Meeting Topic: Field Trip to Fox: Immersive 3D Sound - The Auro-3D Sound Format
Moderator Name: Eric Johnson
Speaker Name: Sven Mevisson, Barco Product Manager
Meeting Location: Fox Studios, 10201 W Pico Blvd., Los Angeles, 90064
October 2013 brought the Los Angeles Section to the grounds of the Fox Studios in West LA. Packing the Howard
Hawks dub stage, attendees had the chance to experience Auro-3D, Barco's next generation sound format, up close.
Powered by the Belgium based company in tandem with Galaxy Studios, the list of recent films featuring Auro mixes
includes Elysium, LucasFilm's Red Tails, the Indian spy thriller Vishwaroopam, and the recently released Ender's Game.
Launched by Auro Technologies as a 9.1 format at the 2006 AES Convention in Paris, the technology was licensed for
cinema use exclusively to Barco in September of 2010. The first plugin version of the format's authoring tools appeared
in October of that year, and in January of 2012 Red Tails became the first feature film released with an Auro-3D 11.1
mix.
Barco Product Manager Sven Mevisson opened his presentation with the key open questions motivating the format's
past and current development: How can a technology create the most immersive sound possible? How can a three
dimensional sound environment be reproduced? How can such a space be simplified and parameterized? Moreover,
given a modeled 3D space, what is necessary to ensure the compatibility and robustness of that representation across
not only variable playback configurations, but also markets as disparate as gaming, mobile and automotive?
Before describing the ways in which Auro-3D addresses those issues, Mevisson established a couple of additional
principles and objectives guiding the format's design. While striving for the highest fidelity and spatialization quality
possible, the partner companies also focused on practicality. Minimization not only of the number of speakers and
amplifiers required for reproduction, but also impact on work flow, both at the scale of distribution (i.e. DCI/SMPTE
compliance) as well as mixing, were prioritized. Though 9.1 was indeed a starting point and seemed viable as the smallest
system capable of producing 3D, 11.1 has become the format's configuration of choice.
At that point Mevisson introduced the three layer structure at the core of the format. Auro-3D partitions three dimensional
space into lower, height and top layers. The lower layer (1) encompasses sounds from the "horizontal
plane", at the listener's ear level, while the height layer (2) "the most immersive layer" includes sound sources and
reflections originating forty degrees and higher above ear level. Finally the top layer (3), commonly known as the voice
of god (VOG) layer covers the space directly overhead. In terms of channels, layer 1 consists of a conventional 5.1 / 7.1
system (L C R Ls Rs LFE). Layer 2 then offers essentially another full surround layer, with HL, HC, HR, HLS and HRS. The
eleventh channel (T) hangs in layer 3, and can be spread across multiple speakers to produce a broader sweet spot.
In the interest of ultimately bringing the Auro-3D experience to consumer level playback systems via established lines
of distribution, Auro created the Auro-Codec. The codec accepts an input 11.1 or 13.1 (added back channels) mix and
produces a 5.1 or 7.1 PCM for incorporation into DCP, Blu-Ray and HDMI compatible Auro print masters. Broadly
speaking, the height layer is folded down into the lower layer, and the top channel is mixed into the corner channels.
Theaters have the choice then of either playing back the DCP directly, or using an Auro-3D decoder to retrieve the full
mix. Mevisson reported 100% channel separation, with coding artifacts under -90 dB.
Pausing from the technical presentation momentarily, Mevisson presented a collection of demonstrative film
and sound clips. An ambient recording of a busy Amsterdam square illustrated the relative contributions of Auro-3D's
layers, starting with the full mix and then isolating each of the three. Pipe organ and symphony performance clips
showcased the format's handling of a single, live acoustic space, and an airplane flyover emphasized the utility of the
top layer.
Several film trailers brought the presentation to the subject of work flow in the studio, and here mixers Chris
Jacobson and Marti Humphrey joined Mevisson. Again, the options come in three. Engineers have the choice of
working either completely inside the box, on a console with integrated controls (i.e. panners), or somewhere in the
middle, with controls spread across software and hardware. Incorporating SRS's Multi-Dimensional Audio (MDA), the
plugin is capable of processing position information for up to 128 individual audio objects. Over 30 individual speakers
can be addressed at mix time, and Auro's tools allow for audio clips to be treated either as channels or objects within
the same mix via standard from automation. The implication here is that a mixer might use standard panning moves as
a starting point and hone them further in Auro's three dimensional space.
When asked to describe a few of the format and tools' most compelling features, aside from a streamlined
workflow in comparison to other available 3D audio mixing tools, Jacobson highlighted the height layer. While the
format is geared to make use of speakers spread throughout a room, in practice most of the important content finds
its way up to those channels around and behind the screen. He said the presence of two complete layers to work with
in that front-of- audience space helps to mediate the challenge of managing those typically crowded front channels.
For example, scoring or ambience can be positioned around dialogue, instead of the two fighting for bandwidth and
headroom in a single position. Additionally, the relative transparency of MDA decreases the likelihood that a 3rd
party's technology would be necessary in order for mixes to be handed off during production and distribution.
The presentation concluded with a brief overview of some recent developments concerning the larger push for
an open standard for immersive audio currently underway. The National Association of Theater Owners, DCI and
SMPTE have jointly requested an interoperable format. Along with Barco and DTS (which now owns SRS), Auro is a
member of a 12 company group working on an open standard, object based format. The group intends to make the
standard royalty free. Both Mevisson and Barco Senior Strategic Business Development Director Brian Claypool emphasized
the potential utility of MDA for its open standard nature, meaning third parties would not need to be involved
in the delivery or distribution process.
Written By: Nathan Dahlin
