On Mon, 21 May 2007 20:14:05 +0100, "Serge Auckland"
wrote:

"Dave Plowman (News)" wrote in message
...
In article ,
Serge Auckland wrote:
We will need a completely new way of generating sounds at home, not
using discrete loudspeakers, before we can realistically recreate a
complete soundfield.

Ambisonics come close using conventional speakers. At a cost.

However, the improvements in sound reproduction have bored some, who
have returned to very obsolete technology in an attempt to stir up some
excitement. So, plenty to exercise us.

There could well be something in that. When I started playing around with
audio, the norm in a domestic environment was pretty poor. FM radio was
near non existent, and TVs had poor loudspeakers. Record players had small
loudspeakers driven off SET valve amps ;-) with pretty nasty crystal
pickups. So it was relatively easy to bring about a real improvement which
gave one a great deal of pleasure. Not so easy these days given even a
modest but decent sound system straight out of the box.

--
*I am a nobody, and nobody is perfect; therefore I am perfect*

Dave Plowman London SW
To e-mail, change noise into sound.

--

Ambisonics came closer than quadrophonics did, but still relied on a small
number of loudspeakers, and the creation of phantom images between the
loudspeakers to fill in the gaps between them. A real soundfield is all
around the listener, with sounds coming from an infinite number of
directions.

At the European AES a few years ago, I heard an experimental sound-field
creator which used something like 200 small loudspeakers arranged round the
periphery of a room, each loudspeaker driven from a separate power amp, and
each loudspeaker being given an individual feed off a large DSP driven
routing matrix (the same matrix that's currently in BH as the main and
programme routers and at Bush House). The demo didn't have any height
information, but the surround soundfield was most impressively realistic.
One demo was of a city street recorded with the Soundfield microphone and
then synthesised in the room. It was the most realistic portrayal I've yet
heard. Walking round the room gave the same sort of effect as walking around
in the open air in a city. To be complete it would have needed several
hundred more sound sources to portray height, but the principle was sound.

Clearly this sort of system wouldn't be domestically acceptable, but I think
we have to get away from the current paradigm of 2 or 4/5 or even 6/7
loudspeakers and to some sort of sound-field synthesiser if recorded music
is to make real progress. Can't see it happening though, as the concept of
listening to music in one place, without distraction seems as old-hat as
wearing spats.

S.

As I understand it Ambisonics is a system to encode directional
information, 3 channels can encode from any direction in a plane
accurately and 4 channels can encode sound coming from any direction
on a sphere accurately.

The reproduction decoder is a seperate function from the recording and
it has always been recognised that the more speakers the better, but
practicallity dictates that commercial decoders are designed to drive
a small number of speakers. However, ISTR that the WW/Integrex decoder
of about 1978 was capable of driving 6 speakers and the current
Meridian designs can drive 7 speakers. There are other non-commercial
decoder designs that can drive many more speakers.

Improving image accuracy with multiple drivers has been tried even
with normal 2 channel systems. E J Jordan described a system using
many drive units arranged in a horizontal line. The speakers were
linked with delay lines, the left signal went in one end, the right in
the other. He claimed that the result gave very accurate imaging
independent of listening position. If it worked as he claimed it
should have been a significant improvement over normal 2 speaker
stereo.

Bill