Audioseminar Banz

Summary of the lectures
of the 8th seminar, “Sound Reinforcement & Broadcasting”, in 2003:

Wolfgang Köhnsen, President of the German audio engineer’s association, VDT and chief sound engineer at Hamburg Staatsoper, was the first to speak. In his presentation, “Sound-Reinforcement Challenges in the Opera Business”, he discussed the evolution of public address systems in opera houses and present-day multimedia applications. Video footage and projection have become a significant part of opera house sound engineers’ and operators’ responsibilities. They are increasingly asked to create illusions even Rigoletto’s Count Almaviva would have mistaken for reality.

Next, Dr. Diemer de Vries of Delft University spoke about wave-field synthesis as the upcoming sound reinforcement approach. Unlike stereo systems, this sound field generation approach, based on speaker arrays, doesn’t just provide a single “sweet spot”, but rather an entire “sweet area” that can be experienced from various locations within a room or even theatre auditorium. This approach enables enhanced sound localization anywhere in a venue, provided one is prepared to purchase and use the kit required to achieve optimum results. The intense discussion after his presentation seemed to hint at the likelihood of wave-field synthesis being implemented in certain halls and theatres.

Dr. David Griesinger of Lexicon, flown in from the US for the occasion, surprised the audience with a simple test: he played back a video of an opera performance where the sound was high-quality and “up-front” at first, but becoming more diffuse as the video progressed. Griesinger used this to exemplify the differences between the ideal listening position and the rear seats in opera houses. He stated that the human brain analyses the acoustics of a given location only once and doesn’t even notice possible subsequent changes. This explains why nobody in the audience actually noticed the sound degradation of his video. Next, Griesinger showcased the acoustics of various famous opera houses, discussed their characteristics, and went on to demonstrate that achieving something the listener would call ‘perfect acoustics’ is actually next to impossible to achieve in real life. Therefore he advocates the use of electro-acoustic measures and techniques to enhance the acoustics for a given application.

Volker Löwer of IFB presented an altogether different case study: public address and acoustics in stadiums and arenas. Using the “Auf Schalke” arena (Gelsenkirchen, Germany) as an example, he discussed the challenges and requirements of multi-functional stadiums that need to provide a public address system suitable for sporting events, clear voice alarm messaging and even rock concerts. He said that the challenges posed by a roof that can be opened, a lawn whose sound absorption characteristics are as yet unknown, and which can be wheeled out of the stadium, as well as huge dimensions with lengthy sound delays, can only be addressed by using a decentralized sound reinforcement system that clearly covers defined areas.

Prof. Wolfgang Fritz of Wiener Staatsoper is also faced with huge sets and dimensions for the Bregenz festival and needs to adapt his PA system to suit. “We are still waiting for invisible speakers that can be set up exactly where they need to be,” was just one of his poignant remarks. He stated that, “Clearly focused directional sound reinforcement is only possible on open-air sites if one accepts flying speakers from the sky, drowning them in the water, or positioning them in places where they can be seen.” But even if this is accepted, people sitting at the back would still have problems with locating sounds, as the first wave front would still be too soft. In his opinion, the festival site is so huge that even wave-field synthesis would not be able to solve the problem. The ensuing discussion showed that several delegates disagreed with that idea…

Wireless microphones have become staple tools for PA and broadcast applications. That’s why Peter Arasin of Sennheiser had accepted the invitation to talk about the future of wireless microphones during both sections of this year’s seminar. He said that the introduction of digital wireless microphones would mainly involve more user-friendly and smarter devices capable of scanning the frequency bands automatically. There will also be the possibility of remote control via a PC, and so on. However, the downside is that digital wireless mikes require high precision manufacturing techniques and greater bandwidth. Therefore he does not expect analog approaches to become obsolete in the near future, and anticipates that they will be used alongside digital systems.

Albrecht Krieger of Staatsoper Unter den Linden (Berlin) was another speaker who appeared in both sections of the seminar, albeit with different approaches. He chose to discuss open-air events involving classical music, based on countless examples of the “Berliner Waldbühne” venue. During the first section, he concentrated on the ways and means he has been using for open-air sound reinforcement. Here again, the localization aspect of sound sources plays an important part. By their very nature, open-air events are subject to weather conditions (wind and rain), which can prove quite a challenge during concerts and opera performances. During the second section, Krieger discussed the challenges he faces in preparing signals for public address on the one hand, and live telecasts on the other. Since each domain requires a radically different approach, one cannot simply feed the FOH mix to the broadcaster.

