M-S Mastering The use of a middle (M) and a side (S) signal instead of the usual L/R (X/Y) signal can result in a much more selective and musical processing. The main reason for splitting the L & R stereo mix into M (sum of L and R) and S (the difference between L and R) is so that different processing can be performed on the M and S signals before they are re-combined into a new L and R. High-energy Middle signals (vocals, snare, bass guitars, etc.) are separated from Side signals (guitars, keyboards, cymbals, etc.). As a simple example, if you insert a high-pass filter on the S channel, when re-combined to L and R, all the low frequencies will now be located in the centre of the stereo image if any were off to one side previously. Another example would be the use of different degrees and type of reverb on the M and S channels. Separate processing of the M and S channels can produce spatial ambiance that is not easily possible to create by direct processing of the L-R stereo mix. A compressor can be used to process one element instead of the whole mix. The same applies to any kind of processing, making the M-S matrix an indispensable element in the mastering stage. M-S encoding and decoding processes are identical and theoretically lossless, so can be performed in principle any number of times without signal degradation. As with analog summing, I have found empirically that more acoustically pleasing results can be obtained using external analog processing for these tasks than performing the M-S coding in the digital domain.