In the summer, I met with several other engineers to test a slew of microphones and stereo techniques on the Wren organ. At the most basic level, the goal of the test was to determine which microphone(s) and which position(s) would be used for the project. The shootout carried the added benefit though of offering the opportunity to compare and contrast stereophonic techniques in a controlled setting.
Today, I present you with two samples from our session. I’ve chosen these samples because they tie in with a cerebral discussion over on the recording-enthusiast website GearSlutz.com concerning the difference in polar pattern between XY and MS. For those who aren’t familiar with the terms, XY is a stereo technique using two cardioid (directional) microphones placed at a right angle (90 degrees) to each other and panned Left and Right.
MS stands for mid-side, a technique in which a cardioid microphone and a figure-8 microphone (bi-directional) are placed perpendicular to another. The Mid is panned Center and the figure-8 is panned both Left and Right, but the phase is inverted on one side. The Both techniques are coincident – that is, the capsules of the microphones are in the same place. *It should be noted however, that both mic techniques can utilize other polar patterns, but those listed above are the most conventional.
The discussion on this page revolves around the whether the polar patterns of XY and MS are the same with regards to amplitude. I will defer to studies by Wes Dooley and Ronald Streicher that demonstrate how MS technique with a cardioid and figure-8 can become nearly the same as XY with two cardioids. See the second paragraph on page 9 here: M-S Stereo, A Powerful Technique for Working in Stereo
What I will demonstrate today is that MS is not the same XY with regards to imaging, and that XY is a poor choice for distant recording applications in comparison with MS.
To this effect, I have posted two samples recorded with a Schoeps CMTS 501 multi-pattern stereo microphone. The microphone was in the same place for both takes, running through Grace preamps into an Edirol R-44. Gain structure was the same. The samples have been approximately level matched in post, as well as having the takes aligned for more convenient comparison. The organist is Rebecca Davy (see The Cast), who was kind enough to come play the same short phrases for us over and over and over…
The clips:
In listening to the two samples, I think you’ll notice:
A. The imaging in XY is narrower than MS. The instrument sounds much smaller and distant.
B. The experience is less immersive in XY than MS. Even the ambiance when the organ isn’t playing seems to have collapsed.
The lesson here: ambiance, reverb, and room noise are important. Capturing them accurately is important. Our minds and ears are used to sorting out information amidst a busy sonic atmosphere (ever heard of the Cocktail Party Effect?). A stereo technique like XY that has a tendancy to move ambience to the center works against this by attenuating the spatial cues which our brains rely on, which is why I would never recommend XY for recording at a distance.
But if the polar pattern of XY and MS are so similar, can’t you just widen XY with an m/s plug-in?
Nope, the imaging still won’t be the same. An XY pair has so much infomation in common with the two capsules that attempted to widen or shuffle can actually make things worse. I’ve included a sample of this as well.
Schoeps CMTS 501 in XY (widened)
**TomMcC over on GearSlutz asked a great question about what ratio the two channels were in the mid:side. As Dooley and Streicher’s paper points on out page 9, if the mid:side ratio is 1:1, the array more equivalent to an XY pair of hyper-cardioids than a cardioids. I’d guess the ratio is somewhere in between, and I probably could have made the MS sound more like XY cardioids by reducing the side. However, I’m not trying to demonstrate that MS can sound like XY. I’m trying to demonstrate that XY can’t sound like MS, even after extra processing. MS offers a greater range of possibilities, and after listening to the clips, I doubt anyone would want to make their MS array sound like XY for distant miking. Want to test your ears? See if you can spots the XY among four full organ samples here.