In his presentation, “New Surround Sound Methodologies”, Dr. Günther Theile of the Institute for Broadcast Technology IRT discussed binaural room synthesis for multi-channel audio applications. Binaural sound reproduction can be used for playing back 5.1- and 7.1-channel program material using headphones. Based on response data of a reference room, a BRS processor employing convolution can calculate the sound image that should be perceived at a given spot in the virtual reference room. That sound image is then fed to a pair of headphones. To achieve maximum realism and allow for optimum perception of front/rear information, the movements of the listener’s head are tracked and processed by the BRS system. This technique could be used for simulating near-perfect listening conditions in environments one usually does not associate with these requirements, such as OB-van control rooms. Another application would be surround sound imaging in headphones for the consumer market.

Multi-channel sound has become almost commonplace for TV stations. Martin Wöhr of Bayerischer Rundfunk and Jürgen Marchlewitz of Westdeutscher Rundfunk showed how multi-channel sound can be ported to broadcast applications. At the IFA 2003 fair in Berlin, they launched a test program using AC-3 encoding that was transmitted via DVB-S at little additional cost. The main reason for choosing the AC-3 format was that most consumer receivers support it. The response from the audience was extremely positive, while audio pros immediately signaled their readiness to pursue this project. Despite several technical challenges, which were mainly due to the compatibility problems of a vast number of receivers, BR and WDR have decided to carry on with their multi-channel broadcasts. BR will use it for simulcast applications, while WDR has decided to create special programs for that service. In the evening, Wöhr and Marchlewitz took advantage of the top-notch surround system provided by MEG for demoing their test programs.

Going one step beyond purely audio-related subjects, Frank Fell-Bosenbeck of the German TV station ZDF discussed the future of HDTV. Thanks to the rise of digital transmission channels, HDTV can now be aired at acceptable cost and looks set to become a reality within the foreseeable future. Fell-Bosenbeck also reminded the audience of the necessity to proceed in that direction, because the growing popularity of plasma screens will soon make MPEG-induced picture deterioration all the more apparent. Apart from the equipment for airing and reproducing HDTV images, one also needs mass-storage media for the production side. Several approaches are currently being used: Blue-ray, OPD as well as solid-state systems, such as SD Memory Cards. Both approaches seem to cater to the call for a closed production chain capable of avoiding unnecessary data compression cascades.

Roland Fischer of the Swiss TV station TSR, explained how the opening of Expo.02 was covered and presented some of the present-day production techniques now possible. The opening celebration had been scheduled to take place in four different locations and was aired by four broadcasting stations using different viewing angles and sound. The main concern for this production was to provide proper synchronisation of the video and audio data and to restrict, if not eliminate, latency and related delays between the four venues. The musical sections of those shows, involving four symphony orchestras and bands performing simultaneously, proved the biggest challenge. Such large-scale applications are only possible in the digital domain, using fibre-optic links between the various locations.

Florian Camerer of the Austrian broadcaster ORF tackled another practical example. ORF was the first station to air the traditional 2003 New Year Concert of the Wiener Musikverein in 5.0 Surround sound. The success of that endeavor was such that it was decided to continue to produce live telecasts with multi-channel sound. A live event, the New Year Concert proved a formidable challenge for the production crew. Four control rooms were used for the parallel preparation of regular 2.0 TV sound, 5.0 Surround sound, 2.0 radio sound and 2.0 audio for the CD recording. ORF has since taken to producing all major telecasts, like the Salzburg festival, an Antarctica documentary, and important sporting events, in the multi-channel sound format. A 2.0 stereo-sound equivalent is generated based on automated down-mix routines. At the end of his presentation, Camerer played several examples of the New Year Concert and the Salzburg festival.

Loudness and level are altogether different things, yet TV stations seem to be unaware of this truism. Whether on a single channel or between channels, the loudness of TV sound would greatly benefit if a standard could be established. This is the subject Jean-Paul Moerman of the Flemish broadcaster VRT and Ralph Kessler of Ingenieurbüro Pinguin had been asked to elaborate on. They demoed a number of striking examples and discussed the possibilities and approaches to measuring and thus eliminating inconsistent loudness. One of their findings came as a surprise: public broadcasters appear to have bigger problems with loudness jumps than private stations, because loudness differences do not only exist between regular programs and commercials, but also among various programs.

The last presentation of this year’s seminar was given by Gerhard Stoll who discussed various audio formats, such as MP3 Pro and DTS. Audio codecs are still evolving at a rapid pace, while a universal codec capable of handling all tasks at the same high level, has yet to materialize. For example, while the Quicktime Music codec is just perfect for music, its quality for speech applications is almost dismal. Comparisons have shown that data rates are no longer reliable indicators of sound quality, because a multi-channel AAC signal streamed at 160 kbs exhibits the same quality as Dolby and DTS at 384 kbs and 448 kbs, respectively. Details of these comparisons and example files are available on IRT’s web page.