M/S Mastering

[::XeS::]

Anyone have used the M/S mastering tecnique?
This way to proceed consists in the division of the stereo file in 2 channels: not Left and Right but MID and SIDE.
With mid and side channels you can set separately the mid instruments (vocals,kick,snare, bass...) and the side instruments (guitars, hats...)..
What do you think?

 

[LSD-Studio]

i'm using that only in case i gotta fix things, like to quiet vox etc) from outside recordings.

 

[chadsxe]

Anyone have used the M/S mastering tecnique?
This way to proceed consists in the division of the stereo file in 2 channels: not Left and Right but MID and SIDE.
With mid and side channels you can set separately the mid instruments (vocals,kick,snare, bass...) and the side instruments (guitars, hats...)..
What do you think?

Its a tool for when your painted in a corner. I would never would wish it upon anyone to have to use it.

 

[Mindmare]

Though I haven't used this method much, I find at least the idea rather appealing. Todays masterings (specially in the metalgenre) are so darn loud there should be law against it, and it is really difficult matching those RMS levels without ruining the music. The M/S approach along with normal L/R compression, offers a way to get more distinct keysignals. Instead of compressing "all the way" in normal stereo, one could compress lesser (say half as much) using each method (M/S first, then L/R) and thereby (hopefully) get a more meaningful end result.

What I'd like to hear is why one should not process in M/S mode?

 

[flhctroll]

Do you have to use some sort of channell mixer on multiple wavs in the session to get this or is there a plug to get the mid/side?

 

[::XeS::]

there are waves plugins that made a sort of M/S but I read they don't work very well...but you can do the same thing manually.
You can get M and S from a stereo file with those equations:

M = 0.5 * ( L + R )

S = 0.5 * ( L - R )

therefore, M is the mix of L and R, 6dB lowered and S is the mix of L and R (R with inverted phase), 6dB lowered.

When you have setted all your things with M/S, you can obtain the L/R image with those equations:

L = M + S

R = M - S

L will be the mix of elaborated M and S
R will be the mix of elaborated M and S (S with inverted phase).

With you daw, you should create 2 tracks (one for the Left channel and one for the Right channel). Now, you have to copy those 2 tracks in 2 others tracks.
1 -> L
2 -> R
3 -> L
4 -> R

Now you can rename the tracks this way:
MIX L1
MIX R1
MIX L2
MIX R2

you have to set the level of the 4 tracks at -6dB and MIX R2 have to be phase inverted.

Made this, you have to create 4 mono groups:
GRP M
GRP S
GRP M2
GRP S2

Now:
---> = send
MIX L1 and MIX R1 --> GRP M
MIX L2 and MIX R2 --> GRP S

GRP M --> GRP M2
GRP S --> GRP S2
with send level setted to 0dB

GRP S2 have to be phase inverted.

Now,if you play the song you listen all normal like before.
If you wanna set the mid channel, insert the plugins in GRP M.
If you wanna set the left channel, insert the plugins in GRP L.
With this approach, you can set M and L and listen directly how it sounds in stereo.

Sorry for the long post but it's an italian article and I've synthetized it for you (hope you undertand something )
I post the link of the article for the pictures of the various steps in cubase.

Tecniche & trucchi per la produzione audio

 

[Wolfeman]

Thanks for that XeS.... I'll probablly never have to use it but it's good to know just in case.

 

[Mindmare]

I use this set of m-s encoding/decoding plugins:

MStools

One converts LR to MS, another converts MS to LR. Very simple to use.

 

[AStacy2]

Use it all the time, great to do before a track hits radio to track down vocal resonances and such. A lot of people don't check this and that just means that the song isn't going to translate as well as it could.