Movie Making Manual/Sound Recording
This Module is part of the Movie Making Manual
||This page or section is an undeveloped draft or outline.
You can help to develop the work, or you can ask for assistance in the project room.
Types of microphone
- Dynamic: Most common / cheapest. Dynamic Microphones have a metal coil within a thin plastic diaphragm which vibrates beside a fixed-charge magnet, generating a voltage. These microphones do not require an external power source.
- Condenser: Most require Phantom Power. Known for producing more detail, especially in the upper frequencies. Condenser microphones use one fixed and one movable electromagnetic plate, the vibration of which generates a voltage. These mics require external power to charge the electromagnets.
- Shotgun: A subtype of condenser mic that is used in filmmaking to catch dialogue during filming. Shotgun mics are generally designs which use a long, notched barrel to increase side and rear rejection, picking up more sound from the front of mic.
- Ribbon: Ribbon Microphones use a thin charged ribbon beside a charged magnetic plate to generate a voltage. Was rare, becoming more common. Catching speed as being 'smooth' on guitars. These mics require external power to charge the electromagnets.
- Also: Mic Patterns
* Omnidirectional: pickup sound in a circle around the mic * Cardioid: pickup sound better directly in front of the mic, less sound pickup as you move around the sides of the mic
Electronics you should be aware of
Balanced versus unbalanced
Balanced audio connections use 3 wires for sending each channel: a positive phase, a negative phase, and a ground. The sending device sends its output over both the plus phase and the negative phase. But (and here's the clever bit) the signal is inverted on the negative phase. The receiving device then inverts the negative phase back (to match the positive phase) and adds it to the positive phase. Any noise picked up along a stretch of wire will be out of phase with itself, (once the destination reverts the negative phase) and be cancelled out.
Phantom power is used for condenser microphones that need power to run. It is also used to power direct boxes. Direct boxes turn unbalanced signal into balanced signal for reducing and preventing hum. You will most often see phantom power in 48v. You can get phantom power from a source such as many sound consoles, especially if you have a recording console. If you don't have it built into your console, you can buy a portable unit that takes up an outlet and has an xlr in and xlr out(with 48v power).
As a quick rule of thumb for a TV mix, dialogue should average 3.5/4 ppm on a meter i.e. around -20/-18dB. You should record it like this and mix at these levels. Peak dialogue shouts etc shouldn't really go above 6 ppm (-10dB). If you are recording or mixing for TV nothing should peak over 7ppm (-6dB). And those peaks should last a minimal amount of time. Music is mixed at the dubbing mixer's discretion, so that it works with the programme and fits with the dialogue, but shouldn't peak above 6.5ppm. A limiter would usually be placed across the output to ensure nothing goes above 7ppm.
If you are mixing for DVD or Theatre release you would use the full digital scale. Dialogues should still average -20/-18dB however your fx peaks can go the full range to 0dB but not over!
So you would have two mixes one for TV and one for DVD.
the BBC's audio delivery specification states:
"Peak sound levels (when measured on a BS 5428 Peak Programme Meter) within the programme must not exceed PPM 6: that is, no greater than +8dB with respect to the reference level of -18dBFS."
In other words there should be no less than 10db of 'headroom' between the highest level you can record and your programme's highest peak.
You can find the full document at: http://web.archive.org/web/20050312163940/http://www.bbc.co.uk/guidelines/delivering_quality/pdf/tv/tv_standards_london.doc
It is preferable to have sound with "hiss" rather than distortion. Distortion cannot be fixed, background noise can be somewhat fixed.
What to record on?
The video camera
I've recorded onto camera quite a few times - the skinny is that you can get great audio from the cameras with XLR inputs and also:
The Canon XL1 does not have phantom power - you need an external mic power supply.
The DVX-100a has small audio controls which are more difficult to manipulate, a little movment seems to make a big difference in level
One of the two DVX100a's I've used had a buzz only when the DV tape was installed. Very annoying, but the other one was quiet.
The Sony PD170 has a good set of audio controls - especially the ch1/ch2 link function when using one microphone (two channels are better than one!)
The PD100 has needs a +48/XLR hot shoe attachment, otherwise it sounds fine.
It's wise to back up audio, and slate scenes without pausing, when using DV cameras - back up on DAT. On a recent 3 camera shoot, I had a total of 6 audio tracks, one into the DVx100a, one into a Canon Optura PI (fine little camera!), but with 1/8" connector and one into a Sony tcd-08 DAT - the DAT won as the best overall sound with the lowest noise and cleanest sound. BM
DAT-Heads concerns itself with Digital Audio Tape decks, with an emphasis on their use for the recording and distribution of live music.
ATRAC 3 is a proprietary lossy perceptual audio codec developed by Sony and used for minidisc and other Sony products such as the PlayStation Portable. Uncompressed PCM offers higher quality, and therefore field recording with a portable DAT or harddisk recorder is almost always preferable to a minidisc recorder. Personally, I'd trust a tape (DAT) more as I have accidentally recorded over material with the minidisk's chapter functions. It's easier to track down ABS (absolute time code) on a DAT, especially for sound logging purposes. DAT = Digital Audio Tape
Hard Disk recording has quickly become the industry standard (aside from analog tape.) Benefits include: Non-linear editing - no need to rewind or fast-forward to find different songs. Cost - You can buy a hard drive with well over a hundred gigs, for around a hundred bucks. 100 gigs can hold 300 hours of uncompressed single-channel audio, making it a very cheap way to record.
With the advent of mp3 players, portable audio has become a household convenience. Some devices can be slipped into a pocket or clipped to a belt, but certain recorders are more bulky and cannot be carried as easily. Take this into consideration and make sure you have, if nothing else, long enough audio cables to allow your recordist to move around as he captures sound during your shoots.
When on location you should always record a "wild track". This should be a few minutes of the ambient noise from your location (the sound of traffic in the distance, birds singing etc). You should ask the assistant director to tell the entire crew to be quiet so you get an undisturbed track.
Why bother? In the edit it's very useful (and often absolutely required) to keep a wild track running in the background to smooth over the changes from cut to cut.
A Wild Track is an audio only recording of an actor's lines while the camera is NOT rolling. There are no sync marks or visible slates. These wild recordings are used to replace dialogue from a filmed take that may be unusable for various reasons. Perhaps a plane flew over during the scene. The Wild Track Or Wild Line will be used in part or in whole to replace the on-camera dialogue. It is also used for Off-camera lines that need to be recorded on-mic. Many times wild Lines are used to replace compromised recordings made during a Wide master shot where the microphone could not be placed in the optimal position. Because the shot is wide, lip-sync of these lines is not as critical and sync errors are not as visible as for a close up.
"Room Tone" or "Presence" is a wild recording of just the ambient sound of a location (no one speaking). It is used in post production to cover unwanted sounds (like the director's verbal instruction during the take). To record Room tone you should place the same microphone in the same position as it was during the dialogue take. Record at the same level used for the actual take but have everyone remain still and quiet.