Mid Side

Anyone play with Mid-Side processing on the mix? I just discovered how to do this with the new channel tool and its rather a neat tool.

I want to try this technique out as well

My only half decent mic is an AT4033 which doesn't have any switchable patterns.

Would an SM58 work? Is it an Omni? Can't remember the spec.

I just received S8:

To be fair: Channel -Tools VST, a new VST for the Mid Side recording technique, now gives crystal clear info and technique for M-S Recording and phasing of waveforms (much easier than Sonitus Phase and Voxengo OverTones). This is a plus for Sonar 8: Channel-Tools is 'God-send' for me.

I've been having repeated dreams about getting the M-S Miking/mixing technique operational, and to show it off with vox hooks and such. (E.g., panning wide during hooks and narrowing center for verses ... Yee-ow

But as Frank stated you must have a omni (figure-8 'Side') mic (like the AKG C414B) operational.

If/when I get this working right, Lord-willing, I'll report.

ORIGINAL: Philip

I just received S8:

To be fair: Channel -Tools VST, a new VST for the Mid Side recording technique, now gives crystal clear info and technique for M-S Recording and phasing of waveforms (much easier than Sonitus Phase and Voxengo OverTones). This is a plus for Sonar 8: Channel-Tools is 'God-send' for me.

I've been having repeated dreams about getting the M-S Miking/mixing technique operational, and to show it off with vox hooks and such. (E.g., panning wide during hooks and narrowing center for verses ... Yee-ow

But as Frank stated you must have a omni (figure-8 'Side') mic (like the AKG C414B) operational.

If/when I get this working right, Lord-willing, I'll report.

Channel

OK, it works, finally. The mics have to be close to each other (vertically).
Channel-Tools VST plugin OUT-Monitor gives you a visual of LT and RT as does the output bus (its waveform preview - On):

Interesting Key Results:
1) You can somewhat see the LT channel sticking-up on your (RT channel) waveform ... and the RT channel sticking-down on the same (RT channel) waveform.
2) Covering one side of the figure-8 mic does not seem to help 'enhance' the other side (I think you're covering the inverse polarity that's used in the algorhythm)
3) The cardiod mid-mic must be present and inbetween the figure-8 field ... close to center.
4) Results are not audibly nor visually that dramatic
5) You will have to adjust the mid-gains and side-gains for slight improvements
6) Don't expect pure LT and RT panning: the panned results equate to about 30% LT and 30% RT (based on 'my ears' and visuals)

IOWs (Tentative Conclusions):
This is too complex (for 90% of us).
Your 'LT' Channel will always bleed to your 'RT' using this technique. You probably CAN NOT get a widened pan effect because the mid-mic must always be present ... and the figure-8 mic also bleeds LT and RT. Its better to buy the M-S Mic(s) pre-built into a solid configuration. I wouldn't play much with this at home.
(Room) Ambience alone is an advantage using this technique. You can angle the phases and such to help (ambience).
Us rock-stars and effects-mongers will probably not use this technique for chorus panning.
We'd probably prefer to use 2 cardiod vocal mics angled (and phase tested) for our particular vocal booths.

Hope this helps.

For anyone that's interested, I've been playing with M/S techniques for mastering tracks.
sample_1.mp3
sample_2.mp3
These are a couple of brief excerpts from a live show I recorded in the summer. Both are the same basic Sonar stereo export mix, sample 1 is 'mastered' with all the normal tools (Ozone 3, Boost 11), sample 2 using M/s to gain access to the center (mono) and LR (stereo) information in the track and treating each set of information in different ways to acheive a very different sound from the same basic track.

I'm having a lot of fun, and it's been a bit of an eye (ear?) opener as to how the different sets of information extracted by M/S processing can be treated.

I'll elaborate if there's any interest.

The guy in this video: http://revver.com/video/895784/home-made-hit-show-tv-011-tomas-de-camino-beck/ gets some great results using the midside technique. This is what has inspired me to give it a whirl.

FWIW some of the cheap new ribbon mics might make a great starter figure 8 mic for the "sides".

There is a short video and a load of usefull info on Mid/Side at http://www.recording-microphones.co.uk/Mid-Side-stereo.shtml

john

Wow, this thread took on a life of its own weeks after I posted.

