Just some notes on this "older" discussion:
With MS techniques you can do all kinds of processing....and do it very well!
The thing is....IT IS ABSOLUTELY NEEDED TO KEEP THE PHASE OF M AND S ABSOLUTELY THE SAME.
With this in mind, you can EQ and more....dNa-CompANY-XT and dNa-Stereo Equaliser are MS capable devices
for which i created with own MS encoders / decoders, and i put a lot of time in the phase guarantee!
(also the dNa-Summer-of-82 has MS processing on ALL the stereo input channels, giving lots of control
in placement. Dante did some nice testing on that, and has pointed that feature out of the summer)
Because processing can cause sample-delays through a not-so-well designed plugin, resulting
in very weird results. (caused by dsp-placements etc)
Simple:
M is created by adding L and R together
S is created by substracting L and R from eachother.
In math:
M=L+R
S=L-R
So...how do you make it back to L/R mix?....basically:
L=(M+S)/2
R=(M-S)/2
That is basically the trick...
It is no magic, just simple math, and a good judgement ear
If you put a encoder and decoder in series on a stereomix, it should be ABSOLUTELY THE SAME!!!!! If not....something is very wrong.
This is why you cannot take an MS encoder, chop in some EQ or comp, or seperate processing on one of the channels, and mangle it back together with the decoder....there is the delay. A good MS device is designed "as is"...like for example the dNa-Stereo Equaliser....all is in the device, and i can guarantee
phase-alignment is stable, no matter what scope-system you load it on, or when you reload your project (which can cause different dsp placements)
which will cause a horrible soundfield.
