The No.1 Website for Pro Audio
Let's talk M/S...
Old 26th March 2006
  #1
Gear Maniac
 
tunasafedolphin's Avatar
 

Let's talk M/S...

...and I'm not refering to multiple sclerosis.

Info up front: I'm not a ME, but a professional mixer and tracking engineer.

The other day I had a client bring by a mix that was in pretty rough shape, asking me to do something with it. Clueless and just opting to throw logic out the window, I just used my ears. I ended up encoding the mix M/S, and high-passing the side at like 90 and adding some air, compressing the center and adding 2 DBs, then back to stereo for final touches...

The mix certainly improved, but it seems to me such treatment is drastic. Correct me if I'm wrong, but typically this kind of thing is a bad idea right? How often are you guys processing the M/S differently and why?

Thanks,
-C
Old 26th March 2006
  #2
Lives for gear
 
5down1up's Avatar
 

M/S rocks ... imho
Old 26th March 2006 | Show parent
  #3
Lives for gear
 

Quote:
Originally Posted by 5down1up
M/S rocks ... imho
+1

It's great for mixing too.
Old 26th March 2006 | Show parent
  #4
Gear Maniac
 

little goes a long way - headphone check won't hurt.
Old 26th March 2006 | Show parent
  #5
m\s can sometimes be a cure and sometimes destructive , a great technique for recreating the stereo image for too "monoish" mixes and i often find it great for mastering accapela \ choirs , but when used gently it can be effective for many sources , i have a custom stereo Pultec eq that sounds amazing for the 2-buss and i have an M\S encoder built in , and i often love the effect of adding air on the sides and some bass to the center , but it's a very tricky thing because you are rebalancing the mix
Old 26th March 2006
  #6
Gear Maniac
 
borism's Avatar
 

how to M/S in protools?

Would any of you guys care to explain step by step, how to apply/work M/S in a Protools environment with a stereo file, a couple of months ago an article in Electronic Musician tried to cover this, but my attempt to apply technique of this article explanation was frustating.
Could you iluminate me?
Thank you
Old 26th March 2006
  #7
Gear Maniac
 
borism's Avatar
 

bump? anyone?
Old 26th March 2006 | Show parent
  #8
Lives for gear
 
Harley-OIART's Avatar
 

Bumpy Bumpy
Old 26th March 2006 | Show parent
  #9
Lives for gear
 
djui5's Avatar
 

Quote:
Originally Posted by borism
Would any of you guys care to explain step by step, how to apply/work M/S in a Protools environment with a stereo file, a couple of months ago an article in Electronic Musician tried to cover this, but my attempt to apply technique of this article explanation was frustating.
Could you iluminate me?
Thank you


You need to bounce your stereo buss to a mono file, bring in this mono file and pan it center, and flip the phase of one side of your stereo buss. You now have a mono center, and the really wide stereo sides. Process accordingly.
Old 27th March 2006 | Show parent
  #10
Gear Addict
 

hi! this thread is nice!
will this technique be 100% mono-compatible ?
are they any m/s - Eqs as software ?

ciao

chester
Old 27th March 2006 | Show parent
  #11
Lives for gear
 
djui5's Avatar
 

Quote:
Originally Posted by chester
hi! this thread is nice!
will this technique be 100% mono-compatible ?

Yeah, that's what the mid channel is for. If you fold down the sides they'll pretty much (not entirely) cancel each other out.
Old 17th June 2006 | Show parent
  #12
Gear Addict
 
johnlink's Avatar
 

Quote:
Originally Posted by djui5
You need to bounce your stereo buss to a mono file, bring in this mono file and pan it center, and flip the phase of one side of your stereo buss. You now have a mono center, and the really wide stereo sides. Process accordingly.
When I follow the directions given above, the bass sounds very weak no matter how I adjust the levels of the left, right, and mono tracks. It sounds as though I have a polarity reversal problem, which is not surprising given that I have reversed the polarity of one side of the mix bus. But maybe I misunderstand the directions. Here's what I did, spelled out in detail:

1) Place the left track of the mix in channel 1 and the right track in channel 2.
2) Pan channels 1 and 2 to the center and bounce.
3) Place in channel 3 the left (or right) track of the file created in step 2 and pan it to the center.
4) Pan channel 1 hard left and channel 2 hard right.
5) Reverse the polarity of one side of the mix bus.

