L-R: Lon Neumann, Jim Starzynski, Tim Carroll, Ken Hunold, Jackson Wiegman
Members and guests of four technical societies packed NBC Universal's 30 Rockefeller Plaza mezzanine conference room to hear a distinguished panel present a review of the CALM Act (Commercial Advertisement Loudness Mitigation Act—passed by the U.S. Congress by unanimous vote) and the progress made in implementing it so far.
Jim Starzynski, who was the lead presenter, stated in his introduction that he felt the CALM Act is "a win-win for everybody", with improvement in the consistency of broadcast sound levels already evident. Starzynski made clear that he felt this had come about because the Congress and the FCC had wisely left the actual writing of the technical standards and rules to the Advanced Television Systems Committee (ATSC), which was composed of highly qualified industry engineers who were determined to "get this right". The result of their labors was ATSC A/85, "ATSC Recommended Practice: Techniques for Establishing and Maintaining Audio Loudness for Digital Television", issued in final form in July, 2011. (A PDF file of the paper can be found online easily by Googling "ATSC A/85".)
Starzynski noted that the ATSC originally intended the annex sections of A/85 (annexes A through K) to be the suggested rules for implementation in the law, but the FCC, after reviewing the document, felt that the body of the paper contained much that should be included as well. ATSC A/85 is now part of the CALM Act, and successor documents are possible as technology evolves. (I recommend this paper to the attention of anyone interested in the issue—it provides a comprehensive overview of the issues involved in establishing and maintaining audio loudness in broadcasting, while being concise and on point.)
The other presenters approached the subject from various angles:
Lon Neumann, Senior Technical Consultant for THX, spoke about what the new rules required from operators and how they would be enforced. The rules apply to all commercials, and compliance can be obtained via two paths:
- Local (using processing equipment in real time) and,
- Safe Harbor (content that is mixed to the standard is passed through without modification)
Neumann went on to detail the initial compliance regime, which consists of spot checks over a period of 2 years, and how enforcement works. With the latter, complaints originate with the FCC and spot checks are made within 30 days. If non-compliance is found the FCC is notified within 7 days, and a second spot check is made again within 30 days. If the operator is still in violation, it loses Safe Harbor status and becomes liable for fines. In short, this law does have teeth. Fortunately, making a complaint is not a trivial matter, and requires filling out several forms with specific documentation of the complaint.
Jim Starzynski covered the various techniques involved in producing CALM-compliant audio. He noted that it is not sufficient to address just the loudness of commercials; a "commercial can't be loud unless the program is soft." Therefore, it is necessary to address the loudness of programs as well. He went over the means used for measuring and adjusting the loudness of long-form programs, commercials, public service announcements and interstitial material (station I.D., etc.). (NBC Universal's technical specs for delivery of short form material—that is, commercials--can be found at http://nbcuni.force.com/commops)
Ken Hunold of Dolby Labs (and current Chairman of the AES New York section) talked about preserving dynamic range while preserving loudness. He described the Dynamic Range Control metadata-based system that is part of the AC-3 encode-decode chain used in Digital TV. When used properly, it allows the listener to have some control of the dynamic range of their programming, typically by changing the settings in their set-top box or other equipment. Some operators have been slow to implement DRC and instead still rely on processing that was typical of analog television systems. Hunold played two clips of a baseball game broadcast, one of which used the DRC system as intended, while in the second, the exact same part of the game had been processed by a local station in the analog manner. The difference between the two clips was very audible, with the analog processing of the latter being glaringly obvious.
Jackson Wiegman talked about the ITU BS.1770 content measurement spec, and also described measurements he had made of channel to channel loudness of several different broadcast channels. In the worst case there was a 10 dB difference between the two worst channels. It is hoped that implementation of the CALM Act will narrow such differences over time, as operators monitor and adjust their equipment for compliance with the law.
Tim Carroll spoke about the potential for operators to comply with the law by using heavy-handed audio processing, titling his segment, "Who cares about quality? I'm Compliant!" He stated that "consumers don't complain about audio processing, they complain about inconsistency" (of levels). Carroll also covered how to implement DRC via the use of metadata.
The following day, inspired by the various presentations, I calibrated my monitoring system to ATSC standards, using the files linked to in the A/85 paper. I then called up some Pro Tools mixes of jazz big band with vocal material that I had been working on and played them without modification though the calibrated monitors. I should mention that my approach in these mixes was pretty close to an "audiophile" standard, with very little use of compression on individual tracks (mostly on the vocals) and no use at all of bus compression. I was very pleased with the results—it is clear that the ATSC standard allows for unmodified playback of high quality mixes with considerable dynamic range. Now if we could just get the music industry to adapt similar standards for compact discs and related media! (One can dream...)
