MS Matrix in Pro Tools OS9

[paule]

Hi all,
someone interested in my experiences about MS Matrix on OS9?
I tried to setup the first time a MS Matrix in order to master a stereo track independently as mid and side information.
Because I would like to let all of you be a part of my learning process or maybe someone can give me an explanation about the following:

Okay, I have tried the MS Matrix setup with some hints of a youtube video (-> link below). But I was afraid all the time that the sound in MS could be different than to original source stereo file.

Therefore I run the following test: I setup the MS-matrix and tried to "erase" my created MS-Matrix with a (out-of-phase) original source. Then I could be sure that I have reproduced the stereo file exactly in mid-sid-matrix and have no loss in quality.
After some hours testing I have achieved the cancelation – but with some tricks:

The usual way is to instantiate an waves S1 MS Matrix plugin in the stereo source. The strange thing was that I have only achieved a complete cancelation when I dropped the "sides"-information by -2.5dB. The "mid"-information was 0dB, but the sides needed to be -2.5dB less. Can someone explain this to me? It seems to me that the Mid-Information (when instantiating the Waves MS Matrix) is +2.5dB higher than the sides!!!!
But why?
In order to get the cancelation done (if you want to test it in Pro Tools) is to drop the source-out-of-phase-stereo-file also -2.5dB. Then flip the phase (and check the delays because of the used plugins) and then you get total cancelation. But why this -2.5dB drop in the sides (or in other words +2.5dB more on the mid information)?

Maybe someone else find this helpful too and maybe some genius can tell me why?
Best, paule.

Waves MS-Matrix (link):
https://www.youtube.com/watch?v=ctNJxPMVWTo

[Protools5LEGuy]

The usual way is to instantiate an waves S1 MS Matrix plugin in the stereo source. The strange thing was that I have only achieved a complete cancellation when I dropped the "sides"-information by -2.5dB. The "mid"-information was 0dB, but the sides needed to be -2.5dB less. Can someone explain this to me? It seems to me that the Mid-Information (when instantiating the Waves MS Matrix) is +2.5dB higher than the sides!!!!
But why?
In order to get the cancellation done (if you want to test it in Pro Tools) is to drop the source-out-of-phase-stereo-file also -2.5dB. Then flip the phase (and check the delays because of the used plugins) and then you get total cancellation. But why this -2.5dB drop in the sides (or in other words +2.5dB more on the mid information)?

You are decoding stereo to M + S.  When you add 2 side tracks (to the mono-center) you are giving 2 (3) signals to become the stereo part, giving more "side" signal that what is needed to the cancellation. That is why the -2.5 dB

Let's check what is "pan law" https://en.wikipedia.org/wiki/Pan_law

[Protools5LEGuy]

http://duc.avid.com/showpost.php?p=1695396&postcount=15

Re: PT 9 Pan Depth
I think SSL is actually -4.5dB pan law, but I've been wrong before..

TDM stereo mixer pan law is -2.5dB. That's why if you use both the stereo mixer and the surround mixer you are forced into -2.5dB for stereo pans (the exception being when you assign a stereo pan to two subpaths of a surround mixer, in which case it's -3dB....got that? <G>)

The surround mixer WAS capable of only a -3dB pan law, in surround or stereo configs. In order to make an adjustable panlaw in Pro Tools they had to just use the surround mixer. So that is why new installations do not install the stereo mixer anymore. If you are "stuck" at -2.5dB and you want to use another pan law you have to quit Pro Tools, move the stereo mixer into the "unused plug-ins" folder, and relaunch.

There are potential compatibility issues with pre-Pro Tools 9 systems, so if you know your session is moving to an older system I would recommend using the -2.5dB pan law. But you have to be careful - for the "new session" default in Pro Tools 9 is -3dB.

Note that older sessions will correctly open as -2.5dB in PT9, even if you are running in the "surround mixer only" config.

What's the compatibility issue? Well, if you create a stereo session with -4.5dB pan law, and move it to a PT8 system, anything that was panned hard-center in the stereo mixer will now be 2dB louder than it was, because PT8 uses a -2.5dB pan law...

I know this is confusing for many....so maybe this will be helpful - I think Pan Law is not widely understood. I hope I'm not being pedantic...but also, please feel free to correct me if I've blundered...

The difference between a "balance control" (blend) and a "pan pot" (panoramic potentiometer) is that the center position has both channels at 100% gain with a "balance control" and a pan pot is, well...there are several choices, dependent on its "pan law."

With a balance control, moving the knob to the left or right attenuates one channel while leaving the other alone. There is no gain increase in either channel by moving the control off-center.

A Panpot continuously varies the gains on both channels across its full travel. You can never have full gain on both channels at any position. If you pan a signal hard left or right, it outputs that signal at unity gain out of one channel at a time.

-3dB stereo pan law means that a signal panned hard left (resulting in 0dB out the left only) will drop 3dB when panned in the center. This is the industry standard, if such a thing exists <G>. It's a compromise between no "panoramic" perception (like a balance control, which is essentially a "0dB" pan law) and "mono compatible" (-6dB pan law).

The purpose of pan law determines two things, from a subjective view point -
1) Determines how dramatic a moving pan is.
• For instance, a pan law of -4.5 dB will sound more dramatic than a -3dB pan law when you are moving a signal around because it will be lower in volume when it passes through the center - or is louder when it reaches the left or right extreme (compared to when it is panned center). Just try panning a mono signal around with the different pan laws in PT9 and you'll see.
2) Determines the level of energy a center - panned signal has after it's panned to the left or right. To a certain degree it helps determine your mono compatibility when creating a "quickie" stereo version of your mono mix.
• Remember that when you mono-ize a stereo mix, any signals that are panned far left or right will drop by 6dB in comparison to the center (or technically, the center comes UP 6dB, but most mono buttons compensate for this).
• If you sum the two channels, then move a center piece of info to the hard left it will lose -
• 3.5dB (if your pan law is 2.5dB)
• 3dB (if your pan law is 3dB
• 1.5dB (if your pan law is 4.5dB
• 0dB (if your pan law is 6dB
• The thing is, using the more mono compatible pan laws isn't necessarily going to make your stereo -mono mix compatibility any better. If you pan it hard left it might sound too loud if you're using a 4.5 or 6 dB pan law, so you'll bring it down anyway. That's why 3dB was chosen for the original industry standard.

-Danny

[paule]

Hi Protools5LEGuy,
thanks for your detailed informations. I am always impressed where other people find the information I was looking for. I suspected the "pan law" as the reason for that but I haven' had these fantastic informations. Thanks a lot!!!!
But maybe my question and your answer will lead someday someone else to this topic.

Thanks a lot,
paule.