complicated the set up is. With any of the position sync's, there is going to be some slop and play. But generally speaking, midi is the least reliable. SMPTE is the professional standard, and most pro gear that does position syncing will have SMPTE. If it don't, then you COULD have some problems with drift. But, that is not a given. Sometimes you just have to try MTC or Midi Clock and see if it works. The deal with midi is that if you have a lot of midi infomation going through the same cable, things start getting too complicated for the sending and recieving devices and you either lose sync, or midi commands get lost. There are ways around that, but this requires some very versatile midi I/O on the devices you use.

No matter what, syncing uses the concept of Master and Slave. No, not S&M stuff! But sort of.... The master device supplies the sync for the other devices (slaves) to follow. In a complicated studio setup, you will have to usually have a Master Time Code device to keep everything happy and in sync. MOTU is probably one of the worlds leaders in Master Time Code devices, and they offer units that can keep any number of different kinds of sync's running well.

Go to Alesis's website and download the manual for the BRC. That manual has some excellent (but sometimes hard to understand if you don't read carefully) descriptions of all the different clocks. It should clear up your questions.

Good luck.

Ed

 

[rjt]

Many thanks... that is a much clearer answer than I have gotten in music stores. I printed the answer out and am off to the Alesis site. Go Beavers... Go Ducks!!!

 

[sonusman]

Hehe...

Don't trust much of what a maroon at a music store tells you. They probably know less then you do at this point!

They are there to hype and sell you equipment, not to tell you how it works or how to use it.

Good luck.

Ed

 

[John Sayers]

Well said Ed - may I just add one point here - When they made SMPTE they left room on the end of the word to put in extra data if need arose (Good thinking) MTC is where that area is taken up by midi so you get the best of both worlds in one code. But it is still only a position code as Ed pointed out.

In the TV studios for example who use SMPTE for everything the SMPTE takes its sync from video sync (or black burst as they call it) which is generated from master control and syncs up the whole station.

Cheers
John

[Edited by John Sayers on 10-28-2000 at 22:13]

 

[sonusman]

Hey John...

Dare we confuse him with the fun world of sync conversions, smpte regenerators, and wild card sync? Oh the confusion there! How about some discussion on digital jitter! Oh noooooooooooooooo....

I was going to mention black burst, but didn't feel it was pertinent to audio production.

I like my BRC because it has just about every sync I/O on it, and makes an excellent sync master device, as well as offering the fullest ADAT recording/playback experience that no other controller in the world offered!

Here is a fun one. I was listening to the pre-mastered mixes of a project I did long ago on the DAT machine it was mixed to. I then listened to the CD. Guess what? The CD version (post mastering) was faster and higher pitched! Oh my.....

I never until that time considered the drawbacks of using another machine to play the DAT tapes during mastering. On this project, the DAT was played on a Panasonic machine which went to an Apogee D/A, which fed the analog compressor and eq. The compressor (it was after the eq, just to clarify) was fed to a Apogee A/D, which inputted to ProTools. The DAT was recorded at 48KHz, and ProTools was recording at 44.1KHz.

Anyway, since the Apogee D/A has to slave to the Panasonics clock, I am figuring that that is where the speed up happened. I have a Fostex D-5 DAT recorder, and I would bet a million bucks that it's sample clock is a little slower then the Panasonics was, thus the speed up of the music and higher pitch!

Weird eh? It was something that I never considered to be a possibility a couple years ago when we mastered this stuff, that the sample clock on two different devices could be that much different. But I can see no other possibility to explain what happened here, unless, oh boy, the duplication plant did something weird when extracting the music from the mastered CDR we sent them. I don't have the mastered CDR to compare to the DAT mixes, so who knows. Maybe THAT was what happened. But I am betting on the Panasonic having had a slightly different sample clock (talking out my ass here?).

If I ever have to mix to DAT again, and the material is going to be mastered in the same manner again, then I am going to bring MY DAT player to the mastering session to remove that possibility to the mix.

Isn't syncing fun stuff?

Ed