Topic: Stereo configs with LDCs

[HighStandDave]

Entertaining the idea of getting some LDCs at some point. Thinking Neumann TLM 102s... Or if I wanna pony up, 414s.

I read here that ORTF with LDCs is less than ideal since LDs handle low end differently than SDs off axis. What would you guys run if it was mostly cardioid? DIN?

[heathen]

Some sort of XY is worth at least considering.  There are mounts that position the mics on top of each other with one upside-down that look handy, though I've never used one of them.

[goodcooker]

I read here that ORTF with LDCs is less than ideal since LDs handle low end differently than SDs off axis. What would you guys run if it was mostly cardioid? DIN?

this article by Neumann shows the graphs of the KM184 cardioid SDC and the U87 LDC in the cardioid setting and the LDC does what most mics do - become more like an omni in the lowest frequencies. https://www.neumann.com/en-us/knowledge-base/neumann-im-homestudio/homestudio-academy/difference-between-large-and-small-diaphragm-microphones

When I was running ADK TL LDCs I usually ran them in hypercardioid setting due to the rooms I was taping in but when I ran them in the cardioid setting I ran them like this
- ORTF if very close - onstage, stagelip, closer than 20ft from the stage
- DIN if any farther back than that - or on a balcony rail
- PAS @ 28cm if the only option was at the sbd in a rough sounding room (or outside with my ADK A51 which were cardiod only)

I don't personally feel that the more open pickup of the LDC at low frequencies impacts the choice of config as much as some do.

[Gutbucket]

The most important factors are the sound of the particular mics and the practical ability to accommodate the stereo configurations you want to use.  Best to select based on those basic fundamentals.  Based on that criteria, LDs work great for a lot of tapers.  A stereo LD mic is big but can be pretty convenient and easy to run, and has proven able produce great sounding recordings.  A stereo pair is more flexible, but not as convenient.  If right or not for you is the question of course.

The "sound" of the mic is a very subjective thing, whereas setup is about practicalities. The ability to switch patterns of single pair of mics can be particularly attractive on the practical side of things, while larger size of LD mics is less compelling.  As for switchable patterns, I find it especially useful to have a supercard/hypercard pattern among the available options.  Many switchable LD mics don't include one.

I've made good recordings using switchable LD mics from all kind of recording positions, but these days really only choose to use them on-stage or at stagelip. That's just me. I choose to use them based on their sound, the pattern I want, and having them available to me.

More of the arcane details on how diaphragm size and dual-diaphragm designs affect stereo imaging in a following post..

[Gutbucket]

Here's the tl;dr on some of the less important, yet technically relevant details you may be interested in:

Technically, to produce the most accurate stereo image, the microphones used in a stereo pair should have a polar pattern which remains smooth and consistent across all audible frequencies.  Such pattern consistency is always a good general goal, and is somewhat easier to achieve in a small, single-diaphragm design.  Somewhat oversimplified, an "acoustically small" sized diaphragm makes for a pickup pattern that remains more consistent at high frequencies, while a "single diaphragm" design (as opposed to a dual diaphragm design that allows for electrically-switched pattern selection) provides for a more consistent pattern behavior at low frequencies.  Broadly, a technically accurate stereo pair is best achieved using single-pattern small diaphragm microphones.

Does that matter?  Sure, everything does to some degree, but its not a deal killer for most concert tapers.  Accuracy of stereo imaging matters more to some folks than others, but typically not nearly as much as the overall sound of the mics and if they are able to easily accommodate what you are trying to do.

Complicating the issue, consistency of pattern will matters more in some stereo mic'ing configurations than others.  The more narrow-spaced the stereo recording configuration is, the more it will tend to rely on the level differences between channels which was generated by a combination of polar pattern and microphone angle.  It matters somewhat less in increasingly spaced stereo pair configurations.

[Gutbucket]

[continued tl;dr]..

But how important is "accurate imaging" to the end result? I'd argue that a good sounding and engaging image is what most listeners value over accuracy.  The broading of the pickup pattern at low frequencies in electrically switched dual diaphragm mics will tend to reduce resulting stereo image width at low frequencies.  Some folks like that.  Some do things to achieve that on purpose.  Its something that was a technical limitation of vinyl record cutting that listeners have become used to. It can be helpful for monophonic playback from a bluetooth speaker, and limited stereo playback from a phone. So maybe its a good thing.  And it can be helpful for some things like close mic'd stereo sources.  Yet maybe not so much of a good thing for concert recordings that aim for an overall sonic portrayal of a live event.

I tend to prefer the opposite of that particular "inaccuracy" - increased rather than decreased stereo width at lower frequencies.  That translates as open-sounding / less-monophonic reverberance and a more stereophonic treatment of low frequency content.  Unfortunately that's the opposite of the trend from a switchable-pattern LD stereo pair, where the narrowing of the pattern at high frequencies produces a wider stereo image up top, and the widening pattern at low frequencies produces a narrowing stereo image down at the bottom.  I want the opposite down low.   One way of correcting that is via the application of "shuffling" afterwards which increases the difference-signal between channels at low frequencies.  There are several threads about that here at TS.  To my ear it improves many coincident and near-spaced stereo pair recordings regardless of which mics were used, yet can be especially helpful with LD coincident cardioid recordings.

