High definition programming is visually striking, but a good audio track will make it unforgettable. Digital High Definition Television (HDTV) broadcasting includes the capability for digital 5.1 surround like DVD’s but with the added benefit of HD resolution for picture.
The start of a good audio track begins on the set. Good microphone selection and placement are always the first keys to getting good audio. Next, the sound must be recorded properly. Video people are used to recording the sound with the picture on the camera. Film people are used to having sound separate from the camera (commonly called “double system”). Now with Highdef, it appears that both are a good idea and don’t necessarily have to be exclusive. A lot of confusion has arisen as the video style and film style methods have collided in HD production.
Camera and Recorders
All high definition camcorder systems have the ability to record high quality audio directly onto high definition videotape. For many productions, this is the only audio recording device that is needed. HD camcorders feature at least 2 tracks of audio and are usually at least 16 bit, 48kHz. The audio is permanently synced to the video which can provide considerable cost savings in post production. When sound is being recorded on the camera, the sound engineer should monitor the sound returning from the camera to make sure it is recording properly. Even if a production needs double system sound it is recommended that you always record sound on the videotape as well. This can be looked at as a high quality backup, a reference for sync checking later, and a locked sound source of audio for downconverts and dailies. Now there are situations where this is not practical, but most of the time having a copy of the audio on the videotape is worth the hassle of connecting an extra audio cable or two.
For double system recording, you can use a Nagra, DAT, DA88/98, DVD RAM or DEVA to mention a few. The frame rate you record depends on the frame rate of the camera that you are using. If the camera is set at 24 frames per second (fps) exactly like a film camera, the audio device should be set at 30fps and record non-drop frame (NDF) timecode. If the camera is set at 23.98 fps the audio device should be set at 29.97 fps and record non-drop frame NDF timecode. This is done because #1 most audio houses want the timecode at 30 frame (29.97) instead of 24 frame, and #2, a lot of audio recorders don’t have a 23.98 mode. Check with the audio house you are going to use if you are unsure. For most non-24P production, match the timecode format of the audio device exactly to the camera.
The issue of free run timecode versus record run timecode is often debated. Free run code keeps going even though the camera is not recording and results in discontinuous code on the tape when the camera is stopped and started again. With free run, there must be at least ten seconds of tape rolled before the action (pre-roll) on every shot for editing systems to be able to lock up accurately. The benefit for this timecode system is that the timecode can be synchronized or “jammed” with other devices recording the same action so that it is easy to sync later because the timecode numbers are essentially the same. This is especially helpful in multi-camera shoots.
With record run, the timecode on the tape is kept continuous even when the camera starts and stops so you don’t need to worry about pre-roll, but because the camera(s) and audio device will start and stop independently and at different times, there will be no common reference. It’s best in any situation to use a timecode slate that displays your audio time code, not camera code, and record it at the beginning of each take. This will be used in post to sync up the audio quickly. Some productions have also fed the audio timecode onto an available audio track on the camera for even faster and more accurate audio syncing in post.
On shoots where accurate free run timecode is needed, there is one master sync/time code generator like an Evertz 5600MSC Master SPG / Master Clock Combo which will provide properly synchronized timecode and sync signals for every device in both 24 and 30 frame formats. This type of system has been used extensively for multi camera sitcom shoots where cameras are fairly stationary and cabling to the timecode generator can be controlled easily.
For situations where there is a lot of camera movement and cabling between camera and sound is difficult, there are devices such as the Ambient Lockit Box which provides highly accurate timecode and sync to each device without wires. Once synchronized, the Lockits will stay in sync all day as long as they are not powered down. If a camera is powered down, you simply need to “jam” sync the timecode and sync reference again when powered up again.
Post Production Sound Mixing
Once in post production, audio for HD can fit seamlessly into either a film style or video style workflow. High definition programs offer a superb outlet for incredible 5.1 sound mixes in the broadcast world and for times when the HD footage will be transferred back to film. Often, audio is synchronized and put onto the downconverted field tapes (now converted to 29.97) and taken into editing, then transferred to the audio house as OMF files for final syncing, mixing, and sweetening to a downconvert of the assembled HD master. Always check sync with either the high definition master or a direct downconvert of that master, not the offline cut.
Also, when mixing for film print, it is standard practice to edit and mix in reels of about 20 minutes in length. When assembling completed audio reels together to be joined to a combined video reel make sure to take extra care that the join is done properly because if not, there can be a frame of offset created at every reel break which can add up over the course of the film. This can occur because of the different ways that tape based and disk based systems handle timecode outpoints. Tape based systems use “exclusive outpoints” where the “out” timecode number represents the timecode of the frame after the last frame of video (the point to be “out by”) and disk based systems often use “inclusive outpoints” where the “out” timecode number represents the timecode of the last frame of video to be included. When the two systems are placed contrary to each other, sync problems can result, so it’s important to know which system is being used to denote in and out points, especially when assembling reels for a feature.
Joining back to the HD Master
Once the audio mix is complete, it can be joined back to the master in a number of ways. Probably the most common is to lay back a stereo or matrixed surround mix to channels 1 and 2 of an HD master, and then encode a 5.1 mix (and maybe a music and effects (M&E) stereo mix) onto channels 3 and 4 using Dolby E. Dolby E is an audio encoding format for production that can encode up to 8 audio channels to fit onto 2 channels of a digital videotape machine. This master is then ready for HD distribution or can be converted for standard definition distribution as well.
High definition production and distribution offers many exciting opportunities for the world of audio to create sound that matches the incredible clarity of HD footage.