So, I sometimes use the SIR1 plugin for reverb (its free!). It takes impulse responses of different locations around the world to simulate reverb. You can even make your own, which I did. I took a pair of SDC's and recorded them in XY inside a giant staircase in my school. It sounds pretty cool.
To make a larger stereo-spread reverb, can I use the M/S technique? I just got a figure-8 ribbon. If I did this, could I just use the doubled side tracks (being flipped phase of course), or do I need the mid too? Or is it whatever I want?
I am intrigued to say the least! Thank you for any input!
It is only in the summing with the mid channel that M/S achieves it's stereo effect. If you merely record with only a figure 8 ribbon, duplicate that track and invert the polarity of the duplicate, you will not get what you want (in fact in mono they would cancel each other out).
But you could use one of your SDC mics as the Mid channel and record in M/S.
Best of luck!
Thank you thatjeffguy, your answer makes sense when you simply think about it.
M/S isn't really any "wider" than X/Y, and it's not M/S without a mid mic. The ribbons I've used are not true figure-eight patterns but more like bidirectional.
Many of the stereo IRs I've used were done with spaced pairs. If you want wider than X/Y that might be worth trying.
You should avoid using a condenser mic for the mid (M) and a ribbon for the side (S) channel, as you will get irresolvable phase issues. As one gives an output proportional to position and the other to velocity, they are inherently 90 degrees out of phase, and the M-S decode matrix will not work to give an L-R result.
There should be no problem in using a pair of ribbons (e.g. M160 + M130) or a pair of condensers (e.g. one NT-2A in cardioid, one NT-2A in fig-8), as these produce consistent phase signal pairs. Once decoded to L-R using the standard matrix (L=(M+S)/2, R=(M-S)/2), they can be used like a conventional X-Y pair in the IR measurements, but with the advantage of being able to control the image width by varying the amount of S channel going into the matrix. The image can be both narrowed and widened relative to that produced by an X-Y pair.
You can easily widen X-Y by copying both channels, reversing the panning on the copies, inverting polarity on the copies and mixing in with the original pair. When mixed at equal levels you have nearly the same thing as just the side mic channel of a M-S pair. That is, it's a stereo difference channel.