Multitrack recording and editing
 

Hi,

I want to make a low budget movie with several children having unpredictable dialog in high mobility situations. So I was thinking about recording each actor on a different track and synchronizing and mixing the tracks in post (like Robert Altman did). Until now I always recorded in camera. I have a Z1, two G2 wireless systems, a 302 mixer and a boommic. I’ve got two basic questions. How to record the multiple tracks (about 6)? And are there any problems bringing them into a NLE, synchronizing and editing them?

About recording. I could use more wireless systems. To record this I could try to power a studio multitrack recorder (like the 8 track Korg D888) in the field but that seems to limit my mobility to much. I could try to use a laptop and some kind of audio interface; not very mobile either it seems to me. I could use field recorders (a Deva is out of my budget however), like two SD 7 series recorders linked together, or maybe an Edirol R4 linked to some other recorder if this were possible.

However, instead of buying more wireless systems I could also use separate mono recorders on each actor (like zaxcom zfr100). That looks like the most mobile solution to me, and saves a lot on wireless systems, but nobody seems to be doing it. Still it is the solution I’m most looking at, if I could get some form of remote control and monitoring. I could use one ZFR100 with stereoadaptor (or another recorder) together with the 302 mixer. On this I would put the boom and one actor’s G2. The other G2 I could use to send a wireless reference signal of the boom to the camera. Then I would use 4 ZFR100’s to record the other actors. Maybe I could also use an IFB800 remote control to jamsync all the recorders time code and hopefully this would also permit me to start/stop them all together. I have to check out if this is possible though. Anyway it wouldn’t be absolutely necessary to use this. Monitoring the signal would be a possible inconvenience I imagine. I’ll have to make sure the lavs are properly installed once in a while or get a Zaxcom wireless receiver to monitor the signals. At least that’s how far I got in understanding this route. It sure sounds attractive to me (high quality audio, 24 hours with 2 AA’s, 8 hours recording on SD card, and very mobile) but my experience is limited. Maybe I’m overseeing something. The cost of all this would be about the same than if I would buy extra wireless systems and recorder(s). Can this be done? Is this the future?

Synchronisation with this camera is not so obvious apparently. I’ve been reading about this on this website. The problem is that the Z1 does not have time code out, so I could record a reference track or maybe use a smart slate for this. But I would at least need some form of time code synchronization between all the audio tracks. How are people doing multitrack recording?

About editing. I read somewhere that a NLE can typically only import a stereo pair of audio, and that it is unlikely that the synchronization between two pairs will be accurate enough to mix them because of tiny errors that can create echoes or hollow sound. So maybe letting the recorder(s) have a running time code, like ‘time of day’ to synchronize all audio tracks would not be sufficient, I would still encounter problems in the NLE? Has anybody done this? Is there a workflow that makes this possible?

I'll be glad to hear your opinions.

Johan Bollen

For low budget work I have had good results running BoomRecorder on a MacBook Pro through a MOTU Traveler interface.

Good Luck with your project!

This seems not so different than any other shoot with a Z1 that uses an additional recording device. Yes, time-of-day timecode will get you to close sync.

I would highly recommend that you go through your complete workflow to see everything work before the first shoot.

I've used the Z1 and a Sound Devices 704 for this. The 704 and similar boxes stamp the start time into the header of a BWF file.

The workflow for Vegas, which I use, involves displaying file TC on your video file, placing it on the timeline at the starting TC, then importing the BWF file. During import of the BWF Vegas will allow you to select placement on the timeline corresponding to its header timestamp.

Then, select the audio clip on the timeline, and do fine sync while listening to it in comparison to the camera's audio track, using numpad keys 4 and 6 to slide the clip left or right. For fine work, turn off "quantize to frames" (don't forget to turn it back on before cutting video!)

Repeat for additional BWFs. Vegas is pretty good at this workflow, once you've got it figured out you can go pretty fast. However, it is fair to say that this is advanced editing technique if you're not familiar with NLEs.