Perhaps step 5 should be replaced with step 5':

5') Reverse the polarity of channel 3.

Now I am able to adjust the faders to create the original stereo mix, the mono mix, and everything in between, although I don't see any way to shift the balance between mono and stereo without changing the summed level.

John Link
Old 17th June 2006 | Show parent
  #13
Lives for gear
 
masteringhouse's Avatar
 

Verified Member
Quote:
Originally Posted by borism
Would any of you guys care to explain step by step, how to apply/work M/S in a Protools environment with a stereo file, a couple of months ago an article in Electronic Musician tried to cover this, but my attempt to apply technique of this article explanation was frustating.
Could you iluminate me?
Thank you
This may help:

http://www.masteringhouse.com/master...s/midside.html
Old 17th June 2006 | Show parent
  #14
Lives for gear
 
djui5's Avatar
 

Quote:
Originally Posted by johnlink
When I follow the directions given above, the bass sounds very weak no matter how I adjust the levels of the left, right, and mono tracks. It sounds as though I have a polarity reversal problem, which is not surprising given that I have reversed the polarity of one side of the mix bus. But maybe I misunderstand the directions. Here's what I did, spelled out in detail:

1) Place the left track of the mix in channel 1 and the right track in channel 2.
2) Pan channels 1 and 2 to the center and bounce.
3) Place in channel 3 the left (or right) track of the file created in step 2 and pan it to the center.
4) Pan channel 1 hard left and channel 2 hard right.
5) Reverse the polarity of one side of the mix bus.

Perhaps step 5 should be replaced with step 5':

5') Reverse the polarity of channel 3.

Now I am able to adjust the faders to create the original stereo mix, the mono mix, and everything in between, although I don't see any way to shift the balance between mono and stereo without changing the summed level.

John Link


2) This needs to be bounced to MONO, not stereo (I'm assuming you bounced it to stereo because you said take the left or right side of the file created and pan it center). An easier way to do this (are you using pro-tools?) is to set up a new mono audio track. Set it's input to buss 1. Set the original audio trackes left and right output to buss 1. Record the audio onto the new track. Then go back to the original audio tracks and set their output to the normal 1/2 and pan accordingly, and flip the phase of either the left or right side. Leave the new track panned center (should be titled "Center"). You shouldn't be doing anything to the master fader at all.

Now, to adjust the width of the music overall, you'll adjust the volume of the left/right tracks and leave the center alone.
Old 17th June 2006 | Show parent
  #15
Gear Addict
 
johnlink's Avatar
 

Quote:
Originally Posted by djui5
2) This needs to be bounced to MONO, not stereo (I'm assuming you bounced it to stereo because you said take the left or right side of the file created and pan it center).
The result was mono, because step 2 says to pan channels 1 and 2 to the center.

Quote:
An easier way to do this (are you using pro-tools?)
Yes.

Quote:
is to set up a new mono audio track. Set it's input to buss 1. Set the original audio trackes left and right output to buss 1. Record the audio onto the new track. Then go back to the original audio tracks and set their output to the normal 1/2 and pan accordingly, and flip the phase of either the left or right side.
Flip one of the original tracks, or in the mix bus?

Quote:
Leave the new track panned center (should be titled "Center"). You shouldn't be doing anything to the master fader at all.

Now, to adjust the width of the music overall, you'll adjust the volume of the left/right tracks and leave the center alone.
When I do as you say and reverse polarity of the original right track I have the same situation I reported in my first post. If, instead, I reverse polarity of the right side of the mix bus, again I have the same situation I reported in my first post.

John Link
Old 17th June 2006 | Show parent
  #16
Lives for gear
 
djui5's Avatar
 

Flip one of the audio tracks. Leave the master fader out of this

regarding the shift in level, you have to remember that if you take something away, you need to replace it with something else. So if you turn down the sides 2db, you should turn up the center 2db or whatever to makeup for the loss. It turns into a tricky balance.

