Advice on setup

[Dennis Brioso]

Hello everyone, longtime lurker here. I've been thrown into a job that is a bit over my head and I was wondering if anyone could give me a few ideas on how to tackle it. The scenario is this: It will be a two camera shoot of 8-10 people sitting around or in front of a dinner table. I am to provide audio to a hard disk recorder and provide timecode to both cameras. I think one camera will be a closeup camera and the other will be a reaction camera. I have been given very little information on the shoot so far. My initial idea is to chain to wendt x-5 mixers together and just lav everyone and feed it into a fostex or sound devices recorder. I have been told by the rental house that these recorders can output timecode. So I figured I would jamsync the cameras at the start of the shoot and then just feed the audio to the recorder and mix appropriately. Can anyone think of a better way to tackle this given the limited information so far. The shoot isn't until Thursday night. So far I am the only audio guy on the shoot so I have to take care of all of this myself which has me a bit nervous. Would multiple boom mics work better (assuming they could be kept out of the shot). Any and all ideas would be helpful. I will hopefully get more information tomorrow but so far this is what I have to go. Thanks in advance

[Dan Keaton]

Dear Dennis,

Can the microphones be visible in the shoot?

Is the shot scripted and staged or is it an impromptu roundtable discussion?

Using lavalieres on each person can be a problem, if they are mixed down.

If the people are close together, one lavaliere will get the sound from that person and the sound from people around them. If you mix these lavalieres together into one channel, you may have problems.

If this is a roundtable discussion, with no script, then you will not be able to predict who is going to speak next.

Depending on the quality of sound that you need, a good hypercardioid, such as a Schoeps CMC-641 could be setup over a portion of the table. You will need more than one to ensure that you cover all of the individuals.

I believe that recording into separate channels for each microphone is the best to ensure that you will have audio that you can use.

I welcome suggestions from others.

A short shotgun, instead of the hypercardioid, would not give you as good a sound.

I would rent four CMC-641's, then use four good boom microphone stands, such as the Atlas EB36w (http://www.bhphotovideo.com/c/produc...one_Stand.html) then record each channel separately into a Sound Devices 744t.

You will need a mixer, such as a Sound Devices 302 or 442 since one 744t has only two microphone preamplifiers.

One problem with the above suggestion is that with four boom stands and two cameras, the boom stands will probably be visible. To avoid this, you may be able to suspend the microphones from the ceiling.

If the microphones can be visible, you could set good micrphones on the table, but you may have noise from the individuals hitting the table.

You could set the camera's timecode to 24 hour time of day timecode. Then just jam sync the cameras from your recorder. I know that the Sound Devices 744t can do this, if you have the right cameras and the proper timecode cables.

If your shoot is scripted so only one person speaks at a time, then I would use one Schoeps CMC641, mounted on a boompole or good microphone boom stand.

[Wayne Brissette]

Originally Posted by Dennis Brioso

Hello everyone, longtime lurker here. I've been thrown into a job that is a bit over my head and I was wondering if anyone could give me a few ideas on how to tackle it.

There are actually a lot of factors that are unknowns here. OK, so you have 7 or 8 lavs, connected via two Wendt X5 mixers. You have a two channel output going to a Fostex or SD multi-channel recorder. Since you're only recording two channels that does make life a bit easier. But let's talk timecode first.

Be prepared to ditch timecode. A majority of gigs I've done recently have been done on cameras that won't take timecode. I think it's more of the way productions have been cutting costs by using more of the prosumer/low-end HD cameras, which run their own TC and won't accept TC from any other devices. If this happens, just slate it and go (you'll have to double slate for since you're dealing with two cameras). But your recorder should be the master timecode device if you do end up with devices that can be jammed with timecode. Don't loose sleep over timecode, there are features that are still hand sync'd using the slate. It works and works very well.

OK, now onto audio. Do you have a boom op? Do you have two? If you have two I would be tempted to use them in addition to (not as a replacement to) the individual lavs. If you have the option of providing a boom, with this many people you really need two boom ops. You might be able to get away with a single boom op if they are really good (or really lucky, as I was in an 8 person interview when I was the boom op). If you choose to go this route you need to put them on a totally separate channel. Since you're going two channel, pick either left or right and stick to it. Put booms on the left and lavs on the right. These will be feed to the SD/Fostex recorder, then sent out to both cameras.

