M/S stereo. How can I know which mic is the bidirectional from only the raw files?

[J-Popp]

Hi,

I just got a project that I'm about to mix for a friend of mine. It seems like one of the microphones on both the OH and AMB pair are phase inverted, which makes me believe that they are recorded in an M/S configuration. Can I be sure of that? And...how can I know which microphone is the bidirectional?

[Grantsos]

I suppose, run them through a matrix and see if they "work" properly... If you're running software most plugs will enable you to swap the input setups.

If you then put an inverter on the suspected S signal, the stereo image should swap perfectly.

The M signal should sound tighter and less roomy than the S.

Just some initial thoughts.

[Evie McCreevie]

If he's a friend of yours, ask him...

Otherwise, take one of the pair, split it to two channels, pan hard left and right, invert one channel. Bring the other of the pair up the middle. Balance these three channels, keeping the split pair even.

If that sounds ropey, swap the tracks, and give them the same treatment.

You could be all day faffing about... If he's a friend of yours, ask him. And tell him it should have been marked on the track names / track sheet!

[Aural Reject]

I may be missing something here, but one channel wouldn't be phase inverted....but the signal would appear off axis.....is that what you mean?

[Evie McCreevie]

I may be missing something here, but one channel wouldn't be phase inverted....but the signal would appear off axis.....is that what you mean?

No, I'm a bit tired, but pretty sure my suggestion is correct.

The signal from the figure of eight mic needs to be split, panned hard, one side inverted.

As pointed out, this signal from the fig-of-eight in its raw mono form, should sound 'roomier' than the other mic (cardioid).

[Aural Reject]

I know how M-S works

But the OP suggests that one mic is phase inverted....which isn't the same thing as a mic sounding off axis...unless this holiday bottle of red has got to me....which isn't beyond the realms of possibility.

[Hugh Robjohns]

It seems like one of the microphones on both the OH and AMB pair are phase inverted, which makes me believe that they are recorded in an M/S configuration.

Okay... first important lesson to share with your friend, and to remember yourself: It is critical to log what each channel is suppose to be carrying and, in the case of stereo or surround tracks, the channel allocation and format. If a project isn't properly documented, send it back!

Next, as AR says, if a pair of tracks is MS encoded, then one channel (MID) should sound tight, on axis and focussed -- a good usable mono sound -- and the other (SIDE) should sound very off-axis, ambient, reverberant possibly and considerably more distant.

If you listen to MS tracks as if they were LR, then you could get an out-of-phase kind of sound depending on where the sound source(s) are relative to the mics. But the best way to identify what is going on is to listen to each channel separately.

The other possibility for an out-of-phase sound from an MS source is that they have already been decoded to LR, but with way too much S. This is actually surprisingly common. A lot of people look at the direct outputs from an MS pair, notice that the Side signal is typically much lower than the Mid signal, and crank the gain to improve the signal-noise ratio. If you then decode these thinking they are proper MS signals, you have far too much Side signal and you end up with ultrawide stereo with a hole in the middle and a nasty phasey sound!

Once you have worked out which track is Mid and which is Side, you can then decode through a matrix to produce normal left-right. This should provide a clear, stable stereo image with a reasonable width. Adjust the realtice levels of Mid and Side to tatse. If you invert the polarity of the Side mic the whole LR image should swap over clearly and become RL.

Hope that helps

Hugh