Strange Idea...

[OneRoomStudios]

Hey Harvey,

I had a strange thought today. Wouldn't it be theoretically possible to use Mid-Side style phase cancellation to "remove" (read: attenuate) the rear lobe of a hypercardioid pattern? If you were to process a hypercarioid track the way you process the bidirectional track in a M/S recording, and you muted the rear lobe side, wouldn't the front be all that's left? I realize that a perfect phase cancellation would be necessary for complete muting of the rear lobe, but it seems that some amount of attenuation could be possible. Could this make a real "unidirectional" pattern (or close to it)?

I realize I'm probably missing some important point otherwise this would have been implemented already. Minds much greater than mine have been at work making cool audio tools for quite some time now. Would the phase cancellation not be enough to make any difference?



-Peter

[OneRoomStudios]

Stupid question? :(

[Nik Keefe]

My head hurts - talk me through that one again?

[OneRoomStudios]

I recently had a revelation, which would make that question irrelevant, but what I was asking requires background info for anyone who doesn't know or forgot the following:

If you look at a hyper-cardioid polar pattern, it is narrower than a cardioid, but at the expense of having a "tail" or rear lobe. This is because it leans slightly more in the bipolar direction than the cardioid (all polar patterns are some mixture of bipolar and omni).
"M/S" or "Mid-Side" refers to a stereo technique that employs a bipolar mic and a cardioid mic. The cardioid faces forward and captures the center of the stereo image. The bipolar is angled with one of its null sides facing the source, hence it only picks up the "sides" of the source. When processing a M/S signal for stereo, it must be run through a mid-side "decoder" or "matrix." Basically what the matrix does is double the bipolar signal, flip the phase on one, and pan them out to the sides and as a result the "middle" parts of each lobe cancel each other out (that's where the cardioid comes in).

So, with that explained, what I was asking is if you were to take a mic with a hypercardioid polar patter, double it, reverse the phase on one, and pan them so that the "tail" side was attenuated (you'd have to mute the "tail" side), would you be left with nothing but the "front"?


Ok....I think that's actually a lot more complicated than it was before, but maybe it will help.

[Nik Keefe]

I understand that MS theory, but I'm still not quite sure how you'd cancel out just the 'rear' of the mic ... but that may not be your fault - it's a while since I asked my brain to think this intensely!!!! :)

[hargerst]

Ok, I had to think about this one for a while, but I think that when you cancel out the reversed signal, it would attenuate the rear lobe. But, it would reduce the front lobe by an equal amount as well, leaving a lower level front lobe. When you bring up the volume to compensate, you would also bring up the rear lobe, leaving you exactly where you started.

[Nik Keefe]

Ok, I had to think about this one for a while, but I think that when you cancel out the reversed signal, it would attenuate the rear lobe. But, it would reduce the front lobe by an equal amount as well, leaving a lower level front lobe. When you bring up the volume to compensate, you would also bring up the rear lobe, leaving you exactly where you started.
Well that's pretty much what I thought, but I wasn't sure because I'm hardly an expert on that kinda thing. I'm glad I wasn't going crazy! :)

[OneRoomStudios]

Ok, I had to think about this one for a while, but I think that when you cancel out the reversed signal, it would attenuate the rear lobe. But, it would reduce the front lobe by an equal amount as well, leaving a lower level front lobe. When you bring up the volume to compensate, you would also bring up the rear lobe, leaving you exactly where you started.

Hmm...but that would only happen if you did this real time, right? If you used an M/S decoder plug-in that permanently changed the waveform of a stereo track, and you just muted the "tail" side, the "front" side would still have the tail attenuated wouldn't it? In other words, if I used an audiosuite plugin in ProTools do "decode" the hypercardioid signal as a bipolar, it would alter both sides of the waveform, creating a stereo track - with the front on one side and the tail on the other, right? Then again, I could be very confused and wrong...

[Marik]

Hmm...but that would only happen if you did this real time, right? If you used an M/S decoder plug-in that permanently changed the waveform of a stereo track, and you just muted the "tail" side, the "front" side would still have the tail attenuated wouldn't it? In other words, if I used an audiosuite plugin in ProTools do "decode" the hypercardioid signal as a bipolar, it would alter both sides of the waveform, creating a stereo track - with the front on one side and the tail on the other, right? Then again, I could be very confused and wrong...


The way pattenrs work in M/S is that after matrixing they get rearranged, as a fraction of mathematical summing, and for each combination will have their own equivalent. For example if you use cardioid as a mid, then after matrixing you get a X/Y equivalent. If you use fig8 as a mid, then you get a Blumlein equivalent. I don't see why if you are using hypercardioid as a mid you would not get a hypercardioid X/Y, and have that rear lobe.

If somehow you will be able to cancel this rear lobe, then the whole pattern will change to cardioid, and after matrixing you will get a cardioid X/Y.

Did I understand your question right?

[OneRoomStudios]

I don't see why if you are using hypercardioid as a mid you would not get a hypercardioid X/Y, and have that rear lobe.
I was talking about using the hypercardioid as the side - I'm not as interested in actually using it in a M/S stereo application as I am in how the M/S processing would affect a hypercardioid pattern. As I said before, I could be completely wrong, but it seems to me that if you processed a hypercardioid as a side in a M/S setup (ignoring the mid), you would be left with a stereo signal with the tail on one side and the front at the other. But the amount of phase cancellation for that to occur may not be enough.