Maintain proper polarity at each stage of the audio signal chain.
This will ensure that end-to-end, a positive going sound pressure wave impinging on a microphone will ultimately result in a positive pressure wave emitting from a monitor loudspeaker. It will also ensure that every patch point will have the correct polarity, making it very difficult to inadvertently reverse polarity when cross-patching.
Why is this important? In a multi-channel system, correct polarity for all channels means that the generated sound field will bear a close resemblance to that intended by the audio mixer.
If any of the channels are out of polarity when they mix with the in-polarity signals, either acoustically (as in sound field created by a surround sound system) or electronically (as in a mono TV set), the resultant mix will have cancellations in parts of the frequency spectrum. These cancellations are dynamic, in that they change as the spectral content of the program material changes over time. The resultant signal often is lower in level and has a filtered kind of sound.
Cancellations that occur in the voice frequency range, especially in the critical 500 Hz and 2000 Hz octave bands, are very problematic. These are the main speech intelligibility frequency bands, and any cancellations here will result in low to no dialog level. Also sound effects tend to be rather band-limited, and if these are the bands that get cancelled, so will the effects.
If unintentional out-of-polarity signals end up in a left/right matrix, a surround sound decoder may process the signal with unpredictable results. Sounds intended for the front may end up in the rear, or vice versa, and even shift around. Dialog may drift in and out of center, or disappear entirely. These are just a few examples of what could happen.
