The Acoustic Guitar Forum

Go Back   The Acoustic Guitar Forum > General Acoustic Guitar and Amplification Discussion > RECORD

Reply
 
Thread Tools
  #1  
Old 11-29-2011, 07:20 AM
bbrown bbrown is offline
Registered User
 
Join Date: Sep 2010
Location: Forest, VA
Posts: 1,731
Default 'Mid-Side' with Audacity?

What do folks think of this technique?

Video is only about a minute long....................http://www.youtube.com/watch?v=nFiIJAIKN0o

Does it make sense?

--Bill
__________________
Collings D2HG - German Spruce/rosewood
Cervantes Crossover I - Spruce/rosewood
Voyage Air VA-OM6
Martin D16 GT


YouTube: http://www.YouTube.com/wfbrown1234
SoundCloud:
https://soundcloud.com/william-franc...gwindy-version
Reply With Quote
  #2  
Old 11-29-2011, 10:05 AM
RRuskin's Avatar
RRuskin RRuskin is offline
Charter Member
 
Join Date: May 2006
Location: Seattle WA
Posts: 1,953
Default

The whole procedure is much ado about almost nothing. He's manipulating the same signal several times. Whatever you get with this will not be stereo. True MS has discreet mid & side signals.
__________________
Rick Ruskin
Lion Dog Music - Seattle WA
Reply With Quote
  #3  
Old 11-29-2011, 11:26 AM
Fran Guidry Fran Guidry is offline
Registered User
 
Join Date: May 2002
Location: Walnut Creek, CA
Posts: 3,417
Default

It's somewhat common today to use the mathematics of mid-side recording to derive the mid and side signals for tweaking in post production. The video in question simply showed how to derive mid and side from a stereo pair, that is: side = l-r and mid = l+r .

Now that he has the two components, he can adjust their relative levels to widen or narrow the image width, apply dynamics processing, reverb, and/or EQ separately to the mid and side signals before decoding them back to conventional stereo.

The decoding can also be done in Audacity by reversing the steps shown in the video. Duplicate the side track and invert one copy. Duplicate the mid track. Sum (mix together and render) the non-inverted side with one of the mids to make the left channel. Sum the inverted side with one of the mids to create the right channel. Combine (pan hard left and right and render) the left and right channels to recreate the stereo file.

Back in analog days mid-side required transformers or active networks, both of which impact signal quality, so there was a cost to each decode and/or encode. With digital signal processing this process is lossless, requires no additional hardware, and so has become a lot more common.

Fran
Reply With Quote
  #4  
Old 11-29-2011, 11:40 AM
redavide redavide is offline
Registered User
 
Join Date: Dec 2008
Posts: 379
Default

I use the following method with Cubase without the need of any dedicated M/S encoder/decoder -- I'm certain you could do the same with Audacity . . .

1) Take a SDC mic and a LDC condenser in a figure 8 configuration;
2) Point the SDC at the guitar (or whatever you're recording) and find a sweet spot;
3) Put the LDC directly under it, getting the 2 mics as close to each other as possible, with the left and right nodes of the figure 8 perpendicular to the sound source;
4) After recording, duplicate the LDC track and open the mixer;
5) Link the 2 LDC tracks, pan one hard right and the other hard left;
6) Put one of the LDC figure 8 tracks out of phase;
7) Mix the linked LDC tracks with the SDC track in a ratio you like -- as you increase the LDCs, the stereo effect will become more pronounced.

Here are a couple examples of something I recorded with this method. It's Youtube, so the quality is Youtube-limited but I think you can clearly hear the stereo effect. . .

This one has quite a bit of the LDCs mixed in:
http://youtu.be/VGOZgui1YRw

This one has relatively little:
http://youtu.be/wt0H3GAdQ7I
__________________
redavide
http://www.youtube.com/user/redavide1
Reply With Quote
  #5  
Old 11-29-2011, 01:24 PM
bbrown bbrown is offline
Registered User
 
Join Date: Sep 2010
Location: Forest, VA
Posts: 1,731
Default

Thanks for the replies.

Fran, I'll have to study your reply and look up some some stuff to understand this. I always learn a lot from the exercise.

David, the first tune was removed. The second sounded great to me - I already am a subscriber to your channel so I got to hear this wonderful tune again. I love your music and the passion you bring to your playing.

So, it seems that the technique in the video with a single mic might be worthwhile?

I tried it but not sure I could appreciate much difference; but there are probably many other reasons for that, due to my equipment, etc.

