3DAS is built with Max/MSP/Jitter.


Version 1 (built by Dave Hunt and Nick Rothwell) was released in 2008 as 3DAudioScape by surroundAV,

a subsidiary of Paul Gillieron Acoustics (PGA).


It has been used extensively by Martyn Ware. They are still using version 1, and Martyn has promoted it effectively.


PGA are a busy acoustics company, and have found themselves unable to continue to support 3DAS. As the main developer and programmer Dave Hunt continues to develop and refine it.

Custom solutions, advice and consultancy  can be provided. Contact Dave Hunt.  Contact

It was originally designed to make Ambisonics usable, accessible and affordable. There are currently four versions, using the same front end and storage files to control several different spatial audio algorithms: Ambisonics, Ambisonic Equivalent Panning (AEP), Binaural, and Delta Stereophony (similar to Wave Field Synthesis).


It runs on Mac OSX, using Core Audio and Core Midi and any multi-channel audio interface. A Windows version is contemplated, but not yet available.


It allows audio sources to be positioned and moved in 3D space in real time. Their positions can also be controlled remotely by Midi or OSC messages. A virtual room, which  provides a global reverberation model, is provided. The dimensions and absorption of this room can be varied, and sources are constrained to remain within it.


All versions have 16 audio inputs, routing up to 16 mono or 8 stereo (or a combination of mono and stereo) source inputs to 3-dimensional panners. The ambisonic version includes a decoder to up to 16 loudspeakers, or a re-coding to a stereo binaural output. The AEP and Delta versions can also output to up to 16 loudspeakers. The binaural version produces a better binaural output for headphones or earbuds.

 Audio sources can be any input of the audio interface for live spatial audio processing, or the outputs of a DAW running on the same (or a different) computer for spatial audio “content creation”.


In the ambisonic version the results can be recorded and played back within 3DAS. Ambisonic encoding and decoding are entirely separate, so the “content” can be played on an entirely different speaker configuration by modifying the decoder settings. The binaural version can record the binaural output as a “stereo” audio file.


It is modular, with a 3D graphical display of sources and loudspeakers, and control panels for the spatial location of sources and other functions.


These provide a “scene description”, which can be applied to other spatial audio algorithms, or hybrids of these algorithms.



3D Visualisation