Levels. Make sure you set your levels by using the tone button on the Wendt, set the SD/Fostex levels to match this, then check each camera and ensure the levels there match.

I'm sure it will be crazy with that many people mic'd up. The TV pilot I worked on for MTV last fall had 10 people in one scene all mic'd. With a two person crew (sound crew) it was all we could do to keep things going smoothly, but we managed it, and it ended up sounding great.

Hope this helps. If you have more specific questions, post them.

Wayne

(I just read Dan's post and he is correct about the lavs. They can be problematic when in close proximity of each other and not on separate channels -- I'm spoiled by the Deva and forget things like that.)

[Dennis Brioso]

I'm going to push to try and bring at least one more guy onto the job, if not two. You guys are right, I'd much rather boom this thing but I'm still not sure if that is feasible (again I'm working with very little information here) The piece is being shot with Varicams so they will be able to accept timecode, but I have no idea if the mic's can be visible or not. I will get more info tomorrow and probably seek more advice. Thank you all so much. This job is a step up from what I normally do and while I am a bit intimidated, I believe I am up for the challenge Again I will post an update tomorrow.

[Bernie Beaudry]

If you choose to go with lavs I would highly recommend renting two auto mixers. Either two eight channel units or an eight and a four. This many lavs could be mixed manually but you would have to be very very good plus lucky. The auto mixers open up the mics as they are talked into and leave the last one open so the background doesn't change dramatically. Any ac noise in the rooms isn't as much as an issue. I've done shoots with up to 25 participants this way and it makes life a lot easier. The real challenges are then getting everyone miked up and the hardwire cables routed so they aren't making a mess on camera. Multi line snakes are helpful for this. This all presumes that exposed lavs are acceptable. Make sure that all of the lav brands and models are the same if possible. I don't have any suggestions for time code or recorders. With this method you could easily feed the same mix to both cameras and not even use a back up recorder unless you really think you need it. Two people would be advisable just so the participants can get miked properly and quickly.
Good Luck!
Bernie

[Dan Keaton]

If the lavaliers have to be hidden, then the problem of picking up clothing noise will arise.

Lavaliers are more reliable when:

1. The individuals are not close together. This does not apply if each lavaliere is recorded on a separate channel and only one person talks at a time.

2. They are not hidden under clothing.

I have not found a foolproof way to hide a lavaliere and not get clothing noise. (I would love to learn the best way to do this.)

Lavalieres under heavy clothing will sound muffled.

[Bernie Beaudry]

Originally Posted by Dan Keaton

If the lavaliers have to be hidden, then the problem of picking up clothing noise will arise.

Lavaliers are more reliable when:

1. The individuals are not close together. This does not apply if each lavaliere is recorded on a separate channel and only one person talks at a time.

2. They are not hidden under clothing.

I have not found a foolproof way to hide a lavaliere and not get clothing noise. (I would love to learn the best way to do this.)

Lavalieres under heavy clothing will sound muffled.

I agree, if the mics can't be seen then don't try to hid the lavs under the clothing. It would take longer to rig them and chances are the many wardrobe possibilities would kill any chance of rustle free sound. The auto mix idea will reduce the chance of the lavs interacting with each other. If this is a dramatic scene and your can suspend four hypers or cardiods from the ceiling and record them to four separate tracks as suggested above, this would give you nice sound as well. I did a free form discussion for a documentary around a dining room table with one Schoeps hyper suspended from the chandelier over the table. In this case it was five people in an apartment. I was very pleased with how it sounded. You do need to be sure you use really good mics so even the off axis sound is useable. If you can use stands to rig the mics go for that unless there is an easy way to hang from the ceiling and you're sure where people are sitting. The worst thing is to have everything set, have the participants show up, and the director decides to move things around for picture. Now you have to re rig the ceiling mics. Been there, done that. Isn't fun.
Bernie