Resolving Audio Timecode in NLEs from DAT or File-based TC-Enabled Audio Recordings

[16mmFilm]

I'm wanting to get some equipment for timecode-based double-system sound recording, and at the moment I'm thinking about pairing up a Denecke timecode slate or an iPad running MovieSlate with either a timecode-capable file-based audio recorder (like the Zaxcom ZFR100) or a timecode-capable portable DAT recorder (like the Fostex PD-4). Before making a decision, though, I thought I'd see if someone could help explain the way these systems capture/store the timecode data, and how it translates to non-linear editing.

If I chose the DAT option, I'd need to manually ingest the audio tracks, along with the timecode audio track. If I went with a file-based recorder, my understanding is that the timecode is embedded into the metadata of .bwf files. That said, it seems that going the .bwf file route would be the easier solution, but I'm curious if most Mac-based editors like FCP, or maybe even some basic, open source editors, can easily import .bwf files and translate the timecode metadata.

On the other hand, if I wanted to go the DAT route and had to manually ingest the audio and TC track, are any editors set up to decode / resolve a analog TC audio track and apply the resulting TC to the audio clip? It seems like this approach would easier said than done, and that I'd probalby need to use some third party plug-in or application to resolve the timecode for a non-linear editor.

Finally, for .bsw files that contain timecode data, does the timecode encapsulated in the metadata indicate only the precise frame at which the audio clip started recording (in other words, only a single timecode value), or does TC data exist for the entire duration of the clip? When it comes to applying this to a non-linear editor, I'm curious if the editor just looks at the timecode value recorded at the very beginning of the clip then calculates the TC for the remainder of the clip, or if TC data actually exists for each frame/sample of the entire audio clip.

Thanks to anyone who might be able to offer some advice!

 

[Steve House]

BWF file timecode is stored as the timestamp of the first audio sample. It is a single point and there is no timecode stream recorded in parallel to the audio, thus the timecode provides a single point position reference but does not provide any speed reference. TC data does NOT exist for any part of the stream beyond the first sample.

 

[Rick R]

As Steve stated, most TC recorders use BWF TC stamp. One could record LTC on an audio track however an external TC generator. Some Avid NLEs can read an audio track LTC, convert to a TC stamp and place it accordingly in the TL. Don't know about FCP.
A TC recorder by itself with/or without a TC slate does not guarantee sync. Many TC capable cameras (REDs for instance) can drift significantly or lose sync all together, so TC lock boxes are wanted on all devices. In any instance, ALWAYS clap slate as well.
I don't trust DAT machines. When I was 'forced' to use one, ( I still have a POS PortaDAT in storage) a non-DAT redundant back-up was rolling as well. DAT tapes are a little difficult to come by these days as well, so I wouldn't go there.
I would only use a iPad MovieSlate for log times.

 

[Noiz2]

Wow did I just see someone seriously thinking of using a DAT... in 2014?

DAT was fantastic when it came in but it was always finicky and you probably couldn't find any tapes today. IF you do what to go down that route let me know because I still have a bunch of tapes in a box. I also have a few DAT machines but none are TC DAT's so...

TC, in the digital world, is positional only and doesn't maintain sync. It was used with analog tape for both but it isn't high enough resolution for digital.

If you want sync you need to be using word clock for sync and TC for position.

For post you would need a TC DAT for reading the tape and a way to read the TC stream into the computer while you digitize.

Also unlike most decent flash recorders generally solid speed wise so you will probably get more drift.

All in all a fair amount of work for no real benefit.

Now, going off your user name, you might be looking to use this to film actual film. That is another ball of wax since film cameras are not very stable in the first place. The old way was to have the camera motor locked to "crystal sync" and send a "pilot tone" to your analog Nagra so the deck would track the cameras speed wobbles. Then resolve it in post. Generally TC wasn't used on set but in post. You resolved the analog 1/4" tapes and transferred to mag. After that you were hardware locked by sprockets. When sound moved to digital sound editors moved to video tape so that you could use TC and "BlackBurst" to sync the DAW to picture. Then you would use a "synchroniser" to lock the DAW to a tape deck (mostly DA88's when I was doing it) to lay back units for the mix stage.

I still have a lot of the above since it has about zero resale value. If you want to go there I can make you a deal ;~)

If camera and sound recorded are file based and at least prosumer then I would just use a dumb slate. They are stable enough that you can shoot any normal scene with no drift. You can often shoot over a half hour and not see any drift. Post sync your slate and you are good to go.

 

[Noiz2]

The other thing to note is that once your in the DAW or NLE the TC is of no use. They are all using TC stamps on the files and not continuous TC. Only analog and video tape (DAT is audio recording using video tape recorder mechanics) uses continuous TC for anything.

 

[16mmFilm]

Wow did I just see someone seriously thinking of using a DAT... in 2014?

