Location Recording Q&A

I get quite a few questions sent to me about various aspects of sound recording as a result of this blog, and i’m happy to answer them as best i can. I thought i’d start posting some of them here for public consumption. Most of the answers here are based on how I work - they are not definitive answers, nor are they always right, but i hope they are of some use to others engaged in recording location sound. If you have a different answer to any of the questions here, please let me know - i’m always keen to learn new ways of doing things.

Anyway, below are some recent questions and my answers…

My main question is really about the sample rate / bit depth that you record at for a TV/film shoot. For my own field recordings/foley work, I tend to record at 192KHz/24 Bit. But I’m guessing location recording needs to be captured at 48KHz/24 Bit. Is that correct? Is there any time that you would record at a higher sample rate?

Standard sample rate/bit depth for broadcast tv is 48khz/16bit, and thats what i do most of my recordings at, whether on a hard disc recorder like my 744T or on-camera sound as most cameras i tend to work with (eg the DSR’s, f900’s, red cams etc) will record at this setting. It sounds good to me, and unless i’m in a studio with some very good monitors i can’t tell the difference between that and any higher settings, and even then it’s pretty subjective, unless you have a pair of golden ears!

The only reason i would tend to change those settings is to add more headroom for the post production process; in which case i’d simply shift up to 24bit. At 24 bit you will get about 20db more headroom without adding any obvious noise, compared to a 16 bit setting. It’s often a question of file size, and if you’ve got huge files it can be a pain in the post process. I will sometimes get a call from the sound post-production supervisor or similar to discuss this sort of thing, and i’ve never been asked for other sample rates or bit depths.

Do you always give a feed to the camera and capture audio on both your SD 552/744t and the camera, or are you only using your equipment to capture audio with timecode. Am I right in thinking that you wouldn’t need to use timecode if you gave a feed to the camera?

There are a few answers to this, and it is largely dependant on what i am shooting; the short answer is that i do either/or, not usually both, and yes if sending sound to the camera, a separate timecoded recording would not be necessary. If sending sound to the camera thats the master recording, otherwise i’ll do a separate recording on the 744 with timecode jam-synced to the cameras timecode

For most of the tv stuff i do, i record all my sound directly to the camera via a 10pin hirose umbilical cable connected between mixer and camera. It’s the easiest/safest/cheapest way to do it, but it does have a couple of drawbacks - the main one being that you are tethered to the camera which can limit your range of movement and also prove a bit annoying in tight locations where the cable can snag etc. Camera operators don’t really like it, particularly when someone stands on the cable and they are jerked backwards via the rear xlr’s! It’s also a bit worrying when you turn round and realise that the camera you are tethered to is on the other side of the road shooting something, and your cable is across the road in the line of the oncoming traffic… anyway, i digress.

The umbilical cable is the most reliable connection between mixer and camera, and if i am working this way i don’t tend to make a backup recording on the Sound Devices 744t or 552. It’s not really necessary. I did at first (professional paranoia!) and it can’t hurt to do so, but you shouldn’t ever need it. I am constantly monitoring the return feed from the camera (usually via the phono outs on a camera, or at a push the headphone output - i have either connector on the camera end of my umbilical cable, and if you switch the headphone monitoring on your mixer to the camera return you will hear what you are sending to the camera)  and if there is a problem with the connection between the two, or the feed itself i should know about it immediately.

The sound that is then recorded on the camera is your master recording and is used in the final film. At least it has been through the mixers high quality preamps before being recorded onto tape/HDD, and there is no additional fiddling in post to sync sound and picture. A lot of cameras these days have 4 channels of audio, so if needed you can still send 4 isolated tracks or similar. This is the standard approach when shooting video.

For film, and for situations where the camera sound is not high enough quality (eg Red One Camera) or you need more flexibility and don’t want to be attached to the camera there are other options: the first is to record sound locally (ie in the bag) on something like the 744T. To do this you need to jam the 744’s timecode with the cameras timecode - a BNC/Lemo cable is connected to the camera’s timecode output and the other end to the 744. You then ‘jam’ the two timecodes so that they are running simultaneously. The 744 will usually run for a day or so without drifting by more than frame, so i do this most mornings first thing if working in this way. Then in post you can simply sync the two sets of rushes via the timecode - easier and more accurate than a clapper board. The SD552 recording function on the mixer doesn’t generate it’s own timecode, and must be fed incoming tc if needed. This means it’s not as flexible as the 702/744 for recording separate sound that must be synced. I often use the internal recorder on the Sound Devices 552 simply as a wildtrack recorder, or as a backup in a desperate situation. With the addition of a Ambient Lockit box or similar, it could be used as a primary recorder with some limitations.

Also when you’re recording separate sound onto a 744 or similar, you have to get the rushes off the device and to the post-production team at the end of every day - so either a laptop with firewire and a dvd burning capability or a dvd-ram burner attached directly to it is needed. You also end up storing GB’s of old rushes ‘just in case’. i like using the 744t in this way but it’s not always what production want. I keep it simple - if i can get the sound on to the camera without compromising it, then that’s what i will do.

My last question is about your experience of mics. You talk a lot in your blog about using lapel mics. Do you use these a lot or do you mostly boom?

Again, it’s a case of horses for courses; i do use lapel mics a lot but thats the nature of tv quite often. For example i do a lot of documentary and reality tv where you don’t know what is going to happen. You will probably have anywhere from 1 to 5 main contributors on screen at once, and keeping them all covered with the boom can be difficult. Shot-size is constantly changing during shooting (eg a big wide to establish a location, tighter singles to cover some dialogue, then back out to a wide shot as something in the background starts happening), overhead lighting (as found in many offices and homes) will cause problems with shadows, and reflections in the back of shot are all constant issues to make booming tricky.

The way i tend to work in this situation is to have the main contributors on radio mics panned to channel 2, and then the boom panned to channel 1 to cover all incidental sound, and any other people who come into shot and talk. Sometimes i find myself booming people already wearing my radio mics, but to be honest the boom sounds much better and will probably be the preference in post. Also i sometimes put radio mics on people due to the acoustics of a room (big and echoey acoustics can be reduced by this close mic’ing) or a loud bg noise (eg a piece to camera next to a main road often sounds better in terms of the voice to noise ratio when using a close mic that is shielded somewhat by clothes and body). With a lot of radios being used in the same situation, the sound can often go very ‘thin’ as a consequence of being picked up by all the mics in the room at fractionally different times (speed of sound), so you can often end up mixing between them more than you would if booming.

But that’s factual/doc/reality.

For drama i try and do as much as possible on the boom. radio mics are a last resort here due to the necessity of hiding them under clothing and adding rustle, as well as their inferior sound quality - they have that slightly ‘dead’ sound due to a lack of bg, ambience and acoustical detail. A good boom op is worth their weight in gold in such situations, but the problem is, once you have started to use radio mics in a scene, you are obliged to continue using them in that scene even if it is no longer necessary because the change in sound quality will be so marked if you go back to the boom.

— Posted by Matt in Uncategorized | No Comments