When preparing to originate programming in surround sound, there are some fundamental guidelines for loudspeaker selection and placement, equipment for monitoring loudness and mixing techniques.
For the local TV station, it's been a long tortuous transition from analog video and BTSC stereo to transmitting ATSC digital HD video and surround sound. HD video is far more revealing of details and imperfections. The same holds true for digital surround-sound audio — the 3in speaker is no longer the norm in the home TV set. According to the CEA, more than 20 percent of viewers are listening to the audio on high-quality surround systems, with this percentage growing each year.
In the analog world of BTSC, we monitored and mixed for the limited dynamic range required for the vast majority of viewers and for analog audio carrier's technical limitations of 25kHz deviation and 75µs pre-emphasis. BSTC digital audio offers a dynamic range of more than 100dB and with it the opportunity for creating great sounding mixes and the challenge of loudness deviations that exceed the comfort range of the viewers. Loudness complaints are on the rise; legislation is pending in Congress to mandate controls and penalties (H.R. 1084, The Commercial Advertisement Loudness Mitigation Act).
These opportunities and challenges have been under study by the ATSC. Last month, the ATSC released the “ATSC Recommended Practice: Techniques for Establishing and Maintaining Audio Loudness for Digital Television (A/85).” The document covers the ITU-R BS.1770 loudness measurement, target loudness and dynamic range management, metadata management, audio monitoring setup, and two essential “Quick References Guides,” one for station engineers and management, and the other for audio mixers and editors. An upcoming issue of Broadcast Engineering will feature a detailed report on A/85.
Speakers and speaker placement
The typical broadcast audio mixing environment is often far from ideal. There are, however, a few basics than can easily be followed to assure that what you hear is representative of what your viewing audience will hear on the typical high-quality system.
Speaker selection is critical. In a panel discussion at the Audio Engineering Society convention in New York in October, Sam Berkow, principal consultant for SIA Acoustics, offered valuable advice: “For effective monitoring, the LCR speakers should be identical, with the surrounds tonally similar to the LCR. We found that monitors that worked for stereo mixes are not always acceptable in surround. This is because the off-axis energy of the L and R speakers is absorbed by acoustical treatment, but for the C speaker, treatment is less effective because of the glass window. Finding monitors that are well behaved in their off-axis response can really affect the way that they are heard in a real working environment. We also found through testing that better off-axis response results in better subjective preference and less listener fatigue.”
The front LCR speakers should be positioned equidistant from the sound mixer's central position. (See Figure 1.) A time delay resulting from different path lengths produces a comb filter effect at the mixer's ear. If necessary, delay can be used to compensate for less-than-ideal placement. If the surround speakers are closer to the mixer than the front speakers, they too should be delayed.
The LFE subwoofer adds its own challenges. Berkow found that “rooms that worked well for standard stereo mixing may have all sorts of problems when you introduce subwoofers. The position of the subwoofer can be very important in how it interacts with the acoustics of the room and the LCR loudspeakers. Some basic acoustical measurement can be very helpful when positioning subs, and there really are very cost-effective ways to resolve low-end problems in many rooms.”
Speaker equalization is often used to correct less-than-ideal room conditions, but there's a fundamental downside to this approach. What the mixer hears is a combination of direct sound, early reflections off of adjacent surfaces (walls, furniture and the mixing console itself) and room reverberation. Speaker equalization tuned to correct room defects improves the room reverb response at the expense of coloring the direct sound. Most of the spatial cues come from the direct sound, and a non-flat direct response can result in a less-than-ideal mix. Room anomalies are much better addressed through acoustical treatment.
Equipment in the control room with fans raises the ambient background noise, often masking low-level detail, noise and hum. Wherever possible, locate such noisy equipment outside in an equipment room. The same holds true for HVAC; control flow rate and install duct treatment to reduce air noise.
Mixing fatigue is also a critical factor in the quality of the mix. Don't underestimate the value of a comfortable chair and fresh air.
Loudness monitoring and reference mixing levels
For decades, sound mixers used two tools: VU meters (or PPM) and their ears. Now we add the BS.1770 Loudness Meter. The VU meter and the PPM were designed to protect the electronics and the recording media; they are not a useful indication of subjective loudness. The ITU, in a joint committee, designed and extensively tested an algorithm to measure loudness in a manner that correlated closely to subjective hearing: “ITU-R Recommendation BS.1770 — Algorithms to measure audio programme loudness and true peak audio level.” The BS.1770 algorithms integrate audio from all channels except the LFE, and present the operator with a single number representing perceived loudness, on a decibel scale in units of LKFS. BS.1770 meters, in hardware and software versions, are available from several manufacturers. The ATSC adopted BS.1770 metering as a required tool in the sound mixing control room.
Jim Starzynski, principal engineer and audio architect for advanced engineering at NBC Universal, is the network's in-house specialist on audio technologies and practices, and chairman of the ATSC subcommittee that developed A/85.
He says, “Using the BS.1770 metering has been the major culture change in the transition to digital. Put the meter right in front of the sound mixer, typically on the meterbridge. (NBC's engineers prefer a meter that gives a simple easy-to-understand numerical indication.) Use the console's onboard metering solely to keep levels well below the peak clipping level.
“Mix with your hearing. Monitor with the BS.1770 loudness meter, using it as a tool. Look at the meter to set up your aural reference, and` then use your ears to determine the layering of the mix. While mixing, switch frequently to a two-channel downmix. If possible, listen through a stereo AC-3 emulator to simulate the metadata effect on the content.”
The ATSC A/85 Recommended Practice includes a “Quick Reference Guide for Audio Mixers and Editors Creating Content.” Distribute it to your mixing engineers.
Howard Mullinack is a technical editor for Linear Acoustic.