Pseudo Stereo – When Stereo isn't possible or practical "spaciousness" may be an adequate substitute
Pseudo Stereo – When true stereo is not feasible or possible, “spaciousness” can be achieved with a relatively simple playback setup.
For many years the conventional wisdom said that stereo PA was impossible. Indeed the accurate phantom imaging that a good stereo system can achieve in the living room cannot be duplicated over a larger audience area. For most seats the image locations will collapse to one of the loudspeaker locations. However phantom imaging is not the holy grail of stereo. Two, three and even five channel systems have been done quite successfully when the designer is content to allow imaging at the speaker locations only and not across a continuous field in between. These systems can have excellent speech intelligibility and reasonably realistic sound stage presentation, but can multiply equipment costs, as each channel needs to cover the entire room.
For smaller rooms there is a very attractive alternative that works with a different set of compromises. This is the Sum/Difference method. It consists of a single central main speaker or cluster, essentially identical to what you would use as a mono system. This speaker gets the Left-plus-Right mono sum. A second speaker system consisting of two or more speakers along the side walls gets the Left-minus-Right difference signal. Thus anything presented equally in Left and Right is reproduced by the center speaker only, a mono signal with a firm center image at every seat because a real loudspeaker is at that location. The single point speaker system will have high speech intelligibility assuming it’s designed correctly.
Any signal presented in Left only or Right only will be reproduced by both the main central speaker and the side speakers. The side speakers should be deliberately designed to provide a diffuse unlocalizable sound. This gives the sound a spacious room filling characteristic. There is no specific left or right localization. A left to center to right panning action will be reproduced as spacious – direct – spacious. However many of today’s recordings don’t make strong use of side to side panning for the sake of imaging but instead use the sides for spaciousness and effects, while the center is used for the main dominant instruments and voices.The Sum/Difference method is particularly suited to training videos where the talking head is always panned center and stereo spread is only used for the music bed and audience effects.
The Sum/Difference system starts with a normal main mono speaker or cluster front and center. I need not go into any design details since it is no different than a mono system. The side speakers can be high passed at 300 Hz so they can be small. They should have a wide radiation angle and be aimed at the walls or ceiling. Two speakers will work, but more are better. Reverse the polarity of every other one. Ideally these speakers will be aimed at walls which are diffuse, not at people’s ears. Where indirect lighting troughs are used the speakers can be hidden in there, pointing up at the ceiling.
Processing is best done in two channels, sum and difference. Obtaining the sum and difference from the LeftRight is simple in a free wire DSP, unfortunately many makers of fixed path DSPs haven’t caught on to including polarity selection in the cross matrix. When balanced inputs and outputs are available, passive combining can be used to derive the sum and difference in the analog connections. [Figure 1]
Equalize and voice the sum speaker as you normally would. Equalize the difference channel using pink noise and an RTA or other time-blind measurement. Roll off everything below 300 Hz, and the high end may need a gradual roll off starting at several KHz. This adjustment is best done by ear, based on the room acoustics, normal program material and desired effect. Since there will be many deliberate phase shifts and polarity reversals caused by the diffused reflections, there is no point in trying to do any time-windowed measurements. However delays should be set so that no difference speaker arrives at any listener position earlier than 5 milliseconds behind the direct sound from the sum speaker.
In a small room this simple technique, which can often be implemented for only a few hundred dollars, can add significant space and realism to music reproduction, while maintaining the optimum single source for speech intelligibility. ds
The most important of my discoveries have been suggested to me by my failures.
Sir Humphry Davy