--Bill
__________________
Collings D2HG - German Spruce/rosewood
Cervantes Crossover I - Spruce/rosewood
Voyage Air VA-OM6
Martin D16 GT


YouTube: http://www.YouTube.com/wfbrown1234
SoundCloud:
https://soundcloud.com/william-franc...gwindy-version
Reply With Quote
  #6  
Old 11-29-2011, 02:48 PM
jomaynor jomaynor is offline
Registered User
 
Join Date: Oct 2011
Location: Georgia
Posts: 1,193
Default

Quote:
Originally Posted by Fran Guidry View Post
It's somewhat common today to use the mathematics of mid-side recording to derive the mid and side signals for tweaking in post production. The video in question simply showed how to derive mid and side from a stereo pair, that is: side = l-r and mid = l+r .

Now that he has the two components, he can adjust their relative levels to widen or narrow the image width, apply dynamics processing, reverb, and/or EQ separately to the mid and side signals before decoding them back to conventional stereo.

The decoding can also be done in Audacity by reversing the steps shown in the video. Duplicate the side track and invert one copy. Duplicate the mid track. Sum (mix together and render) the non-inverted side with one of the mids to make the left channel. Sum the inverted side with one of the mids to create the right channel. Combine (pan hard left and right and render) the left and right channels to recreate the stereo file.

Back in analog days mid-side required transformers or active networks, both of which impact signal quality, so there was a cost to each decode and/or encode. With digital signal processing this process is lossless, requires no additional hardware, and so has become a lot more common.

Fran
+1

Well stated (and spelled).
Reply With Quote
  #7  
Old 11-29-2011, 03:02 PM
redavide redavide is offline
Registered User
 
Join Date: Dec 2008
Posts: 379
Default

Bill, I see the Led Zeppelin tune has been blocked in the U.S. by Youtube because of some copyright issue (not so, here in Italy).

Anyway, this tune has much more of the LDC sides mixed in.

Try this link . . .


http://s394.photobucket.com/albums/p...ronYrFinal.mp4
__________________
redavide
http://www.youtube.com/user/redavide1
Reply With Quote
  #8  
Old 11-29-2011, 03:07 PM
Fran Guidry Fran Guidry is offline
Registered User
 
Join Date: May 2002
Location: Walnut Creek, CA
Posts: 3,417
Default

Quote:
Originally Posted by bbrown View Post
Thanks for the replies.

Fran, I'll have to study your reply and look up some some stuff to understand this. I always learn a lot from the exercise.
...
So, it seems that the technique in the video with a single mic might be worthwhile?

I tried it but not sure I could appreciate much difference; but there are probably many other reasons for that, due to my equipment, etc.

--Bill
Perhaps I missed something - I understood the original tracks in the Audacity video to be a stereo pair - perhaps captured with an XY array or some other stereo source.

The whole exercise would be bordering on pointless with a single microphone.

Here are some fundamentals - stereo means "solid" - it refers to the ability to create a spatially coherent illusion on playback. Back in the 1920s a guy figured out how to do this with only two channels while various other folks experimented with three or more.

The guy's name was Alan Blumlein, and he invented mid-side at about the same time. http://en.wikipedia.org/wiki/Alan_Blumlein http://mixonline.com/TECnology-Hall-...umlein-090106/ and many other sources.

The key to understanding stereo is that it results from differences between the two channels. Time of arrival and/or intensity differences provide the cues our ear/brain uses to locate sound sources in space.

One mic = no differences = no location = mono, no matter how many channels are used to deliver the sound.

OK, let me back up a bit and suggest two semi-whacky uses for mid-side with a mono source. After encoding (creating mid and side as in the video or through use of a plugin) we might

- apply reverb to the side track only, then decode back to stereo. The thinking would be that we could add spaciousness without softening the impact of the mid signal.

- apply compression to the mid track only, punching up the signal while possibly reducing breathing artifacts because the side signal would represent the ambience and not be affected.

I have no idea if either of these tricks could ever actually be useful but the price is right.

Fran
Reply With Quote
  #9  
Old 11-30-2011, 02:17 AM
redavide redavide is offline
Registered User
 
Join Date: Dec 2008
Posts: 379
Default

Here's a video describing the exact mid-side method I outlined above -- using a recording of a choir and demonstrating how effective it is . . .

http://www.recording-microphones.co....d%20Side2.html
__________________
redavide
http://www.youtube.com/user/redavide1
Reply With Quote
  #10  
Old 11-30-2011, 06:56 AM
bbrown bbrown is offline
Registered User
 
Join Date: Sep 2010
Location: Forest, VA
Posts: 1,731
Default

Fran, I assumed the guy was using one mic, and hence the reason for my intitial question. I figured that he was just manipulating the tracks differently re. phase, to try to mimic a mid-side effect. It did seem somewhat dubious but I was curious what y'all thought about the technique.