The original intent was to play with the Sonar Channel Tool and I have. I built a bussing system which creates a mid/side split to the mix and then applies effects and EQ different on the sides. In fact I went a little beyond that and created a secondary buss structure for front to back as well. Anything on the back buss gets varying degrees of a room verb. Only the vocal and featured instruments run in the front buss which is drier sounding.

Overall I can vary the width and depth of the mix in ways, prior to the Channel Tool, I was not able to do. Of course you have to watch things getting too out of phase so I check with my monitoring mono switch to make sure things don't get too out of hand.

Fun stuff.

Philip -

So sorry, I forgot about this thread.
Talking about M/S mastering techniques - the secret is using Voxengo's MSED tool to separate out and independently treat the center track and the L+R stereo tracks as three separate tracks in Sonar.

This means bouncing the original track to tracks twice.
Put the original stereo file in a stereo track in Sonar.
Insert MSED in the FX bin.
Bounce your original stereo file thru the MSED to a new mono track, using the center setting in MSED at -6db, with the side control set to -∞



Bounce your original stereo file thru MSED to split mono to two additional mono tracks using the center setting in MSED at -∞, with the side control set to -6db.



Now you have four tracks in Sonar -
the original stereo,
one center mono Mid, panned to mid
one Left Side, panned hard L
one Right side, panned hard R





Right away you can play with the widening effects, balancing the mid level against the sides, playing with the pan % in the sides, mixing in some of the original stereo track, etc.

BUT THERE'S MORE....
you can also independently treat the center track, with limiting, compression, thickening, bass enhancement, anything you want, without affecting the Side tracks.

Since the center track usually contains all the important 'core' stuff - in rock & pop at least - kik, snare, vocal, bass - this means you have a chance to really work on this - almost like a separate 'stem' at mastering.

With the side tracks - they usually benefit from treatments which add 'air' or 'space', maybe some harmonic sythesis à la Ozone or BBE, or multiband compression emphasising the 'sparkle'. Be aware that any stereo effects, such as wide 'verbs or leslie spkr FX will really jump out at this point, you may want to remix the track to de-emphasise these.

I would just like to point out that if you switch it out of "inline" mode, then you can accomplish the same thing, in realtime, by using appropriate bus routing (and having those output to another bus with the MSED set to decode). Just another option for those that don't like bouncing to tracks and something to keep in mind, since the "inline" mode was really just designed for people that wanted to alter the balance between M and S without doing separate processing on each.

- Original Track: has Voxengo MSED set to "Encoder" mode. Now M/S gets switched to L/R. So the left channel of the output is "mid" channel and the right channel of the output is the "side" channel.
- Mid Buss: From the original track, create a send that goes to the "Mid Buss". Make sure that the send to this buss is panned hard left and that the output of the buss is panned hard left as well.
- Side Buss: From the original track, create a send that goes to the "Side Buss". Make sure that the send to this buss is panned hard right and that the output of the buss is panned hard right as well.
MS Decoder Buss: Set the "Mid Buss" and "Side Buss" to output to this buss and make sure the first plug-in insert is Voxengo MSED. Set the plug-in to "Decoder" mode.


This allows you to EQ, process, etc. the mid and side individually without having to render tracks in between. Watch out for volume differences caused by panning laws.

I'm sorry to drag this one back up from the mists of time, but in Creator's example above, when he splits out his mid & side channels, do these then get routed directly to the mid & side buss?

I understand the example posted by macflooze perfectly, but creator's is leaving me scratching my head a bit, though this is definitely the way to go for me, as I don't really want to start bouncing my stereo wav down to split mono if it can be done just through routing.

Ah! The man himself!!

Thanks for that - might have a go at this tonight then.

What do you do with the actual output of the orignal stereo track?

Thanks for commenting.

My favorite uses for mid-side in the mix/master:

Drum compression- this makes it much easier to compress the kick and snare in the center without having the cymbals "pump" with each hit

Multiband compression- similar to my reasons for drum compression in mid-side, you can use a multi-band compressor in mid-side before your limiter when mastering to get higher gain and pumping only where/how much you want

Mastering- fixing it in the master. I've found that with Mid-Side, linear phase EQ, and transient designer, I can isolate a signal that I want to add more processing to. Say one song on an album has noticeably less reverb on the vocals and seems out of character compared to the rest of them. Using this method I can add reverb to the vocals with very little affect to anything else.

Peace,

Tunes