Yup, guilty as charged From what I've found, most portable file-based, TC-enabled audio recorders are in the $1k price range, so I was considering something a little less expensive. Pre-owned portable TC DATs like the Fostex PD-4, on the other hand, can be picked up now and then for around $150 - $200, so I was considering those as an option as well. I realize that the DAT format itself is borderline obsolete these days, but to the best of my knowledge blank media is still readily available online, so I thought that using a TC-enabled DAT might be a more cost-effective option that would provide TC functionality along with decent quality audio recordings.

Now, going off your user name, you might be looking to use this to film actual film. That is another ball of wax since film cameras are not very stable in the first place.

I am, in fact, shooting 16mm film. Fortunately, I own an Arriflex SR3, so camera frame rate stability isn't a concern. The SR3 has a pretty advanced speed system which can run at crystal sync speeds anywhere from 5 to 75 FPS.

The other thing to note is that once your in the DAW or NLE the TC is of no use. They are all using TC stamps on the files and not continuous TC. Only analog and video tape (DAT is audio recording using video tape recorder mechanics) uses continuous TC for anything.

Good point. That said, I know that TC slates (like the Denecke slates) are still very popular and widely used in many types of video and film production. Assuming that productions using the TC slates are recording audio to file-based systems that may be TC enabled but are just stamping a TC mark at the beginning of each audio file, what's the big advantage to using a TC slate? Is it used more as a general reference to help match up each take with its associated audio file instead of being used specifically as a synchronization method?

Thanks to everyone for all of the information. In the end, I may just end up using a "dumb" slate for my needs and forego timecode altogether. It's certainly a much less expensive option, and would probably help prevent me from unnecessarily over-complicating the audio situation anyway.

 

[David W. Jones]

I'm flashing back to my first portable DAT and all the SFX I recorded with it.

I wouldn't wish that workflow on anybody in 2014!!!

 

[Noiz2]

now the up side with DAT was that like analog transferring meant you got to listen to everything (often more than once) so you really got to know the library. With files it's easy to have a lot of stuff that you really haven't listened to.

16mm

TC was used primarily for music (to lock decks together) and for video since it didn't have sprockets. Film really didn't use TC much. Crystal sync and pilot tone was the norm with a dumb slate.

Even really high quality crystal sync is really sloppy compared to word clock, plus the kind of hunt and seek lock that was how analog worked doesn't work for digital at all. It needs to be clocked VERY consistently or all kinds of bad things happen. So with video recording audio recorders (analog) would slave to black burst and could do the hunt and seek TC lock. All the softness of this is where you have the camera and sound folks calling our "speed". There was an indicator on the Nagra that the machine had stabilized and was at "speed".

When you have two devices that are essentially video recorders you need to lock both to the same sync.

With DAT and a film camera really the best way would be to lock the camera to the DAT. But that is never going to happen. So you will have drift. BUT you are on film so you can't shoot very long anyway so the drift will probably not be bad enough to cause problems.

But wait there's more! Since you are on film, does it have some kind of TC encoder? Probably not and if it does then you are going to need some $$$ TC reading hardware to get it off the film.

So with out that rare TC encoder/reader what you have is a smart slate with a TC number and you are going to manually sync them up.

If you go with a dumb slate you will also have all the info and you will have the slate slap (and your slate person is going to read it so it's in the sound also) and you are going to manually sync the slap. This last is actually easier since you can see the slate on both and you can't actually "see" the TC.

Now you can make things a bit easier by just having a real time clock on the slate and using a flash recorder. IF you set up the recorders clock to the right date and time then you can easily find the take you want based off the file creation date/time.

The other plus with files is you can record at 24 bit, which you can't with DAT.

You are right that you can find some online, and they are a lot cheaper than they used to be! But used DAT machines... Remember they are basically VCR's. So they need to be cleaned and maintained like old VCR's and the heads wear out, and get filled with gunk. If you ever worked with DAT you know it was a love hate relationship. If a tape gets a crease in it the machine will "eat" it at that point when you play it. You can't splice it so with a LOT of TLC you could usually get something off the tape, but not always. PortaDAT had a weird quirk that it would pretend to record but not actually record IF you managed to get a gap in the recording. Most decks were more forgiving but you always have to be carful if you are using a tape that has been used before because you can get these gaps in formatting that some decks won't go beyond. And then there is cost. Yes the deck is cheap (used) and as long as it doesn't have problems in your production then it could be a deal. BUT you have to buy tapes, and you really can't reuse them. So the upside is you will have a backup (as long as you have a working DAT) the down side is it's a REAL TIME file transfer. And then you will want to back that file up.

With the flash recorder you are moving files at file "file transfer" speeds and you can copy to two drives at once. And then you can reuse the card. You really want to keep your DAT's to 60 min or less (longer tapes are thiner and have a lot more eating problems. All the DAT tapes you can buy now are OLD. Nobody is making audio DAT tapes these days so what you are buying is old stock. That is not a plus for tape products.

Really if I were going "old school" I would probably jump back to a Nagra running MagraMaster. It's better sounding (mostly) than DAT and if you are going to go old might as well really go there.