David, I could indeed appreciate the difference. And the Cubase video did a good job of explaining the concept.

--Bill
__________________
Collings D2HG - German Spruce/rosewood
Cervantes Crossover I - Spruce/rosewood
Voyage Air VA-OM6
Martin D16 GT


YouTube: http://www.YouTube.com/wfbrown1234
SoundCloud:
https://soundcloud.com/william-franc...gwindy-version
Reply With Quote
  #11  
Old 11-30-2011, 10:37 AM
Doug Young's Avatar
Doug Young Doug Young is offline
Charter Member
 
Join Date: Apr 2005
Location: Mountain View, CA
Posts: 6,680
Default

It looks to me as if he started with a real stereo signal. If the 2 tracks were identical (one mic), then when he inverts the phase and sums to mono, it would collapse to null. Instead it collapses to something that looks like an expected "side" signal. All he's doing here is showing you a manual way of doing what something like the Voxengo plugin would do if you "encode" a stereo signal to MS. This is a pretty common procedure, and there are lots of plugins out there that convert to MS without you even knowing, do some processing in that mode, then convert back. It's just a form of signal manipulation that can be useful at times. Nothing to do with micing techniques.
__________________
Doug Young
----------------
Music on Pandora
You Tube Channel
website: http://www.dougyoungguitar.com
Fingerstyle Christmas Tunes: A DADGAD Christmas
CDs: Closing Time, Laurel Mill
Pickup tests: http://www.dougyoungguitar.com/pickuptests/
Reply With Quote
  #12  
Old 11-30-2011, 11:42 AM
Fran Guidry Fran Guidry is offline
Registered User
 
Join Date: May 2002
Location: Walnut Creek, CA
Posts: 3,417
Default

Quote:
Originally Posted by Doug Young View Post
It looks to me as if he started with a real stereo signal. If the 2 tracks were identical (one mic), then when he inverts the phase and sums to mono, it would collapse to null...
Good catch, Doug.

Fran
Reply With Quote
  #13  
Old 11-30-2011, 11:34 PM
redavide redavide is offline
Registered User
 
Join Date: Dec 2008
Posts: 379
Default

Bill, maybe you noticed that in the Youtube video, the author refers us to another site for more information on his process (you can see it by clicking on "show more" on Youtube) . . .

For what it's worth, this is it:

http://www.drumnbass.be/forum/thread.php?threadid=17006
__________________
redavide
http://www.youtube.com/user/redavide1
Reply With Quote
  #14  
Old 12-01-2011, 06:32 AM
bbrown bbrown is offline
Registered User
 
Join Date: Sep 2010
Location: Forest, VA
Posts: 1,731
Default

Thanks David.

This short explanation from that thread was helpful: http://www.brainworx-music.de/en/wha...g3vs8rmpvoig84

This is all way beyond anything I'll ever use (I'm using a Zoom H4N!), but I find it interesting nevertheless.
__________________
Collings D2HG - German Spruce/rosewood
Cervantes Crossover I - Spruce/rosewood
Voyage Air VA-OM6
Martin D16 GT


YouTube: http://www.YouTube.com/wfbrown1234
SoundCloud:
https://soundcloud.com/william-franc...gwindy-version
Reply With Quote
  #15  
Old 12-01-2011, 10:57 AM
Fran Guidry Fran Guidry is offline
Registered User
 
Join Date: May 2002
Location: Walnut Creek, CA
Posts: 3,417
Default

Quote:
Originally Posted by bbrown View Post
Thanks David.

This short explanation from that thread was helpful: http://www.brainworx-music.de/en/wha...g3vs8rmpvoig84

This is all way beyond anything I'll ever use (I'm using a Zoom H4N!), but I find it interesting nevertheless.
Did you know that your H4n has a built-in mid-side decoder?

http://www.homebrewedmusic.com/2009/...-the-zoom-h4n/

Fran
Reply With Quote
Reply

  The Acoustic Guitar Forum > General Acoustic Guitar and Amplification Discussion > RECORD

Thread Tools



All times are GMT -6. The time now is 11:26 AM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Copyright ©2000 - 2018, The Acoustic Guitar Forum
vB Ad Management by =RedTyger=