mid side compression question

An audio community like no other.

Moderators: rick, Mark Bassett

mid side compression question

Postby Henry » Wed Aug 17, 2005 11:11 pm

I read a mention about mid side compression, I believe it is being able to compress the middle of the song leaving the stereo image alone? Can anyone explain what the technique is and how this can be done digitally, ie. in the box? Thanks.
Henry
Registered User
Registered User
 
Posts: 127
Joined: Fri May 06, 2005 11:52 pm
Location: Sydney

Postby chris p » Thu Aug 18, 2005 9:10 am

Decoding MS and LR is just sum and difference, both ways, and can be done in analog or digital.

To decode MS into LR, it's simply L = M + S and R = M - S

To encode MS from LR (like the MS compressors do), M = L/2 + R/2 while S = L/2 - R/2. You can then compress the mid bit separately as you wish, and then decode the compressed signal back to LR.

The point of it all for compression, as far as I understand, is to avoid some of the compression stereo phase problems becuase you are only in effect compressing a mono signal.

An analog MS to LR schematic can be found here.

I've just bought the Paia MS microphone kit which adopts the usual MS recording technique - rather than a cariod and an omni, it used 3 cardiods but sums them all into a balanced LR signal within the box. I suspect at some point I'll upgrade the supplied electret mics with better capsules, but the principle of the thing looks sound.[/url]
User avatar
chris p
Frequent Contributor
Frequent Contributor
 
Posts: 882
Joined: Tue May 10, 2005 6:15 pm
Location: Sydney, NSW

Postby PeterR » Thu Aug 18, 2005 10:57 am

I just bought 10 transound 165a capsules with the specific intention of making a MS capable set of mics (3 side address cards for easy placement - making a mount for them, but can use them seperately if I like - decode using fancy bussing in PC).

These might be good upgrade for your Paia Chris? They arrived a few days ago. All up it cost about $80Aus to get 10 of them here.

the good thing about these electrets is there is no crappy FET built in.
PeterR
Registered User
Registered User
 
Posts: 57
Joined: Sat May 07, 2005 12:18 am
Location: Ipswich, Qld

Postby Kurt » Thu Aug 18, 2005 3:11 pm

There are a few MS encoders around (that convert Left+Right to Mid+Side) in software, there's a "suite" including MS Compressor here: http://mdsp.smartelectronix.com/mstools/

Worth a poke arounfd the rest of the site too, there's alot of cool, free, stuff.
User avatar
Kurt
Valued Contributor
Valued Contributor
 
Posts: 1224
Joined: Sat May 07, 2005 1:02 am
Location: Canberra

Postby Adam Dempsey » Thu Aug 18, 2005 4:16 pm

Traditionally used more to tame or eliminate out of phase low frequencies, stemming from requirements in the vinyl cutting process. Used it this morning on an electronic track (backing track made on a Sony PS1!)
Also great for pushing vocals 'forward' with gentle mid EQ without affecting what's panned in the mix.
As an engineer friend often says: "The possibilities are mindless".
User avatar
Adam Dempsey
Registered User
Registered User
 
Posts: 139
Joined: Mon May 09, 2005 5:39 pm
Location: Melbourne

Postby Mark Bassett » Thu Aug 18, 2005 11:22 pm

Mindless indeed, especially the ability to adjust image width as a function of frequency, panning the center image and even rotating the entire sound stage.

MS can also be used to 'correct' the lack of bass width that often accompanies single point stereo microphone techniques. i.e boosting the bottom end of the S component (known as Shuffling) can give the stereo image more spaciousness - basically introducing out of phase low frequencies which can cause issues (see above).

This is more mixing related but I did a MS recording of the Devonshire St tunnel at Central Station in Syd a few months back using a Schoeps MS rig (see below) and a Nagra V 24/96. When i walked out of the tunnel I was lucky enough to capture two buskers (double bass and sax) jammin on the side of the street. In the studio i was able to vary the 'street noise to busker ratio' using an MS processor. I could eliminate busses driving past almost completely while leaving the buskers in tact. I ended up automating the S channel to get the perfect mix of surry hills street sounds and sidewalk jazz. I shuffled the bass which definately gave the whole recording more 'believeability' (however subtle) while pushing up the S in the tunnel for obvious reasons.

Image
Schoeps MS Setup
User avatar
Mark Bassett
Forum Admin
Forum Admin
 
Posts: 538
Joined: Fri May 06, 2005 6:15 pm
Location: Byron Bay

Postby Henry » Fri Aug 19, 2005 7:55 pm

Thx guys. But I'm still not sure I unerstand it?
Henry
Registered User
Registered User
 
Posts: 127
Joined: Fri May 06, 2005 11:52 pm
Location: Sydney

Postby Kurt » Fri Aug 19, 2005 10:30 pm

Just have a play with it, best way to understand anything :)
User avatar
Kurt
Valued Contributor
Valued Contributor
 
Posts: 1224
Joined: Sat May 07, 2005 1:02 am
Location: Canberra

Postby Mark Bassett » Tue Aug 23, 2005 4:31 pm

To break it down MS is simply a different way to think about sound. Blumlein proposed that music isn
Last edited by Mark Bassett on Mon Sep 12, 2005 10:31 pm, edited 1 time in total.
User avatar
Mark Bassett
Forum Admin
Forum Admin
 
Posts: 538
Joined: Fri May 06, 2005 6:15 pm
Location: Byron Bay

Thanks Mark

Postby JulienG » Wed Aug 24, 2005 1:45 am

That was a great post, and certainly gave me a few new ideas, mainly the concept of doing processing to the ms signal other then stereo width. I always knew you could do it, but it never actually struck me in that way. I suppose that it would be very useful with live recordings where you have a limited number of input channels.
JulienG
Regular Contributor
Regular Contributor
 
Posts: 263
Joined: Sat May 07, 2005 3:02 pm
Location: Sydney, Australia

Postby Henry » Fri Aug 26, 2005 1:54 am

Thx!

You da MAN!!
Henry
Registered User
Registered User
 
Posts: 127
Joined: Fri May 06, 2005 11:52 pm
Location: Sydney

Postby Peter Knight » Sun Sep 04, 2005 12:21 am

Hey has anyone tried compressing the M and S signal with different compressors, eg a Distressor on the Mid and an LA2 on the Side? To the mastering guys, would this seriously compromise your options in mastering?
Peter Knight
Registered User
Registered User
 
Posts: 119
Joined: Sat May 07, 2005 1:02 pm
Location: Perth


Return to The Turtlerock Forum

Who is online

Users browsing this forum: No registered users and 1 guest


cron