Taperssection.com

Gear / Technical Help => Post-Processing, Computer / Streaming / Internet Devices & Related Activity => Topic started by: Bdifr78 on April 04, 2006, 02:24:40 PM

Title: How do you mix down an M/S recording
Post by: Bdifr78 on April 04, 2006, 02:24:40 PM
Hey there folks,

I an not a wiz with the whole post production side of things.  I just got an LSD2 and really want to run M/S, but I don't know what to do with it after I pull the recording.  I know how to get to the M/S plug-in on Sound Forge, but once I am there it's like japanese to me.  Is there a good tutorial or reference that anyone knows of? I also want to do a bastard M/S with one km140 and one 414 ;D


Thanks,
Joe
Title: Re: How do you mix down an M/S recording
Post by: kindms on April 04, 2006, 02:32:22 PM
well basically once you have it on the computer its a matter of listening to it and dialing in the width etc using the plugin. I dont think there is really a "right" way to do it. You will just want to either have a really good playback or headphones to do it.

when you start messing with the stereo image it will be pretty in your face wahne you get it to where you want it
Title: Re: How do you mix down an M/S recording
Post by: mmmatt on April 04, 2006, 02:36:48 PM
basically all you are doing is putting the fig 8 mic and the card into both channels.  The fig 8 will be reversed polarity on one and as recorded on the other.  The amount of fig 8 vs card will determine how wide the image is.  Think of the crad as mono, and the 8 as a hard-panned l/r.

Matt
Title: Re: How do you mix down an M/S recording
Post by: Brian Skalinder on April 04, 2006, 02:38:48 PM
I don't know how the SF interface works for M/S, but I find it useful to start out listening with two "end-points":  one super-wide (lots of S in the mix) and one super mono (lots of M in the mix) to get a feel for how the extremes sound.  Then I start to narrow the gap between the two end points until I find a range that sounds good to my ears, i.e. neither too obviously wide nor too obviously narrow.  Finally, within that "acceptable" range I adjust the mix until I find the point where it sounds best to my ears.
Title: Re: How do you mix down an M/S recording
Post by: Bdifr78 on April 04, 2006, 06:47:31 PM
basically all you are doing is putting the fig 8 mic and the card into both channels.  The fig 8 will be reversed polarity on one and as recorded on the other.  The amount of fig 8 vs card will determine how wide the image is.  Think of the crad as mono, and the 8 as a hard-panned l/r.

Matt

So adjusting the width is like pulling more of the mid channel into the mix, or the opposite?  What confuses me is that both channels are independent on the raw recording, but if I understand it correctly after it is mixed both channels should be distributed to your tastes in both tracks of the recording?  How do you get one channel into the other?  Do the M/S plugins do this for you and then you just adjust it the way you like?  I get the concept, and it sounds sweet, just still a little hazy on what exactley goes on in the mixing.
Title: Re: How do you mix down an M/S recording
Post by: MattD on April 04, 2006, 09:00:54 PM
The plugins do the mixing for you. You tell it how much of each source (M-S) you want and the plugin output is L-R.

Maybe an analog explanation might make things clearer for you:
To do this on a mixer, you need three channels. The mid channel is panned center in the mix. The side channel is panned 100% left. The third channel is the phase inverse of the side channel. Channels 2 and 3 (side and side inverse) change volume together. Adjust relative volumes of 1 vs. 2-3 to taste. If you're doing this in a DAW, make sure your sum doesn't clip. Most plugins will handle the clipping issue for you.
Title: Re: How do you mix down an M/S recording
Post by: mmmatt on April 04, 2006, 09:03:13 PM
basically all you are doing is putting the fig 8 mic and the card into both channels.  The fig 8 will be reversed polarity on one and as recorded on the other.  The amount of fig 8 vs card will determine how wide the image is.  Think of the crad as mono, and the 8 as a hard-panned l/r.

Matt

So adjusting the width is like pulling more of the mid channel into the mix, or the opposite?  What confuses me is that both channels are independent on the raw recording, but if I understand it correctly after it is mixed both channels should be distributed to your tastes in both tracks of the recording?  How do you get one channel into the other?  Do the M/S plugins do this for you and then you just adjust it the way you like?  I get the concept, and it sounds sweet, just still a little hazy on what exactley goes on in the mixing.

I've never done it, but it is my understanding that you could do it manually. Or the plugin will do it for you.  You would have to split the wav to two mono sources.  Take the fig 8 side and make a copy of it.  On one of the figure 8 channels you flip polarity.  Now you have a left figure 8 and a right figure 8.  Then you mix the card to each channel to taste.  The more card, the more the image will be centered... too much and it will sound mono.