A good sound recordist can mix your 6 wireless mics while recording. Whether recorded to camera or to an external audio recorder this is how it is usually done. I've never seen a battery powered rig in use for more than 4 tracks, perhaps because you do need individual level trims for each mic, and as the mixer gets larger and the record deck gets larger AC mains supply and a cart makes more sense. Don't forget that if the sound recordist can't monitor each record channel individually, you don't really know what's going down in the recording.

David, I have thought about this solution but I wish it would be more portable, something you hang around your neck. We will be filming them in sometimes difficult circumstances while their talking, like walking in the snow on a mountain, coming down a river in the jungle, in a busy market street. The soundperson will have to follow them sometimes while walking.

Seth, thank you for your explanation about the Vegas workflow. Very useful. I have some experience with this NLE but not with this procedure involving time code stamped files. But what about these echoes I was talking about. I read about them in a Jay Rose book. What if I would have, let’s say, 6 tracks of audio, one for each actor. Each of the six audio files imported the way you described. People would be talking more or less at the same time (stepping on each others ‘lines’), and I want to bring up a little one word of one actor one second, then another word of another actor etc…but never die out a track completely. Will I have problems if I want to do this? I don’t want the soundperson to do the mixing in the field. It would be difficult to do given the circumstances, and I want to do it myself in post. I want a clean lavaliere track of each actor and a general track through a boom. Well…unless someone can convince me it can’t be done easily.

I am thinking about the advantages of recorders versus wireless. Recorders would be more mobile (you don’t have any receivers), the audio would be of better quality (except when using very expensive wireless gear), no problems with radio interference or not allowed frequencies, less batteries to use (that would be 24 AA’s for 4 G2 systems against 4 AA’s for 4 ZFR100’s during the same time period, if I calculate correctly). One of the disadvantages I see is that you don’t have an immediate feedback if the recording is interrupted (battery dies, SD card ejected of full for example). So you have to check the recorder once in a while, put in fresh batteries and SD card each day for example, and see if the red and green light are burning when they need to. Also there could be circumstances when you can’t have a boom or plantmic close by and the actors are at great distance so you can’t follow the dialog. Not very common but not impossible. Does anybody see other problems? I haven’t decided yet. I would go with wireless though I would need also extra recorders, and though I would love to have a SD 7 series recorder (or maybe two in this case) maybe there are other ways…

Johan

Hi Johan...........
 

I can think of a simpler way of going about getting your 6 tracks, but haven't got an elegant answer for synching them.

Camera: 1 wired mike plus one Senn G2 (2 tracks)

Zoom H4 No. 1 : 1 Lav + 1 Senn G2 (2 tracks) [H4 carried by one of a pair of actors]

Zoom H4 No. 2: 1 lav + 1 Senn G2 (2 tracks) [H4 carried by one of other pair]

The only was I can think of to sync them is by audio cue (clapper etc). Hey, it's a bit Heath Robinson, but just might work.

Cheers,


Chris

boom recorder.

Ty Ford

This is major sound editing. I'm not trying to convince you of anything other than you should try it before engaging in the complete production. Echoes need to be controlled in post by fine synchronization and by turning down, muting, or splitting and trimming the mic track clips not in use. This is a very manual, *highly* time-consuming process. The time this takes depends on the style of production and editing, as well as quality of the recording and the skill and experience of the editor.

That would be 16 batts for 4 G2 systems, 8 batts for 4 ZFR100.

With actors carrying ZFR100s, or other recorder, you also don't have immediate monitoring of overmodulation, a mic plug falling out, a mic falling into the wrong position, or other mic/recorder failure. This is only ok if the recording doesn't matter very much.

This technology is meaningless without supporting workflows. Just because you *can* plant a recorder on 6 kids doesn't mean that it's even close to a good solution. More gear is no substitute for experienece. I continue to suggest that you test these ideas so that you'll get some experience, or, hire a sound engineer to help you as you put the field recording approach together. Don't buy what you can rent.

***edit, see Ty's comment below for clarification***

Sorry. I thought it was "know." Boom recorder software for a Mac.

