Ambisonics

Ambisonics is a full-sphere surround sound technique that represents a 3D sound field mathematically using spherical harmonics. Unlike traditional channel-based audio (stereo, 5.1, 7.1), Ambisonics separates the sound field capture from the reproduction, allowing the same audio content to be decoded for headphones, stereo speakers, or arbitrary multi-speaker arrays.

Why Ambisonics?¶

Traditional audio panning places a sound at a specific speaker location. This works well for one listener position but breaks down when:

• The listener moves or turns their head.
• Multiple listeners need different perspectives.
• The speaker layout is unknown or non-standard.

Ambisonics solves these problems by encoding the entire sound field. The decoder then reconstructs the appropriate speaker signals for any given listener orientation and speaker layout.

Spherical Harmonics¶

Ambisonics represents the sound field around a point using spherical harmonics — mathematical functions defined on the surface of a sphere. Each "order" of Ambisonics adds more angular resolution:

Higher orders capture more directional detail but require more channels and processing power.

B-Format¶

The encoded Ambisonic signal is called B-Format. Amplitude uses the modern AmbiX convention:

• ACN channel ordering (SN3D normalization)
• SN3D (Schmidt Semi-Normalization) for energy consistency

The W channel is omnidirectional. The X, Y, Z channels encode front-back, left-right, and up-down intensity respectively.

The Amplitude Ambisonic Pipeline¶

Amplitude uses Ambisonics as an internal mixing format. Here is how audio flows through the Ambisonic pipeline:

graph LR
    A[Mono Sound Source] --> B[AmbisonicPanningNode]
    B --> C[AmbisonicRotatorNode]
    C --> D[AmbisonicMixerNode]
    D --> E{Output Target}
    E -->|Headphones| F[AmbisonicBinauralDecoderNode]
    E -->|Speakers| G[AmbisonicDecoderNode]
    F --> H[Stereo Output]
    G --> I[Multi-channel Output]

Encoding (AmbisonicPanning)¶

The AmbisonicPanningNode takes a mono sound source and encodes it into B-Format channels based on its 3D position relative to the listener:

B-Format channels = mono_sample × spherical_harmonic(direction)

Each source contributes to the global sound field independently.

Rotation (AmbisonicRotator)¶

When the listener turns their head, the entire sound field must rotate in the opposite direction to keep sounds stable in world space. The AmbisonicRotatorNode applies a spherical harmonic rotation matrix:

rotated_BFormat = rotation_matrix(yaw, pitch, roll) × BFormat

This is far more efficient than rotating each sound source individually, especially with many active voices.

Mixing (AmbisonicMixer)¶

The AmbisonicMixerNode simply sums all B-Format streams channel-by-channel. Because Ambisonics is linear, summing sound fields is mathematically correct — there is no phase cancellation or comb filtering between mixed sources.

Decoding¶

Binaural Decoding¶

For headphone output, the AmbisonicBinauralDecoderNode decodes the Ambisonic field to binaural stereo by convolving each spherical harmonic with pre-computed HRIR data:

left_ear = sum_over_orders(BFormat_channel × HRTF_left)
right_ear = sum_over_orders(BFormat_channel × HRTF_right)

This produces a spatially accurate headphone experience.

Speaker Decoding¶

For speaker arrays, the AmbisonicDecoderNode uses a speaker-specific decoding matrix derived from the speaker positions. Amplitude includes presets for the standard layouts exposed by ePlaybackOutputChannels (mono, stereo, quad, 5.1, 7.1).

Energy Compensation¶

Higher-order Ambisonic channels contain less energy than lower-order ones, and decoding to a finite virtual loudspeaker array can flatten the spectrum further. To keep loudness perception consistent across orders, Amplitude applies energy compensation inside AmbisonicBinauralDecoder rather than as a separate pipeline node:

• Max-rE weighting optimizes the horizontal plane.
• In-place gain compensation keeps overall loudness perceptually constant when switching between BinauralLowQuality, BinauralMediumQuality, and BinauralHighQuality panning modes.

Advantages in Games¶

Ambisonics is particularly well-suited to game audio because:

1. Listener rotation is cheap: Rotate the field once per frame, not once per source.
2. Multi-listener support: Decode the same field differently for each listener.
3. Ambisonic beds: Pre-recorded Ambisonic ambience (e.g., rain, crowd) mixes naturally with dynamic sources.
4. Future-proof: The same content works for stereo, surround, and whatever format comes next.

Limitations¶

• Sweet spot: Speaker decoding has a limited sweet spot where spatialization is accurate.
• CPU cost: Higher orders are expensive to encode, rotate, and decode.
• Content availability: Most sound libraries are mono or stereo, not Ambisonic.

Next Steps¶

• Follow the Ambisonic Setup Guide to configure your project.
• Review the Pipeline Reference for node configuration.
• Learn about HRTF and Binaural Audio.