Audio Developments AD149 - Manual (Page 18)

17
differences. Unlike a dummy-head, this microphone has a flat frontal frequency
response and lacks those cues which give front/rear information in headphones
listening and which only lead to colouration in loudspeaker listening.
None of these near-coincident microphone techniques is strictly mono compatible.
[Although the use of omnidirectional microphones is rightly preferred, wherever
possible, by many engineers, it is worth recalling a few of their frequently overlooked
characteristics. They possess an excellent low-frequency response and, although
fairly immune to wind and blasting effects, they are particularly adept at responding to
traffic, air-conditioning plant and structure-borne vibrations. They provide a good
sense of space and 'openness', but, for a given direct/reverberant sound ratio, they
require the closest placement to the sound source. They have the maximum
potential for pickup of unwanted sounds - although with minimum colouration.]
[When deciding upon an omnidirectional microphone, there is a choice to be made
between the type which is designed to have a flat response to an on-axis signal (direct
or free field) and the one designed to have a flat response to a reverberant sound
field (random incidence or diffuse field), ie with an HF tip-up in the on-axis frequency
response. The MKH20 contains treble accentuation circuitry permitting usage in both
direct and diffuse sound fields and also gives the microphone a directional effect at
high frequencies.]
THE MATRIX PROCESS
Conventional left-right stereo signals may be produced directly by using X-Y
techniques, or indirectly by using M-S techniques. In the latter case, the signals have
to undergo a further process variously called matrixing, encoding or decoding. The
matrix amplifier adds and subtracts the M- and the S-signals - after hard sums,
[M+S=L] and [M-S=R]. If the positive lobe of the figure-of-eight microphone is to the
right, then [M+S=R] and [M-S=L].
If these left and right signals are now added to derive the equivalent mono signal,
easy sums show that [L+R=2M]. If the matrix amplifier is used for this process, then
automatically [L-R=2S]. In other words, an existing L-R signal can be matrixed to M-S
and then subjected to any treatment which is applicable to M-S signals - manipulation
of width of image and ambience etc, for the FINAL L-R presentation. This is a very
convenient way of effecting such changes on L-R signals. When using X-Y
techniques, such changes have to be made by repositioning the microphone pair.
When the signal is in M-S format, traditionally (if a tradition there be) the stereo
spread is controlled by varying the level of the S-component. A narrower
spread is obtained by reducing the level of the S-component. Similarly, the width is
increased by increasing the S-component. This technique also has the effect
of altering perspective and ambient content, without affecting the mono signal.
A refinement of the method is to control the width by a balance control - increasing S
at the same time as decreasing M and vice-versa. This has the effect of maintaining