Audio technology encoding sounds as discrete 3D objects with spatial metadata rather than fixed speaker channels. Dolby Atmos supports 128 objects, automatically adapting to varying cinema configurations.
Technical Details
Dolby Atmos supports up to 128 simultaneous audio objects at a maximum sampling rate of 48 kHz/24 bit. DTS:X operates with an unlimited number of objects, but practically mostly uses 32-64 objects. The metadata is transmitted in BWF format (Broadcast Wave Format) or as a separate ADM file (Audio Definition Model) according to the ITU-R BS.2076 standard. Each object carries information about size, diffusivity, and priority. The spatial resolution is typically 0.1 degrees for azimuth and elevation. For cinema playback, the Cinema Processor (CP850 or RMU) converts the object positions to up to 64 loudspeakers.
History & Development
Dolby introduced the first object-based cinema sound in 2012 with "Brave." DTS followed in 2014 with DTS:X for "Crimson Peak" (2015). The breakthrough came in 2014 with Christopher Nolan's "Interstellar," mixed in Dolby Atmos with 128 simultaneous objects. In 2016, the ITU-R established the BS.2076 standard for object-based audio metadata. Netflix began Atmos streaming in 2017, followed by Amazon Prime in 2018. Since 2019, all major streaming platforms have supported object-based formats.
Practical Application in Film
"Mad Max: Fury Road" (2015) used 76 audio objects for the chase scenes, with each vehicle moving through the cinema space as a separate object. "Blade Runner 2049" (2017) employed object-based audio for raindrops, which were positioned as 32 individual objects above the audience. The mixing process takes place on consoles such as the Avid S6 or SSL System T with Atmos modules. Disadvantages: 30% longer mixing time, higher CPU load, and limited compatibility with older cinemas. Advantage: Automatic adaptation to different speaker configurations without remixing.
Comparison & Alternatives
Channel-based audio (5.1, 7.1) assigns sound to fixed speakers, while object-based audio remains flexible. Ambisonics (1st order = 4 channels, 3rd order = 16 channels) encodes complete sound fields, whereas object-based systems handle individual sources. Binaural Audio only works with headphones; object-based audio scales to any speaker arrangement. MPEG-H Audio (since 2020) combines channel-based, object-based, and scene-based approaches. For arthouse productions, 5.1 is usually sufficient; blockbusters rely on Dolby Atmos; TV productions increasingly opt for object-based home cinema formats.