Regards,

Ty Ford

I like the creativity of your proposal Chris. It’s kind of a mixed solution, but I still would like some time code capable recorder. A stereo ZFR100 instead of a H4 might work. I don’t want there to be a cable between the camera and the soundperson to be more mobile, so more channels elsewhere needed. Some of the children would be walking around with a G2 receiver AND a recorder. That bothers me somewhat.

About boom recorder, I’m not convinced Ty. At least it would give me 6 tracks in perfect sync. That’s a big plus. Could I import the files in Vegas? Is there an alternative for PC? Still, as I said before, it doesn’t seem to be mobile enough for this project. Can you walk around with it while recording (the notebook, Motu, batteries, receivers…)?

In general I like buying more then renting. Being in Bolivia I don’t have many possibilities to rent and experiment with the different systems before I buy. Yesterday I tried to sync two audio tracks recorded with two different cameras, no time code, and I could get rid of the echoes, and yes it would be time consuming doing this with 6 tracks for a whole movie. It’s time consuming but possible it seems for someone with a PC and an NLE. I still would want the audio tracks to be in sync with each other as much as possible though before I import them into the NLE.

I’m just trying to work out a workflow for this project. And because I don’t have the necessary gear already I started considering the possibilities, even the uncommon ones. That’s how the issue wireless against ‘local recording’ came up. The biggest argument against local recording would be the monitoring of the audio signal. It’s a reflex to want to hear what you record as you record. I would give some of that up for higher quality audio and more mobility. A different workflow has to be designed. Controlling the lavs before and after several takes etc…. Wind noise or clothing noise would be issues to be concerned about. But it’s not so bad in the end I think. The ZFR100 has a limiter, while my G2 transmitter doesn’t have a limiter, and I can’t change the levels of it while I’m recording either. Best of course would be to do both local recording and transmitting the signal for monitoring. The ‘tranceivers’ of Zaxcom can do this but to expensive for me right now.
I read someone on this website saying that in NYC the common wireless systems (like the G2) would work only a few feet. This will only get worse in the future I think. Also people are complaining about how horrible their microphone can sound when put through one of those mid priced wireless systems. The idea to do ‘local recording’ is inspired by my desire to get ‘film quality audio’ and mobility at a reasonable price, not to try something new just because it can be done. If I wasn’t surprised by this project I probably wouldn’t consider buying ZFR100’s or new G2’s. I would save my money and little by little buy the real good wireless/recorders combos.

I have difficulty finding the information I need to make my decision. I was hoping the IFB800 would be a remote start/stop control, apart from sending out time code to sync all ZFR100 recorders. (That would create files of equal size with same time code stamp, I hope), but I still don’t know if it does that. I also hope one Zaxcom receiver would give me monitoring of all the recorders. I was told this could work by a very friendly Billy Sarokin, who has experience with the Zaxcom systems but not with the ZFR100. The Zaxcom website doesn’t help much either, and until now I didn’t get an answer to my email to them. Nobody seems to have actually tried this, and I can’t try it from here. So I’m starting to look at a more ‘conservative’, somewhat less ‘elegant’ approach to get the 6 channels of audio…

I could get a 4 channel recorder (like the SD 744T) and put the boom, and three wireless systems on it. That would mean I get two new channels of wireless, mabe the AT 1800 double receiver system. I could send a reference signal of the boom to the camera. I suppose this can be done with the 744T. Then get two ZFR100 recorders for the last two channels. The 744 could jam the time code to the ZFR100’s once in a while. I’m thinking about not using the 302 as a front to the SD recorder. I’m not sure if it will be necessary. It will add weight. Some children would have recorders and some wireless, so it’s not a perfect solution, more Salomon kind of solution hé? I hope it will give me some flexibility afterwards, than going all mid priced wireless or all ‘local’ recording’.

So that’s where I’m standing for now. I hope you don’t get bored with my writing. Feel free trying to change my mind. It works.

Johan

Oh, 2 AA's power a ZFR100 for 24 hours. To power a G2 system for 24 hours you would need 12 AA's. I think my math is correct now.