Is your bass still funny when you do this method the right way, leaving the master buss out of this? If so, do you have delay compensation turned on?


BTW, just an aside, I don't work with a master fader. I use an aux track for those duties Try it, you might like it better.
Old 17th June 2006 | Show parent
  #17
Gear Addict
 
johnlink's Avatar
 

Quote:
Originally Posted by masteringhouse
Yes, now I've got it working. If I set both the M and S faders to 0.0 I get the original stereo mix. If I set the M fader to 0.0 and the S fader to -inf I get the mono mix. If I set the M fader to 0.0 and vary the S fader between 0.0 and -inf the overall level stays the same. If I set the M fader to 0.0 and raise the S fader above 0.0 I get the original stereo mix with added stereo component, and the the overall level goes up.

Very cool that it is all done with routing, with no need to bounce first.

John Link
Old 17th June 2006 | Show parent
  #18
Mastering
 

Quote:
Originally Posted by johnlink
Yes, now I've got it working. If I set both the M and S faders to 0.0 I get the original stereo mix. If I set the M fader to 0.0 and the S fader to -inf I get the mono mix. If I set the M fader to 0.0 and vary the S fader between 0.0 and -inf the overall level stays the same. If I set the M fader to 0.0 and raise the S fader above 0.0 I get the original stereo mix with added stereo component, and the the overall level goes up.

Very cool that it is all done with routing, with no need to bounce first.

John Link
And now that you've got that you can play with EQ and compression on the M and S channels.

BUT HOLD THE PRESSES! HERE'S THE SIMPLEST SOLUTION YOU CAN FIND:

Try this width control that only requires 2 stereo faders! It's far less complicated but it is exactly equivalent to the M and S equations. Surprised that others have not found this solution, which is SO SIMPLE:

Stereo Source goes into a stereo fader.

The same Stereo source goes into a second stereo fader REVERSED Pan and Inverted polarity. Add the second fader in and the width increases. Simultaneously the level goes down, so raise both faders in a group until you restore the level. Using the modifier keys to override, plus two stereo faders in Pro tools and you can do this with grouping.

Miracle, eh? This is a simultaneous ENCODE and DECODE without having to go through all that multiple and messy bussing!

BK
Old 17th June 2006 | Show parent
  #19
Lives for gear
 
masteringhouse's Avatar
 

Verified Member
Quote:
Originally Posted by bob katz
Try this width control that only requires 2 stereo faders! It's far less complicated but it is exactly equivalent to the M and S equations. Surprised that others have not found this solution, which is SO SIMPLE:

Stereo Source goes into a stereo fader...

Bob,

I can understand the usefulness of this for stereo width enhancement, but how would you process the M and S components separately with this technique?

Is processing the first stereo pair really the same as processing a summed mono track?
Old 18th June 2006 | Show parent
  #20
Lives for gear
 
Matti's Avatar
VST: Voxengo MSED (free) encoder / decoder for example.
Matti
Old 18th June 2006 | Show parent
  #21
Gear Addict
 
johnlink's Avatar
 

I've posted three versions of my song "Goodbye": https://www.gearslutz.com/board/showthrea...449#post764449

The three versions are 1) the original mix, 2) the original mix with M/S technique applied in order to expand the stereo field, and 3) a version that alternates between versions 1 and 2 every few seconds. Let me know what you think.

John Link
Old 18th June 2006 | Show parent
  #22
Gear Addict
 
johnlink's Avatar
 

Quote:
Originally Posted by johnlink
If I set the M fader to 0.0 and vary the S fader between 0.0 and -inf the overall level stays the same.
That appeared to be the case because the source tracks (Mariah Carey's "Vision of Love) were pegged at 0.0. With my own "Goodbye", mentioned in my previous post, there is plenty of dynamic range, and I see that the overall level changes as I move the S fader between 0.0 and -inf.

