Topic: Side is reversed on m/s setup

[speedo]

I have an m/s rig with a special 2xlr to 5 pin cable.

I have the figure 8 mic on axis stage left. However, the signal is reversed. Left is right and right is left. Does it matter which way the figure 8 mic is facing? Or is the polarity in my cable perhaps reversed?

Thanks.

[DSatz]

I get confused by the terms "stage right" and "stage left" so let me just say, the 0-degree axis of the "S" microphone should point to the left from the audience's point of view as they face center stage.

If that's what you did, and you still got reversed L/R channels, then it's possible that your "S" microphone signals are getting their polarity flipped on the way to the matrix input. Do you have a meter that you can check your cables with? We can go over which pins should be connected to which.

At least there's no other harm done--just swap L and R outputs, and the result should be exactly what you would have gotten if everything had been correct in the first place.

--best regards

P.S.: Just to be very clear, yes, it certainly does matter which way a figure-8 microphone is facing! Every figure-8 has a front and a back. The back is as sensitive as the front--but if you push a "positive-going" (pressure increasing) wavefront at the front of the microphone, the microphone will put out a positive-going voltage, while if you push the same wavefront at the rear of the microphone, you will get an equal, negative-going voltage instead. In other words, the front and back of the microphone have identical response except w/r/t polarity. And the whole way that the M/S approach works is based on using that behavior of the "S" microphone to discriminate between the sounds arriving from the one side versus the other.

[voltronic]

It sounds like you're using a microphone with a built-in matrix going by your description of the cable. 

The only thing I'll add to what DSatz said is that I personally don't like doing to a L/R matrix when recording.  You lose the ability to manipulate the stereo width / room balance later on, which you CAN do if you record un-matrixed "mid" and "side" channels.  To me, that's the single best thing about recording using this method, especially if you're in a room with great acoustics.  Here's a decent guide:

http://www.uaudio.com/blog/mid-side-mic-recording/

[Gutbucket]

^^
That.

If the matrix ratio you use is 50/50 and it's all done digitially, there is no loss going either way (M/S > L/R  or  L/R > M/S).  If you adjust the ratio away from a straight 50/50 to favor either the Mid or the Side to tweak the resulting pattern to your liking, then there is potential for dynamic-range loss in a following adjustment, but not enough to be of concequence if the change is minimal, which it almost always will be in the real world.  Obviously if you were to use a radical ratio you loose information.  Consider the extreme of matrixing using a ratio of 100% Mid to 0% Side.   You could not rematrix again and retrieve any Side information again after doing that.

[speedo]

The setup is a schoeps mk4 and mk8. I am going through a connbox and that terminates in a 5 pin cable. Then I have a 5 pin female branching to 2 xlrs that go into an sd-702t. I am monitoring in ms mode. This diagnoses is only from monitoring. For instance, the other day an airplane was flying overhead. I heard it on the right side but then looked up and the airplane was on the left.
I can easily flip the phase on the second channel but I am curious why this is happening in the first place.

[voltronic]

I never considered going the other way - that's good to know.  I wonder if something like this is happening in the chain of some plugins when they allow M/S proccessing of L/R material.  I'm thinking of the EQ module of Ozone for instance.

[Gutbucket]

I wonder if something like this is happening in the chain of some plugins when they allow M/S proccessing of L/R material.  I'm thinking of the EQ module of Ozone for instance.

That's correct.  Also used for FM radio transmission, analog television transmission, etc.  In those cases if the side-band carrier with the Side signal is lost, the audio seamlessly reverts to the primary band carrying the Mid signal.  So if reception is spotty, the receiver looses stereo and reverts to mono.  You can hear that happen momentarily on FM radio while driving sometimes.

The setup is a schoeps mk4 and mk8. I am going through a connbox and that terminates in a 5 pin cable. Then I have a 5 pin female branching to 2 xlrs that go into an sd-702t. I am monitoring in ms mode. This diagnoses is only from monitoring. For instance, the other day an airplane was flying overhead. I heard it on the right side but then looked up and the airplane was on the left.
I can easily flip the phase on the second channel but I am curious why this is happening in the first place.

Are you setting things up so the the red dot on the MK8 (the positive polarity side) is facing the left hand side as viewed from the recordist's perspective?  If so then polarity may be swapped in the connector or con-box wiring.

A quick practical fix other that tracing down and correcting that potential wiring error is to simply turn the MK8 around 180 degrees so the red dot faces the right side.  The side signal will then start out polarity inverted and be inverted back to correct polarity by the time it reaches the SD-702 M/S monitor matrix.

I'm not super familiar with the SD recorders, but I imagine they may have a provision for inverting polarity of the inputs if desired.  You might check the setup on the recorder to make sure that is not where the inversion is happening.

[macdaddy]

What if he had the 0 degree of the side pointing right (when viewed from behind the mic, towards the stage).? Isn't there a way to adjust that after the fact when processing with the computer..? I would assume one could invert the phase of just the side channel and then m/s decode..?

[DSatz]

macdaddy, yes, that can be done in any editing software that I'm familiar with.

[macdaddy]

Thx. I asked because it seems to me that even though the popular m/s plugins say they can invert polarity, they don't (msed, for example) specify whether they are inverting polarity for both channels, or only the side. it seems to me in this case of the fig 8 being oriented opposite that the solution would be to invert the side channel's polarity and then do the m/s decoding...

[Gutbucket]

I asked because it seems to me that even though the popular m/s plugins say they can invert polarity, they don't (msed, for example) specify whether they are inverting polarity for both channels, or only the side.

Most likey, the plugins only invert the Side signal, since doing that flips the left and right sides of the output and doing so is a common task. If you were to invert both the Mid and Side polarities, the resulting left and right outputs would no longer be flipped, but the polarity of both outputs would be inverted.  That polarity inversion of both channels usually isn't obviously audible on it's own, if audible at all, unless mixing with other microphone channels.  But it can be with some instruments.

It's the polarity relationship between Mid and Side which determines which side is output as Left and which side is output as Right.  So you could also flip the Left and Right outputs by inverting the Mid signal rather than of the side signal, but in that case both Left and Right outputs would have inverted polarity.

[macdaddy]

Thanks.

Makes total sense, and confirms what I thought. I was more playing devils advocate, esp. Since some of the plugins are unclearly labelled and documentation tends to be lacking...

I still think it safer to invert the mid channel first if need be, and then run the plugin, but that is just me-I like to limit plugin use and control it exactly, rather than guess and trust my ears...

Ymmv

[kindms]

Thanks.

Makes total sense, and confirms what I thought. I was more playing devils advocate, esp. Since some of the plugins are unclearly labelled and documentation tends to be lacking...

I still think it safer to invert the mid channel first if need be, and then run the plugin, but that is just me-I like to limit plugin use and control it exactly, rather than guess and trust my ears...

Ymmv

I use this video all the time doing m/s in audacity. I find it explains things well and gives a good work flow

http://www.youtube.com/watch?v=eVOBRx7zGzw