Ambisonics

1970s: Ambisonics was developed in the 1970s as a surround sound technology by British engineer Michael Gerzon and later refined by Peter Fellgett and others.
Spherical Sound: Ambisonics aims to capture and reproduce audio in a spherical or three-dimensional manner, including both horizontal and vertical sound information.
B-Format: The core of Ambisonics is the B-Format, which represents audio as a combination of spherical harmonic components (W, X, Y, Z). This format allows for flexible manipulation of sound directionality.
Encoding and Decoding: Ambisonics involves encoding audio using B-Format during recording or synthesis and decoding it for playback through speaker arrays or virtual systems.
Immersion and Realism: Ambisonics has been used in various applications, including music production, gaming, and virtual reality, to create immersive and realistic audio experiences.
Challenges and Adoption: While Ambisonics offers a high degree of flexibility, its adoption has faced challenges due to the complexity of encoding and decoding, as well as the need for specialized equipment.
Modern Resurgence: In recent years, Ambisonics has seen a resurgence in interest, driven by advancements in VR and 360-degree video, as well as software-based solutions for encoding and decoding.
Open Standards: Efforts have been made to standardize Ambisonics, making it more accessible and compatible across different platforms and software.
Future Applications: Ambisonics continues to be explored for applications like 3D audio for headphones, spatial audio for augmented reality, and improved audio experiences in various multimedia content.
Ongoing Research: Research and development in Ambisonics continue to push the boundaries of spatial audio, aiming to provide more realistic and immersive soundscapes for listeners.

BBC R&D Ambisonics Intro

Ambisonics and Periphony

Principles

Gerzon, Barton, Fellgett - main theoretical basis
hierarchical approach to directional sound recording
formats
- A-format: microphone recording
- B-format: studio equipment and processing
- C-format: transmission
- D-format: decoding and reproduction
- UHJ - good for mono/stereo compatibility

Differences from stereo/quad

Quad
- is fixed to four speakers
- Uses phantom images between two speakers
- worse with off center listening
- common “hole in the middle” problem
Ambisonics
- encodes from all directions, decodes to your loudspeaker setup

Ambisonic microphone:
- Sennheiser AMBEO VR Microphone
- RØDE NT-SF1
- NEVATON VR
- Zoom VRH-8 Ambisonic Microphone Capsule for H8 Recorder
- Zoom H3-VR Handy Audio Recorder with Built-In Ambisonics Mic Array
- Soundfield SPS200
- Schoeps Double-MS
- Josephson Engineering C700S
Virtual source

Signal formats - A format

B format

B-format Ambisonics

Consists of four signals: W, X, Y, and Z
W (omnidirectional) represents the pressure component of the sound field
X, Y, and Z (figure-eight) represent the velocity components of the sound field in the forward, sideways, and upward directions, respectively
Similar to the sum and difference format of two-channel stereo, where the B-format is made up of three orthogonal figure-eight components and an omni component

Picking out the figure-eight and omni polar patterns

W: omni
X: forward-facing figure-eight
Y: sideways-facing figure-eight
Z: upward-facing figure-eight

Horizontal and vertical spatial information

W, X, and Y provide the horizontal spatial information
Z provides the vertical spatial information

Comparison to MS stereo

X: equivalent to M (mid) in MS stereo
Y: equivalent to S (side) in MS stereo

Recording options in the Zoom recorder

Can record directly to B-format with either FuMa or AmbiX encoding
Can record in A-format

Additional information

FuMa encoding is a more efficient way to store and transmit B-format audio signals than AmbiX encoding
Encoding is not necessary if recording in FuMa

I hope this is helpful!

Feature	A-format	B-format
channels	4	4
Signal rep	Raw output of ambisonic microphone capsules	Derived from A-format using mathematical transformation
Efficiency	Less efficient	More efficient
Ease?	More difficult to work with	Easier to work with

A-format and B-format are two ways of representing ambisonic audio signals.

A-format is the raw output of an ambisonic microphone. It is a set of four channels, each of which contains the signal from one of the microphone capsules.

B-format is a derived format that is more efficient and easier to work with than A-format. It is a set of four channels, but the signals in these channels are not the same as the signals in the A-format channels. The B-format channels are derived from the A-format channels using a mathematical transformation.

B-format is the preferred format for storing, transmitting, and processing ambisonic audio signals. It is also the format that is used by most ambisonic software and hardware.

Here is a table that summarizes the key differences between A-format and B-format ambisonics:

A to B Conversion

X = 0.5((LF – LB) + (RF – RB))
Y = 0.5((LF – RB) – (RF – LB))
Z = 0.5((LF – LB) + (RB – RF))
W = 0.5(LF + LB + RF + RB)

The A-to-B ambisonic conversion is the process of converting the raw output of an ambisonic microphone (A-format) into a more efficient and easier-to-work-with format (B-format).

The conversion is performed using a mathematical transformation that combines the four A-format signals into four B-format signals. The B-format signals are derived as follows:

W (omnidirectional): The sum of all four A-format signals
X (forward-facing figure-eight): The difference between the left and right A-format signals
Y (sideways-facing figure-eight): The difference between the front and back A-format signals
Z (upward-facing figure-eight): The difference between the top and bottom A-format signals

The A-to-B conversion is typically performed in software, but it can also be implemented in hardware.

Here are some of the benefits of using B-format over A-format:

B-format is more efficient to store and transmit, as it uses less data.
B-format is easier to work with, as it is based on a more intuitive representation of the sound field.
B-format is compatible with most ambisonic software and hardware.

C format - aka UHJ

D format - reproduction

Speaker numbers
- Four speakers give adequate surround sound
- six provide better immunity against the drawing of transient and sibilant signals towards a particular speaker
- eight may be used for full periphony with height.

D-format Ambisonics

A format for decoding ambisonic audio signals for a specific speaker configuration.
Not a separate ambisonics format, but rather a way of representing ambisonic audio in a way that is optimized for playback on a particular set of speakers.
Example: Using a D-format decoder to decode an ambisonic recording into seven channels for a 7.1 surround sound system.
Less commonly used than other ambisonics formats, such as B-format, but useful for applications where you need to decode ambisonic audio for a specific speaker configuration.
Used in some virtual reality and augmented reality headsets, as well as in some home theater systems.

Benefits:

Allows you to decode ambisonic audio for a specific speaker configuration, which can improve the immersive sound experience.
Relatively efficient, so it can be used to transmit and store ambisonic audio without using too much bandwidth or storage space.

Conclusion: D-format ambisonics is a practical and efficient format that can allow you to enjoy the full immersive sound experience of your ambisonic recording, especially if you need to decode it for a specific speaker configuration.

Higher order Ambisonics

Current Developments

Open Source Output format - Opus 1.3
Corporate interest
- Google - Spatial Audio RPC
- Youtube - 180 or 360 degree videos with spatial audio
- Occulus audio
- Microphones - Sennheiser Ambeo VR Mic, Zoom H3-VR
- BBC Research on ambisonics

Gaming

Unity, Unreal
Resonance Audio
Steam Audio
Battlefield 1 & Battlefield V - Dev Talks video