Matt

*edit*  what Matt D said
Title: Re: How do you mix down an M/S recording
Post by: Bdifr78 on April 04, 2006, 09:17:16 PM
The plugins do the mixing for you. You tell it how much of each source (M-S) you want and the plugin output is L-R.


Thanks, that's all I needed to bridge the gap.  Mid-Side recording here I come!!!

Real quick though, what situations are just bad for M/S?
Title: Re: How do you mix down an M/S recording
Post by: audBall on April 04, 2006, 10:21:38 PM
hijack!


would running two cp4's and an omni/card into an mx100 be considered MS as well?  There's a center panned to both channels and an independept L and R as well. Would the sum/difference occur in the same way? of course, this would all be done on the fly.

I've never seen this discussed before...sorry if it's a dumb question.   
Title: Re: How do you mix down an M/S recording
Post by: BobW on April 04, 2006, 11:03:51 PM
Here's some chatter on how to do it.
http://www.kvraudio.com/forum/viewtopic.php?t=90106

Here's a Free VST M+S Plug-in   (and a few other cool plugs, as well)

Voxengo MSED

http://www.voxengo.com/files/VoxengoMSED_10_WinVST_setup.zip

Other stuff:
http://www.voxengo.com/downloads/
Title: Re: How do you mix down an M/S recording
Post by: MattD on April 04, 2006, 11:41:17 PM
For the OSX folks, I posted a free AU M/S plugin a while back. If I can find it, I'll link it here.

http://taperssection.com/index.php?topic=53527.0
Title: Re: How do you mix down an M/S recording
Post by: BobW on April 15, 2006, 09:12:45 AM
hijack!


would running two cp4's and an omni/card into an mx100 be considered MS as well?  There's a center panned to both channels and an independept L and R as well. Would the sum/difference occur in the same way? of course, this would all be done on the fly.

I've never seen this discussed before...sorry if it's a dumb question.   


They seem to be very specific about these things.  Classic M+S is a Blumlein figure of eight at 90^ to source and a cardioid center toward it with space between (coincident).  Schoeps has a Stereo mic that has a supercard front pick-up. (CMIT5U)
In concert recording, I wonder how much valid Side information occurs at 30 to 150 feet from source, but I have no experiences to back that up. (I'm dying to hear the results of the CMIT5U from DMB at Roosevelt Island.)
   [EDIT]  Shotgun, but not a stereo mic, my bad.

That does not mean that other techniques aren't good, better or invalid.
The big difference that I see is that the L-R needs to be out of phase. 
(Teddy or Nick, check me on this one  -  Fig-8 has a front lobe/back lobe phase inversion, doesn't it ?)

A classic three mic technique would be a Decca Tree, good for big band and orchestra due to the wide source.
There have been many engineers who have tweaked the Decca tree.
It sounds like you are on to something. Make it your own and run with it.
Title: Re: How do you mix down an M/S recording
Post by: audBall on April 15, 2006, 10:34:20 AM
hijack!


would running two cp4's and an omni/card into an mx100 be considered MS as well?  There's a center panned to both channels and an independept L and R as well. Would the sum/difference occur in the same way? of course, this would all be done on the fly.

I've never seen this discussed before...sorry if it's a dumb question.   


They seem to be very specific about these things.  Classic M+S is a Blumlein figure of eight at 90^ to source and a cardioid center toward it with space between (coincident).  Schoeps has a Stereo mic that has a supercard front pick-up. (CMIT5U)
In concert recording, I wonder how much valid Side information occurs at 30 to 150 feet from source, but I have no experiences to back that up. (I'm dying to hear the results of the CMIT5U from DMB at Roosevelt Island.)

That does not mean that other techniques aren't good, better or invalid.
The big difference that I see is that the L-R needs to be out of phase. 
(Teddy or Nick, check me on this one  -  Fig-8 has a front lobe/back lobe phase inversion, doesn't it ?)

A classic three mic technique would be a Decca Tree, good for big band and orchestra due to the wide source.
There have been many engineers who have tweaked the Decca tree.
It sounds like you are on to something. Make it your own and run with it.


I think you're right about the phase inversion thing.  It now makes better sense. 

Thanks  :)
Title: Re: How do you mix down an M/S recording
Post by: BobW on April 15, 2006, 11:02:41 AM
OK, but I am dead wrong on the Schoeps, it is just a Shotgun, not a stereo mic !