( )

~ O ~ http://www.kineticsoundsystem.com

( )

Thanks a lot.

_____________________________

Patrick Delges

Centre de Recherches et de Formation Musicales de Wallonie asbl

]]>To really get it you want to read the article in Computer Music

Journal (Myatt A and Malham DG: Three Dimensional Sound

Spatialisation Using Ambisonic Techniques, Computer Music Journal,

Vol.19 No.4, MIT Press, Cambridge MA,USA, 1995, pp58-70.

Also vist:www.ambisonic.net/

and: en.wikipedia.org/wiki/Ambisonics

Briefly, the point of ambisonics is that it is based on the

mathematical technique of sampling the surface of a sphere, whis can

be represented by four signals: X = front/back, Y = left/right, Z =

up/down and W = omni – for largely historical reasons to do with the

soundfield microphone this is known as B-format. A few simple

formulae involving sines and cosines (very easy to implement in Max

using cycle~ objects) enable the four signals to be decoded for any

number and configuration of speakers.

This technique was developed by the mathematician and sound expert,

Michael Gerzon in the 1970s, which led to the development of the

Soundfield Microphone, which enables a soundfield to be recorded and

encoded into B-format (A-format is the output of the microphone itself).

Level panning only works properly for speaker angles of 60 degrees or

less (and is also very dependent on correct location of the

listener), so if there are four speakers the average angle is 90

degrees, and level panning will always be sub-optimal. However,

ambisonic signals can decode correctly over four speakers, and they

don’t even have to be in an exact square. Larger numbers of speakers

will give greater accuracy for a larger number of people, as well as

enabling three dimensional positioning. Moreover, the same B-format

recording can be decoded for any number and positions of speakers, so

is easily adaptable to different listening conditions.

Much more details on the ambisonics website and the Wiki article (see

above)

On 22 Apr 2007, at 06:43, Joseph Grimm wrote:

> would anyone be willing to explain how ambisonics works in layman’s

> terms? How would a four-channel ambisonic setup differ from quad

> panning based on simple level differences among four channels?

Hope that helps

Lawrence

Lawrence Casserley – lawrence@lcasserley.co.uk

Lawrence Electronic Operations – http://www.lcasserley.co.uk

Colourscape Music Festivals – http://www.colourscape.org.uk