John Link
Old 18th June 2006 | Show parent
  #23
Gear Maniac
 
Tubefreak's Avatar
 

If you had to do some surgical EQ during mastering, would you do it on normal stereo track or on one of the M/S tracks?

Maarten
Old 18th June 2006
  #24
Lives for gear
 

Quote:
Originally Posted by tunasafedolphin
...and I'm not refering to multiple sclerosis.

Info up front: I'm not a ME, but a professional mixer and tracking engineer.

The other day I had a client bring by a mix that was in pretty rough shape, asking me to do something with it. Clueless and just opting to throw logic out the window, I just used my ears. I ended up encoding the mix M/S, and high-passing the side at like 90 and adding some air, compressing the center and adding 2 DBs, then back to stereo for final touches...

The mix certainly improved, but it seems to me such treatment is drastic. Correct me if I'm wrong, but typically this kind of thing is a bad idea right? How often are you guys processing the M/S differently and why?

Thanks,
-C
I think that a mastering engineer is very limited with only a stereo track. Stems in combination with M/S can help, but on the other hand it's also a good thing that a mastering engineer can't do too much since the feeling and colors of the original mix should be left after mastering. I guess there are different opinions about this. But the problem is when the mix in a good/flat sounding environment sounds much different than in the mixing environment, in other words the ideas behind the mix don't translate to the real world, it might be due to a combination of things, for instance mix monitoring environment, badly consumed sound field etc. I think that's when many mastering engineers today would send it back for mixing, especially when the client demands a loud mastering.

So if you get a mix that is already in good shape, in other words each instrument overall has its own dominant frequency range in the center and the sides, the relative track level balance is right and the overall mix is translating well in the master monitoring environment, I think M/S is at least more transparent than applying things only on a stereo track. So I think M/S is a minimum requirement in terms of isolation. In these cases the point is of course to have the center and the sides isolated so that you CAN target these frequencies differently. I guess a typical problem is a muddy center: the vocals are not clear enough, the kick drum and snare might be much too loud and the bass guitar is rather undefined. By then applying for instance a multiband compressor setup in the right way in combination with an EQ effect, you can adjust the different tracks in the center without effecting the instruments on the sides. So this is possible without compromising the mix too much if the mix is already in pretty good shape.

The problem is that all these changes will affect the overall mix anyway since some instruments occupy a pretty broad frequency range and are sometimes also occupying a lot of space in the stereo field. So it's a matter of how much you CAN optimize. I think this is a tricky work especially with a rather unbalanced quiet mix that you need to gain, but in these cases M/S is becoming really important.
Old 18th June 2006 | Show parent
  #25
Mastering
 

Quote:
Originally Posted by masteringhouse
Bob,

I can understand the usefulness of this for stereo width enhancement, but how would you process the M and S components separately with this technique?

Is processing the first stereo pair really the same as processing a summed mono track?
Good catch, Tom! No, it's not the same! So if you need to separately process the M and the S components, my 2 fader method goes to pot. I'd have to sit down and do the equations, but separately processing the two faders is going to make a strangely variable encode/decode cycle, it would probably create weird width changes with dynamics.

BK
Old 18th June 2006 | Show parent
  #26
Lives for gear
 
masteringhouse's Avatar
 

Verified Member
Quote:
Originally Posted by bob katz
Good catch, Tom! No, it's not the same! So if you need to separately process the M and the S components, my 2 fader method goes to pot. I'd have to sit down and do the equations, but separately processing the two faders is going to make a strangely variable encode/decode cycle, it would probably create weird width changes with dynamics.

BK
Thanks Bob, my sanity has been spared.

Your approach may be a different tool/technique for a different purpose however.

I was thinking of an extreme case (reality always seems to be somewhere between the extremes):

Say that we had a track where both channels were 180 degrees out of phase at the start and we wanted to compress a bass track in the middle. When the track was summed to mono the channels would null and there would be nothing to compress. By compressing it in stereo (linked) we would have the ability to compress it, though it would be based on whatever is loudest between the two.