Well, if there is a 6-channel A-D converter that would work for you on the move, you could record directly into Vegas on a laptop, FWIW.

Quite right.
Were it me... I'd probably do something like 3 wireless channels and one boom into a 744. Wireless link to the cam is OK, too, but I'd have a camera-mounted mic going on the other camera channel at all time to serve as a backup for syncing. Then, I'd figure out who the primary speakers were, give them the wireless, and boom the whole thing as well. I'd definitely use the 442 mixer as well - I'd not go into a multichannel shoot without some input mixing and limiting. Besides, you need to mix for the link to camera, and you need to monitor. I'll continue to say that unmonitored record channels are only good when the audio doesn't matter very much.

Be aware that the 744 has a total of two mic preamps. Channels 3 and 4 are line-level input.

Don't forget a good mixer bag system and battery system for it. A bag, battery system, 442, 704, boom mic, and handful of decent wireless mics is an extremely versatile system for all kinds of field production.

Me, I don't plan on shooting in NY, if I do, I can rent locally. In the meantime, my Senn. 100g2, Senn.500 and Sony 800 series wireless work just fine. I suspect open freqs can be found, even in NY.

But what would you do then if you wanted to record two extra channels? Extra wireless systems and a two channel recorder with time code jammed to the 744T? I read that the SD 7 series recorders can be linked digitally. So that would give me 6 synchronised channels. Maybe other recorders can be connected to the 744T permanently with a time code cable so they would be in sync allways. Is that possible?

Thinking about immediate future use I'm coming to the same conclusion about versatility given the current state of technology and existing workflow habits.
I think I will pass on the 442 for the moment and prefer to find a way to incorporate the 302 I already have if it would be necessary.

Glenn Sanders of Zaxcom responded to my mail about the ZFR100 recorder and the IFB remote control. It thought it might be interesting for some to know that according to Glenn Sanders the IFB800 can send out time code and start/stop commands to different ZFR100’s at the same time. The different files created this way will align in the NLE. The recordings would be exactly in time so there would be no echoes. This makes multitrack recording possible in new and easy ways I think.

Johan Bollen

1- Do you necessarily need to sync by jamming timecode? (Which is not a fool-proof process anyways.) It might be easy just to continuously roll the camera + audio recorder(s) and use a slate. You'd sync the sound manually and that shouldn't be difficult at all to do in post. Have the person holding the slate say what scene/take it is so you have your audio notes in your recording.

You can get problems if you mix different wireless units into the same channel. If the two actors are standing close to each other, then you can get phasing issues. Two mics are recording the same sound with a slight delay [phase shift], but with similar volume... that's where the phasing issue happens.

If you record multitrack, this shouldn't be a problem unless your NLE can only import two tracks easily. But the thing to do is to get an NLE that can handle multiple tracks of audio (or figure out some sort of workaround).

2- Perhaps an alternative solution would be to:
Have 2 boom mics... each boom covers a different set of children.
Put an audio recorder on each child/talent... something like an iRiver or SonyMD might work?? You can manually sync in post. (This would be cheaper, though likely not as foolproof as the Zaxcom recorders.)

i would use wireless sytems with receivers connected to a firewire mixing table (like the alesis multimix firewire) and a laptop recording the 6 chanels (or more) separately, so you can remix later if by any chance something is going wrong.
i would also send the stereo output of the mixing table to the camera with a digital wireless transmitter, so if you are lucky, the audio will be fine directly on the tape and you can monitor it from the camera.
provided you do not purchase a laptop only for this use, the cost should be minimal, except for the wireless mic, depending the quality you are looking for.
(here rental will be easy and cost effective).
The Zaxcom ZFR100 is priced around 1000$, so for this price you can get 2 wireless SONY UWP-C1, long range, diversity, high quality.
And this setup will be highly reusable for other situation.

http://www.alesis.com/product.php?id=93
http://www.adorama.com/CTAIPAST.html
http://www.totalaudio.co.uk/products/wireless/uwp.htm