Got myself in a muddle about what mid and side processing is about..

Just can't visualise whats going on in this process and how changes effect the sound.

Any insights into correlation/gnomemeters(?) wiuld ve appreciated..

This may help http://www.fabfilter.com/.deo/mid-side-demystified

What we refer to as "Mid" and "Side" should properly be called "Sum" and "Difference". Technically, the former is the sum of the left and right channels and the latter is the difference (one subtracted from the other) of left and right. IOW, what's different between left and right.
 
It's how stereo is encoded onto vinyl records, with the sum written to the side-to-side (lateral) part of the groove and difference written to the up-and-down (vertical) part. It was done that way so that stereo records could be played on mono record players, which only pick up the side-to-side information. But because that's L + R, it's also conveniently the mono version of a recording.
 
The reason we care in 2014 is that it's the differences between L and R (or "side") that make a stereo mix sound stereophonic. You can boost the high frequencies and/or boost the overall level of the sides to make a mix sound wider. Or cut low frequencies in the sides to make the kick and bass sound more centered. You can compress the Mid more than the Side, and in some compressors you can get fancy and use the Side as a sidechain signal to compress the Mid and vice versa. They're all tricks to manipulate the sense of stereo width.
 
The term is also applied to a microphone technique wherein a cardiod mic picks up the "Mid" part and a figure-8 mic picks up the "Side". It's an adaptation of M/S vinyl encoding, and handled the same way at the mixer, but shouldn't be confused with M/S EQ.

The Fabfilter article is pretty good... but just in case you want more:
 
Mid-Side processing is used all over the place in audio.
 
A front facing cardiod can be paired with a side-facing bi-directional microphone to record both the mid and side signals. This provides almost infinite control over the stereo width during mixing.
 
Electronically realized mid and side signals are used for FM stereo broadcasting, and were once used to encode stereo material for storage on broadcast cartridges (which were notorious for their terrible stereo performance - the tape would slide up and down and make a real mess!)
 
And the much respected Fairchild 670 provided a way to create Mid and Side signals for processing to make it easier to cut a record... meaning quite literally cutting the vinyl.
 
Let's do a little bit of math (just a little!)
 
The Mid signal, in all of these cases, consists of the mono equivalent, what you'd hear if you put up only one microphone. It can be created, after the fact, but adding the left and right channels (and you will see it referred to as L+R)
 
The Side signal is created by subtracting the Right from the Left. This is also what you get from a side-facing bi-directional microphone.

So why is it useful?

Mid is monophonic - and compatible with mono radios, which is why it is the primary signal broadcast by FM stations.
 
If you add Mid and Side you can, in effect, decode the Left channel:
(L+R) + (L-R) = 2L - since the right channels will cancel out
 
If you subtract Side from Mid you can decode the Right channel:
(L+R) - (L-R) = (L+R) + (-L+R) = 2R - since the left channels now cancel out
 
That's how stereo FM receivers work. It also turned out that even with the tape slip-sliding around if you encoded the stereo into Mid and Side you could recover a pretty darned good facimile of the original stereo from a cart machine.
 
None of which is all that valuable to us today.

But what is valuable is we can now work on four different channels of a stereo signal! We have Left and Right, of course, but we also can have Mid, which is the mono equivalent, and Side, which is the ambiance, if you will. (yeah, not a great description, but my brain is off for the weekend.)

So I can apply equalization or compression or delay or whatever to the MID channel and only affect those things that appear in both the left and right channels, the "center" channel if you will. That's pretty powerful! And I can do the same to the side channel, but that's not always as useful<G>. All the while I can also still process or effect the left and right channels.

Clear as mud??
 
PS- Carver and Polk Audio (among others) used to use this matrix idea to create a "bigger than stereo" effect. It sounds cool, for a little bit anyway.

Here is the way I think about it … you record two tracks and then process so you end up with three tracks.
 
You record with 2 mic’s, a cardioid pointed at the sound source (called the mid) and a second mic switched to a figure 8 pick-up turned 90 degrees from the sound source pointing at the side walls.  This second mic is the side.
 
To process after recording, open up a third track and copy the side mic track into this new third track.  You now have 2 identical side tracks.  Pan one side track hard left and pan the other hard right and invert the polarity of one of the tracks.
 
You now have 3 tracks that may look like this:
 
Mid – panned right down the center
Side – panned hard right
Side – panned hard left (polarity inverted)
 
Center the image with the mid track and the start creating the illusion of a sound stage by raising the faders on the side tracks together.
 
If you collapse to mono, the two side tracks will cancel each other out so there are no phase issues.

Is it possible that one can use Channel Tools to create mid/side tracks by cloning a track and placing CT on both tracks, then turn mid way down on one, and side way down on the other? It seems like that would be true, but I'm not certain how CT is processing the audio and if the end result id the same as using a true MS encoder.
 
Regards,
Dan

You can use Channel tools to do something like this:
 

hint: Right click view image in new tab to see the image at full size.
 
You can put effects in the bins of the mid or side bus.
 
There are other free Mid Side decoders. Voxengo has a good one called MSED.
 
Although this routing is easy to do, you may find that any FX with built in mid side processing saves a lot of time and routing and so they may be preferable.
 
best regards,
mike

Somehow I'm not "demystified".
I'm not complaining!
I'm glad (well not really, more relived I guess) that I'm not the only one who's
head hurts when trying to "wrap" it around this.
I guess I AM one of the "Geezers".
 
T
fine print
AM is a registered trademark of bapu etc., etc.
end fine print
:-}

Tom, you're younger than me. Don't go throwing around the "geezer" label until you've earned it. When you can no longer lift an amp into the van, when you start wearing earplugs to concerts, when you are truly mystified by 99% of what's on the radio, only then will you qualify for your membership card.

I'll give it a go to explain it simply as a microphone technique.  
 
We know that flipping polarity cancels out sound.  That is negative voltages added to equally positive voltages results in no voltage. 
 
In the membrane of a figure 8 microphone, sound coming from the front of the mic results in standard polarity.  When the sound waves push the membrane, you get a positive voltage, when they pull it, you get a negative voltage.  The opposite happens on the rear lobe.  Pushing results in negative, and pulling results in positive.  
 
In mid side, the figure 8 microphone points the front lobe to the left and back to the right.  The channel this microphone is recorded to is duplicated.  One is panned hard left, and the other copy hard right.  On the right channel, the polarity is switched to make the rear/right lobe act like a second front lobe, but exactly opposite.  If these two channels were centered now, you would hear no sound because they cancel each other out,  but they are panned left and right on separate channels.
 
What we need now for a stereo image is a center reference. So a mid microphone is added.  It can be any pattern but is traditionally cardioid.  Since it is panned center, it occurs on both left and right channels equally.  Now anything that is similar on the left and center is equally canceled out on the right.  And anything that is similar on the center and right is equally canceled out on the left, resulting in a stereo image.  
 
Hope that makes sense.
 
MS was the very first stereo experiment Alan Blumlein divined.  Created before there was even a stereo playback format or stereo mixers.  Quite ingenious.