Mid-side, sum and difference, etc... Tips to a noob?

I've been reading up a but on techniques, practical uses and the such, not finding a WHOLE lot in the way of when/where to use it... So, first question is HOW do you encode in DAW? What plugins are you guys using and why? Secondly, WHEN? What type or genre of music can benefit from MS processing? And then WHAT? What do you typically do? Squash the center more, leaving the sides more dynamically exciting, widening the sound-stage? Or are you trying to do the opposite? Cut a dynamics "hole" into the center so you vocals and such can pop out more?

Be gentle, I'm still very new to mixing in general (only been teaching myself for about a year now)...

Man I had a nice reply going then my browser pooped on me.....

But here is what I use http://www.voxengo.com/product/msed/ I like it alot.  To record I'll set up a stereo track and record the mid to one side and the side to the other.  Then pop this plugin on it and it gives me control of the signal.

As for when, I use it just about everywhere I can for a stereo recording, but my fav right now is drum room.  2trk mixes can benifit as well nice tool for mastering.  I'd use this on just about any genre really.

What do I do with it.....sometimes I leave it alone sometimes I squish the mid and leave the sides, it's really a taste thing at this point for you.

Not as long as my first reply but it gets the point across.

-Casey

Any MS plugin will do.  They are all the same. 

Drum Room. Acoustic instruments. Piano.  Any time you want a very natural image.

As far as MS processing goes, it can be used different ways in different situations.  It can be used to de-mud the center without affecting the sides.  I can be used to bring the sides and ambience out more in the mix.  I can be used to bring out the vocal more.  It can be used to fatten a kick drum.

In other words, play...  Have fun... Learn... Mess a few mixes up... See what happens.

The plugins are doing very little - IMO, it is very easy to just set up three tracks and have all the control in your faders. I set one track as the mid (with the record source set to the mid mic) and two tracks as the sides (recording from the same input source, i.e. your side mic). Pan the two side tracks fully L and R and invert the phase on one of the tracks (if the front of your side mic is facing the left, invert the right track). I usually send the three tracks to a bus and adjust the level of the sides with respect to the mid to get a nice stereo image. I think if you have more compression on the sides you will bring out the 'room sound' more by accentuating the tails. This may or may not be what you want depending on source, style, taste.

Drum Room. Acoustic instruments. Piano.  Any time you want a very natural image.

Click to expand...

+1
For certain mixes, recording a guitar amp with m-s will really bring out a full, natural sound.

Ther's a HUGE difference between using M/S processing on a stereo signal, and recording a set with M/S techniques.
The former consists in converting L/R to M/S, processing the resulting signals with EQ, dynamics, whatever, followed by converting M/S back to L/R. This offers the possibility to modify the stereo image without altering the overall balance, or making the mix punchier without narrowing or shifting the image, but this can also be done in L/R format, albeit less easily.
Recording in M/S mode is a different thing; it takes a cardioid or omni mic tocapture the M signal (very often a sub-cardioid is prefered) and a fig-8 mic for the S signal. the big advantage is that the signal is 100% mono-compatible, and in a general manner, most of the signal energy being defined by only one microphone, the result is punchier and less affected with phase interference.

Good stuff guys, keep it coming...

I've already messed up more than a few mixes... I've been using the waves S1 MS matric plug, which does nothing but encode/decode and drop the level -3dBm each instance. I'm trying to figure out how to encode, then patch the mid(left) to one track, side(right) to another so they can be processed in dual mono. Anyone know how to patch L/R to separate mono tracks in Cubase 5?

I prefer MS in the situations with less than ideal room, or instrument. It gives a flexibility of changing stereo width, or separate EQuing of the M (instrument) and S (room) channels, so I prefer to keep M and S separately, and then matrix only during mastering stage. For S I much prefer ribbons over condensers set into fig8.
Mathematically, the MS with cardioid is equal to XY, so in principle, you can de-matrix any XY recording, make those processing manipulations and then re-matrix again.

Don't afraid to experiment for M with fig8 (equal to Blumlein), or even omni (I think, that would be binarual). For matrixing you can use any DAW software. Copy the S channel to separate track, flip its phase, and then pan the M to the center and both S to extreme right and left, respectively. Changing the levels of the M and/or S (both L/R simultaneously) you will change the stereo width.

Best, M

There's a HUGE difference between using M/S processing on a stereo signal, and recording a set with M/S techniques.

Click to expand...

Ah, I really missed the boat - the OP was talking about M-S processing during mixing / mastering, not recording.
I have no experience so I'll keep quiet from now on and just be reading...

dmp said:

There's a HUGE difference between using M/S processing on a stereo signal, and recording a set with M/S techniques.

Click to expand...

Ah, I really missed the boat - the OP was talking about M-S processing during mixing / mastering, not recording.

Click to expand...

Huh, same here. I'd think the MS processing will work rather for coincident stereo techniques, but who knows...

Best, M

I usually only hear of it used in mastering, as another way to get at parts of a mix that can't be remixed.  I used a tube limiter set once as an AGC in mid-side while mixing something specifically for vinyl, and liked the effect I was able to achieve.  No hard and fast rules here.

Interesting technique from Joel Hamilton for using sum/difference to mess with stereo overheads: http://messageboard.tapeop.com/viewtopic.php?t=29416&postdays=0&postorder=asc&&start=0

JH uses M/S techniques without an M/S set-up.
What he does is basically cross-feeds L into right and R into left OUT-OF-PHASE. That's equivalent to turning down the M fader in an M/S set-up.
Since M represents the "meat" of the signal and S the "air", the result is that the sound is less punchy but more open.
Particularly with spaced OH, where low fequencies are less out-of-phase than HF, this results in a relative attenuation of LF.

http://www.brainworx-music.de/en/plugins/bx_solo

cool little plugin... Nice to let you know what you're listening to, but other than that it's pretty limited. But that's all it was meant for, a little M/S teaser.

sorry for the stupid question. Do the the plugins assume that left is mid and right is side? I mean if I record a m/s onto a stereo track do i have to
ensure that mid is left and side is right or should i record to 2 seperate mono tracks (although I have not figured out how to get them into the same plugin).

generally left is mid and right is side when you're talking about a stereo channel.

tazwolf said:

sorry for the stupid question. Do the the plugins assume that left is mid and right is side? I mean if I record a m/s onto a stereo track do i have to
ensure that mid is left and side is right or should i record to 2 seperate mono tracks (although I have not figured out how to get them into the same plugin).

Click to expand...

If you record two seperate mono tracks, you could pan M hard left and S hard right, then send them both to the same stereo buss, and then put your encoding plugin on that buss.

Or you could just duplicate the S channel, reverse the polarity on one, and pan them hard left and right. This is basically all that the plugin is doing anyway.