A way of recording or building a full sphere of sound around a point, which you can then decode to whatever speaker setup you have.
One mic captures a 360 sound sphere as W/X/Y/Z, rotate the scene, then decode the same master to headphones, surround or a full dome.
What it is
Ambisonics records a full 360-degree sphere of sound as a format-independent blueprint you decode to any speaker rig or headphones.
Key facts
Speed of sound = 343 m/s in air at 20 C (rises ~0.6 m/s per +1 C)
B-format = the storage format: 4 channels W, X, Y, Z (W=omni pressure, X=front-back, Y=left-right, Z=up-down)
1st order Ambisonics (FOA) = 4 channels; localisation is blurry, sweet spot small
Channel count formula: ch = (N+1)^2, where N = the spherical-harmonic order. 2nd order=9 ch, 3rd=16 ch, 4th=25 ch
Higher order = sharper imaging + bigger sweet spot, but more channels and more speakers needed
Format-independent: ONE recording decodes to stereo, 5.1, 7.1.4, a dome array, or binaural headphones
Scene is rotatable in real time (yaw/pitch/roll) before decoding, ideal for VR and 360 video head-tracking
Ambisonic mics use 4 capsules in a tetrahedron, output A-format, converted to B-format
AmbiX (ACN order, SN3D norm) is the modern channel convention; FuMa is the older one. Match it end to end
+6 dB = double sound pressure; +10 dB = roughly twice as loud; -3 dB = half the power. Binaural decode uses HRTF to fold the sphere to 2 ch
How it works
A tetrahedral mic's 4 capsules grab raw A-format (4 directional signals).
Convert A-format to B-format: W, X, Y, Z spherical-harmonic channels.
Optionally rotate the whole sound scene (yaw/pitch/roll) to track head or camera.
Pick a decoder matched to your playback target (stereo, surround, dome, binaural).
Decoder maps the sphere onto your actual speaker angles or HRTF.
Same master plays back on any rig with no re-recording.
Real examples
VR/360 YouTube video: head-tracked Ambisonic audio rotates as the viewer looks around.
Live concert capture: one SoundField mic grabs the whole hall ambience for the broadcast bed.
Immersive theatre/dome show: decode HOA to a 30-speaker dome for full overhead sound.
Film atmos beds: record a forest sphere once, fold down to 5.1 and stereo for different deliverables.
Game audio: store ambience as B-format, decode in-engine to the player's exact speaker setup.
How it helps in live sound
For ambience capture, place one tetrahedral mic (e.g. Sennheiser Ambeo, Rode NT-SF1) at the mix sweet spot.
Record the 4-ch B-format flat, decide stereo vs surround fold-down later in post.
Use Ambisonic plugins (IEM Plug-in Suite is free, Sennheiser Ambeo) to rotate/decode in your DAW.
For an immersive PA/dome, count speakers first: 2nd-order HOA needs 9+ outputs, 3rd-order needs 16+.
Stick to AmbiX (ACN/SN3D) ordering end to end so mic, plugins and decoder all match.
Binaural decode the same file for a headphone monitor feed or VR preview, no re-record needed.
Everyday analogy
It's like photographing a room with a 360 camera instead of one fixed lens, you crop and aim the final view later to fit whatever screen you have.
Watch out
Myth: Ambisonics is just another surround format like 5.1. Truth: it stores a speaker-agnostic sound FIELD (W/X/Y/Z), and you decode it to any rig afterwards.
Fun fact
Ambisonics was invented in the 1970s by Michael Gerzon and the UK's NRDC, decades before VR existed, then sat largely unused until 360 video and headsets suddenly made it the perfect tool.
Key takeaways
Stores a 360 sound sphere, not fixed L/R channels.
B-format = W (omni) + X, Y, Z (the three axes).
Channels = (order+1)^2: more order = sharper image, more speakers.
Format-independent: decode once-captured audio to ANY playback rig.
Whole scene is rotatable, which is why VR/360 loves it.
One mic captures the full venue ambience in a single take.