If a stereo image is made up of 2 separate audio files, how is it possible to EQ the 'sides' separately while leaving the 'centre' natural? Makes no sense at all :s
That is the "magic" that the Mid/Side processor does
It takes the stereo image and allows you to treat it as "the middle" and "the sides" - rather than "left & right"
If you think about the stereo image going in to the processor it is actually quite straight forward...
In a stereo signal the sound that appears to be in the middle of the stereo field has equal amounts in the left and right stereo channels - so any sound with equal amounts in the left & right channel is treated by the processor as the "MID" information
Therefore the "SIDE" information is audio that has differing amounts in the left & right channels
The Mid/Side processor is then able to treat these 2 streams of audio (Mid & Side) in completely different ways
At the end of the process the signal is combined back to a standard stereo "Left & Right" signal
Some good examples of use are...
Adding some reverb to just the "side" information to give the track some "space" but retaining a solid focus in the centre (eg the vocal)
You could even use Mid/Side processing on the reverb return - eg using an EQ you could scoop out the vocal frequencies in the MID so that it doesn't "muddy" the vocal but leave it present in the SIDES
Another good use is on BASS, especially synth bass
If the synth bass has a stereo effect you can use MID/SIDE processing to keep the bass mono and punchy in the centre of the stereo field and then be a little more adventurous with the SIDES for some stereo "excitement", whilst keeping the weight of the bass centred
How about running all your guitars to a BUS and then using MID/SIDE processing to bring up the volume of just the SIDE signal of all the guitars up for a chorus or solo section - ie increase the volume of all your guitars but just in the SIDES, without increasing the volume of the MID portion of the guitar signal, so the vocal is still clear
A common trick of MID/SIDE processing is to add a high frequency EQ boost to just the SIDE portion of the signal for a particular instrument or vocal
This gives the impression of a wider stereo spread, whilst leaving the important CENTRE information in tact
On an acoustic track if you are mastering and want to "warm up" the track you can experiment with adding reverb to just the SIDES so that you don't muddy up your bass and kick (which are likely to be in the centre - MID - portion of the signal) - or add reverb to the MID but scoop out the "lows" to keep it clean
There are other ways to achieve these "tricks" but MID/SIDE is a great technique to get your head around as another tool in the box
Hope this helps TONE