Topic: Matrix sync question

[hardrain62]

Hi gang,

Tried to do a little research on this before starting a new thread, but can't seem to find anything with my problem.

Last night I taped a show using two separate rigs (RODE NT5 > Tascam DR-100 and CMC-8/AT943 > Sony PCM-M10) and I'm now in post-production on what is my very first matrix recording. I basically synced them up (in multi-track mode in Cool Edit Pro) by ear at a point in the beginning of the recording, and to my ear, it is perfectly in sync in the beginning, but then towards about 75% into the show, it goes EVER-SO-SLIGHTLY out-of-sync. And when I mean slightly, it really just sounds like a bit of reverb and not really noticeable. But I'd really like to get it perfect. What have I done incorrectly? Was it really never truly in sync from the beginning?

Would really appreciate any help on this!

Thanks so much!!!

David

[Brian Skalinder]

You're likely hearing the results of clock drift.

FWIW, the sticky at the top of this forum should have links to several threads on syncing two sources with different clocks / clock drift.

[page]

Hi gang,

Tried to do a little research on this before starting a new thread, but can't seem to find anything with my problem.

Last night I taped a show using two separate rigs (RODE NT5 > Tascam DR-100 and CMC-8/AT943 > Sony PCM-M10) and I'm now in post-production on what is my very first matrix recording. I basically synced them up (in multi-track mode in Cool Edit Pro) by ear at a point in the beginning of the recording, and to my ear, it is perfectly in sync in the beginning, but then towards about 75% into the show, it goes EVER-SO-SLIGHTLY out-of-sync. And when I mean slightly, it really just sounds like a bit of reverb and not really noticeable. But I'd really like to get it perfect. What have I done incorrectly? Was it really never truly in sync from the beginning?

It has to do with the clock crystal used in each machine, you would have this even if you had used the same model for both units, it's a unit by unit issue. To fix it, you will have to adjust the speed of one recording to match the other (aka, the "stretch method" or "resample method" depending on where you learn it). I suggest consulting this as a general reference if you don't use Audacity, the math behind it is sound.

posting anyway even though Brian beat me to it while I was composing a small novella and digging up the link. 

[kirk97132]

yep drift... one of the driving forces behind a lot of people getting multitrack decks.  There are a variety of way to solve the problems depending on personal preferance and the DAW you are using.  No matter which it is a pain in the ass and will require some time and patience on your end. 

[mattmiller]

yep drift... one of the driving forces behind a lot of people getting multitrack decks.  There are a variety of way to solve the problems depending on personal preferance and the DAW you are using.  No matter which it is a pain in the ass and will require some time and patience on your end.

It can be done in 15 minutes, especially if there's a point near the beginning and near the end of the recording where the crowd is quiet.  In this case, it's easy to identify identical points in the MIC and SBD sources to use as measurement endpoints.  The more constant the crowd noise, the harder it is to identify these points at the sample level of the zoomed-in waveform.  But in the most difficult of scenarios, it's still no more than an hour of work to get the sources close enough.