A system of sound reproduction based on the spherical harmonics of the sound field.  Although most easily described in terms of reconstruction a soundfield at a point, the design is actually aimed at reproducing directional cues in a practical but effective manner.


Eric Benjamin, Richard Lee and Aaron Heller, authors of a number of papers reporting on listening tests of various classic ambisonic playback rigs.  The papers published so far are available from

Classic Ambisonic Decoder

The best decoders for first-order ambisonics are still those designed by Michael Gerzon, which optimise the directional cues by maximising rV in the lower part of the frequency range and rE in the upper part.  Shelf filters are used to combine the two regimes.  Hardware decoders are generally classic in design, but many current software ones are not

Cooktown Recording & Ambisonic Productions

Richard Lee - designer of the SoundField Mk3a microphone among other things.


The process of generating suitable speaker feeds for a given rig from the ambisonic source signals is know as decoding.  Decoding can be seen as generating the outputs of microphones pointing in the directions of the speakers; if the speaker array is regular then the virtual patterns of the microphones are all the same: mildly hypercardioid for energy decoding (see rE), strongly hypercardioid for velocity decoding (see rV), and cardioid for "controlled opposites" decoding (which is deemed to be suitable for sound projection in large areas)

Grand Vizier (or GV)

Dave Malham of York

Greene-Lee neckbrace

A (notional?) clamp to hold the listener's head stationary.  This ensures that the head remains at the sweet spot (or as near it as it can fit) and that confusing head-movement cues are not introduced.  (Named for Richard Lee and Robert Greene)


Higher-Order Ambisonics - i.e. second-order and higher


International Telecommunication Union.  In sound reproduction this refers to the standardised layout of 5 speakers for home theatre usage (centre front, L & R at +/- 30 degrees, L & R surround at +/- 110 degrees or so).  This layout was never designed for true surround reproduction as is understood in ambisonics, and is extremely unsuitable for this usage because of the very irregular spacing of the speakers; and  domestically it is extremely rare that speakers are set up accurately to this layout anyway.  As a result a quite extraordinary amount of effort is put into generating useful decode coefficients for this layout...


Michael Gerzon - he who mainly started it all.

Malham Cap

A tight cap (like a swimming cap) placed over the head and ears, with hole over the ear canals.  The purpose is to remove pinna cues, and thus prevent them interfering with other better-understood directional cues. (Named for Dave Malham)


Mark 1 Human Head.  At present, in spite of its deficiencies, no alternative design is on the horizon


Single channel sound reproduction


Near-Field Compensation.  Basic ambisonic theory assumes plane waves, and thus is subject to proximity effect (as is familiar from directional microphones); the effect is present both in recording and playback.  At first order (POA) this is a comparitively minor problem, but the effect becomes successively stronger at each higher order, and therefore needs to be taken into account at some stage in practical HOA systems


Plain Old Ambisonics - i.e. first-order only


Plain Old Stereo

Quad or Quadraphonics

A range of surround systems developed in the 1970s, using four channels for horizontal surround, and (usually) matrixing them down to two channels in a comparitively ineffective way.  Two of the best known were SQ and QS


The Energy (gradient) vector.  Decodes maximising this are optimising for directional cues based on loudness differences, as used by the upper range of human hearing.  This kind of decode is used in the upper frequency part of a classic ambisonic decode


The Velocity vector. Decodes maximising this are optimising for directional cues using phase differences. As this mechanism is dominant in human hearing in the lower part of the frequency range, this kind of decode is used in the lower frequency part of a classic ambisonic decode

Shelf Filters

Classic ambisonic decoders have different coefficients in the upper and lower frequency ranges to suit different directional hearing mechanisms.  Phase-matched shelf filters are used to change from one regime to the other, typically at around 380 Hz, though for some purposes as high as 700Hz is used


Two channel sound reproduction; a system of sound recording and reproduction that works better than it has any right to


The matrixing scheme associated with Ambisonics.  The two-channel version was a compromise between the BBC's Matrix H and the NRDC's 45J.  The BBC's had been chosen from a range of possibilities by listening tests, and the NRDC's was based on theoretical principles


Wave Front Synthesis; an alternative method of reproducing directional recording playback.  WFS is currently only practical in a single plane, and requires lots of channels.  It is presumed that at a high enough number of channels (seriously many), the benefits of HOA and WFS will converge


Surround signals for home theatre (see ITU) consist of 5 directional signals and one non-directional additional bass signal (LFE) for special low-frequency effects such as earthquakes (commonly misinterpreted as being a subwoofer signal).  As the LFE signal is band-limited by definition, it is referred to as .1 of a signal, so the five speaker signals plus LFE add up to 5.1, the common designation for home theatre audio.  Some people attempt to overcome the deficiencies of this signal set in surround imaging (which, to be fair, it was never designed for) by multiplying its deficiencies into more channels: 7.1, 10.2...  Zillion.1 is Richard Lee's term for the long-term convergence of these doomed attempts
Ambisonic Info header