I suppose one could switch the compressor to the S component in a regular M/S configuration (would be like processing a mono track again), but it shows how your technique might be useful over traditional M/S. I wonder how inserting delays might sound using the two techniques? Hmm, mad scientist back off to the lab ...
Old 19th June 2006 | Show parent
  #27
Gear Addict
 
bonne's Avatar
Quote:
Originally Posted by bob katz
Miracle, eh? This is a simultaneous ENCODE and DECODE without having to go through all that multiple and messy bussing!

BK
Nice one, Bob. I'll be checking that out.

Here's a simpler M/S encode/decode without the elaborate bussing that Tom suggests on his website:

You need two stereo source tracks, three group tracks and a stereo out track (master buss).


1) Load your source to source track 1. This will be used for reference and to check the transparency of your M/S encode/decode network (see below).

2) For MS encoding load your source to source track 2 and insert the Waves S1 MS Matrix plug on it. Create two sends from it, both prefader and both at unity gain. Pull the fader on the source track all the way down.

3) One send is routed to the first group track, called Mid, and the track panned all the way left. The other send is routed to the second group track, called Side, and panned all the way right.

4) The outputs from group tracks Mid and Side are routed to a third group track called L/R Stereo. The Waves S1 Imager is used as an insert effect on that track for decoding back to stereo. Input on the S1 Imager is set to M/S.

5) Output from group track L/R Stereo is routed to the Stereo Out (master) track.


This is a simple way of doing it ITB. To check for any routing errors reverse the polarity on source track 1 and it should null the sound coming through the M/S setup. You're done and ready to insert eq's, compressors or other effects on your Mid and Side tracks.

Good luck!!

Jørn
Old 19th June 2006 | Show parent
  #28
Mastering
 

[QUOTE=bonne]Nice one, Bob. I'll be checking that out.

Here's a simpler M/S encode/decode without the elaborate bussing that Tom suggests on his website:

But why get that complex? If you're going to use the plugin, you might as well do everything ON THE STRIP without any bussing. The Waves S1 can be configured as a stereo to MS converter and vice versa. So you can then insert equalizers or compressors on the rest of the strip and then another S1 to convert back. Or, for less CPU power, you can just use the Waves MS plugin and manipulate the two channel gains of any equalizer or compressor plugin that follows. Be aware that Waves MS plugin loses 3 (or is it 6?) dB for safety, which you can make up in the following processor.

Of course, you can use a buss if you wish to MS process a stereo group, but you can still keep all the MS processing on a single buss strip using the Waves MS plugin.

BK
Old 19th June 2006 | Show parent
  #29
Gear Addict
 
bonne's Avatar
But why get that complex?

Because I couldn't find a software compressor that would process left or right (M or S) separately like you can with the Waves eq's. Hadn't thought of manipulating the two channel gains as you suggest.

Be aware that Waves MS plugin loses 3 (or is it 6?) dB for safety, which you can make up in the following processor.

Not a problem in the setup I posted. Polarity reversed reference source track nulls the unprocessed sound coming out of the MS encode/decode. The Waves plugs seem to be transparent.

Cheers

Jørn
Old 19th June 2006 | Show parent
  #30
Lives for gear
 
masteringhouse's Avatar
 

Verified Member
Quote:
Originally Posted by bonne
Here's a simpler M/S encode/decode without the elaborate bussing that Tom suggests on his website:

You need one stereo source track, three group tracks and a stereo out track (master buss).

1) Load your source to the source track and insert the Waves S1 MS Matrix plug on it. Create two sends from it, both at unity gain. Pull the fader on the source track all the way down...


Jørn
Ah that's cheating Jørn, my approach is free!

Just create a PT template setup this way and you only have to do it once.
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 36 views: 20492
Avatar for Andrew Klimek
Andrew Klimek 14th August 2014
replies: 68 views: 17043
Avatar for KJandKT
KJandKT 10th July 2010
replies: 1 views: 1844
Avatar for fragletrollet
fragletrollet 25th October 2013
replies: 89 views: 14276
Avatar for bgood
bgood 15th May 2017
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
🖨️ Show Printable Version
✉️ Email this Page
🔍 Search thread
🎙️ View mentioned gear
Forum Jump
Forum Jump