The differences between the DTS MDA object-based sound format and the similarly capable Dolby Atmos are primarily that of company strategy. In this, DTS appears to be moving ahead of Dolby.
Object-based sound poses a major challenge to the content-sphere. Channel-based sound is dead simple, assigning a recorded channel of sound in its entirety to a loudspeaker. The sound mixer pans a sound object in the mixing room, and the outcome is recorded in the final sound track, which is then available to content owners. In sharp contrast, with object-based sound, the sound mixer pans a sound object, and no final sound track is created. Instead, the object is stored and distributed, along with metadata that says how the sound mixer intended it to be panned. The rationale for adding this complexity is that the end user’s speaker array may not be identical to that used by the sound mixer. So the art of rendering object-based sound is to best match the intent of the sound mixer with the available speakers in the end user system.
There are two distinctly different techniques for putting object-based sound to work, which are not mutually exclusive. In the first technique, it is recognized that a large number of speakers are required to move sound spatially about a room. If channel-based sound were used for this room, an equally large number of channels would be needed, one channel per speaker. When object-based sound is used for the room, the actual amount of sound data needed is far less, because only sound objects are required, with accompanying metadata. So effectively, a large number of speakers can be addressed without carrying a large amount of sound data in distribution. Even if the number and placement of speakers used by the end user are identical to that used by the sound mixer, with the intent to carry the soundfield heard by the sound mixer verbatim to the end user, the object-based method adds value by reducing the size of the distribution.
In the second technique, the number and placement of speakers used by the end user are not necessarily that used by the sound mixer. When dissimilar speaker systems are used, the object-based mix can be rendered to produce the best possible approximation of the soundfield heard by the sound mixer. So whether the end user can afford many speakers, or just a few, a reasonable version of the mix is heard, albeit not that which the sound mixer heard. This is the DTS MDA approach.
Philosophically, the Dolby Atmos approach is designed to preserve director intent. The DTS MDA approach can also preserve director intent if speaker systems match, but it is marketed to emphasize “renderer” intent. The DTS MDA approach allows exhibitors to install what they can afford, while Dolby Atmos is more black and white: install the full Atmos array, or an approved subset thereof, or stick with 5.1 or 7.1. Both approaches can be served by the same object-based architecture, which is one of the blind spots in the Atmos/MDA/Auro 3D debate. The difference in all three of these systems has less to do with technology and more to do with company philosophy.
Company strategy, however, is where the game is getting hot. Dolby’s DNA requires it to own its technology, which led to such goofs as monopolizing the Atmos sound mix for a movie, much to the pain of the studio distribution arms that own the rights. Dolby has yet to take visible steps to fix this mess. However, its active engagement in the SMPTE 25CSS standards process may indicate that it’s simply waiting to see what evolves on the standards front in the way of assistance.
DTS is taking a very different approach. Object-based sound is inherently complex and data-driven. It is not dead simple channel-based sound. Without an open format for object-based sound, the only path available for object-based sound is that which Dolby took with Atmos, which rights owners are not pleased with. In sharp contrast to Dolby, the path DTS is taking is to give MDA away, with the understanding that an object-based sound format is more valuable as an open format than a closed one. This would solve the rights owner issue, and insure that DTS has a fair opportunity to compete in providing consumers with object-based sound.
DTS acquired MDA through its US$125M acquisition of SRS in 2012. It has long hinted that it will give MDA away, but has not backed its intent in writing, up until now. In a written statement to the FCC this month, DTS requested that the US carry its MDA specification to the ITU-R for international standardization, declaring that MDA would be made available on a “open and royalty free” basis. (ITU-R is organized under the United Nations as the International Telecommunications Union.) This appears to be the first time DTS has made such a declaration in writing to a formal body, making it quite significant. Other than SMPTE and its proposal for ITU, DTS says it has also submitted MDA for standardization to ETSI (the European Telecommunications Standards Institute).
There is some irony in DTS’ action to give away MDA object-based technology to the ITU-R. Dolby’s CTO, Craig Todd, is a vice-chair of the ITU-R committee that will receive the US request for standardization of MDA.