Research into loudspeaker stereo has shown that spaciousness, depths, and directional accuracy depend on factors that can be easily calculated for a particular recording technique. Factors such as the time and level differences between the channels, and the correlation of the reflected energy, can be combined with information about typical playback rooms to predict how a given overhead microphone technique will perform. We find it is in general not possible to meet all the requirements of an ideal technique without employing either non-standard microphone patterns, independent equalization of the L-R and L+R signals (spatial equalization,) or both. Since small time delays do not provide directional information below about 500Hz, techniques which separate the microphones by small distances require additional compensation. With spatial equalization coincident and semi coincident techniques can make recordings of high quality. This is a paper about loudspeaker stereo. I will examine the spatial properties of various microphone techniques with the explicit expectation that they will be played back for the most part through loudspeakers in various positions in a relatively small room (about 40 cubic meters.) Spatial information at low frequencies is perceived quite differently in larger rooms, and entirely differently in earphones.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.