Too bad its not the other way around, with the pattern narrowing down at the lowest frequencies.  That would achieve much the same as the application of shuffling.

Going even deeper into intentionally applied imaging inaccuracies, in my multichannel mic mixes where I'm intentionally manipulating the imaging by mixing and manipulating multiple mic channels, I've found I tend to like a wider image at the very bottom (as per the application of shuffling, or in my case by way of wide-spaced omnis in addition to narrow spaced directional mics), yet also like a wider portrayal at the very top, in combination with a more narrow, tightly-focused image through the midrange.  Sort of an hourglass-shaped trend in stereo width.  That's decidedly not "pin-point accurate" but produces a very attractive and engaging sound that is wider sounding and more engaging overall than would otherwise be possible without loosing good central focus of vocals and other critical midrange content.

[Gutbucket]

Here's the polar for Neumann TLM 102 (cardioid only), take a look and you can see the narrowing at high frequencies and the widening at low frequencies.

[DSatz]

Hi. Just would like to help clear up what that huge polar diagram says. If you want to get some idea of how a microphone will behave in a real-world, three-dimensional environment, you use this type of graph in combination with the 0º frequency response graph. Polar diagrams can be a little tricky to read, though, mainly because their 0º response always LOOKS identical for all frequencies. That's because the graph is intended to show only the relative response of the microphone at various spot frequencies and angles, relative to the 0º response. And since that 0º response is flat only in certain special cases, you have to read it in a slightly non-intuitive way.

But with a little practice, you can definitely learn to look at the combination of a mike's on-axis and polar response diagrams and get a feel for its behavior at some distance in a reverberant space. This is an easy case, since Neumann has thoughtfully smoothed their 0º curve to be a straight line at all the standard spot frequencies but one (8 kHz, +2.5 dB). Here I've converted the values from the polar graph into a tabular format, just taking what the diagram says and not (yet) accounting for that +2.5 dB elevation at 8 kHz:

	45º	90º	135º	180º
125 Hz	-2 dB	-6 dB	-12 dB	-17 dB
250 Hz	-2 dB	-5 dB	-10 dB	-15 dB
500 Hz	-2 dB	-7 dB	-17 dB	-21 dB
1 kHz	-2 dB	-7 dB	-14 dB	(< -20 dB)
2 kHz	-2 dB	-4 dB	-12 dB	-18 dB
4 kHz	-2 dB	-4 dB	-7.5 dB	-9 dB
8 kHz	-3 dB	-8 dB	(< -20 dB)	-9 dB*
16 kHz	-7 dB	-18 dB	(< -20 dB)	(< -20 dB)

note on 8 kHz response at 180º: a polarity inversion (distinct rear lobe) occurs here, as well as at higher frequencies between 135º and 180º)

When you map those data points onto the 0º frequency response diagram and take the 2.5 dB elevation at 8 kHz into account, you get a graph something like what's attached below:

[DSatz]

Now, to make this visually clearer, I've connected those points with straight lines, BUT that is almost guaranteed to be wrong since we really only (somewhat) know the data AT those points. Nonetheless it is a way to visualize a possible interpretation of the data we have, and to notice some overall trends:

[DSatz]

So: Depending on how much of your sound will arrive directly on the axis of each mike, and how much will arrive at other angles, the actual frequency response of the mike will vary among those various curves. The more you're at a distance in a reverberant environment, the less the 0º response determines the behavior of the mike, relative to the sum of those other curves.

The on-axis curve looks to me as if the microphone was designed primarily for relatively close-up use. Conventionally such curves are corrected to a 1-meter measurement distance, where proximity effect occurs. That raises the measured response a few dB higher than it would be for more distant pickup. When used at a 1-meter distance or less, that proximity effect raises the bass response to what is shown (or higher if one is recording closer in, as with a vocal soloist). But proximity effect is mainly a front hemisphere thing, and in an open space with sound arriving from all angles, it won't be a factor, and the bass response on axis will be somewhat less than the 0º curve indicates.

As I said, the closer you mike someone or something, the less (relatively speaking) the off-axis response tends to matter, all other things being equal. The farther away you are, the more the diffuse-field (i.e. all angles of arrival averaged together) matters.

That's all I want to say, given that I'm associated with a competing microphone manufacturer AND that on the other hand, one of the chief microphone designers at Neumann is an honored friend. I will point out that Neumann makes other microphones that are clearly better suited to more distant stereo recording, as shown by the far more nearly parallel response curves that they have throughout various angles of incidence.

[EmRR]

short form answer:

Everything becomes omni at some point in the low end.
LDC's get more directional at higher frequencies more quickly than SDC's, relatively speaking, for better or worse.  If you're running PAS it could be of benefit to have beamier treble. YMMV.....

[DSatz]

The effect of the pattern "blooming" outward at low frequencies is not just that more total bass is picked up from off-axis, but that the bass in a stereo recording tends toward mono, since both mikes are picking up mostly the same low-frequency information in pretty much the same phase and at the same time. That's great for cutting a vinyl LP, but not for modern recording media where you can decorrelate the low frequencies all you want. What you ideally would want, in order to get a feeling of spaciousness in a stereo recording, is exactly the opposite characteristic from that.

Since no real-world microphone has such directional characteristics, the best available choice is mikes that maintain their nominal directional pattern down to the lowest audio frequencies. In and around the cardioid part of the spectrum, single-diaphragm microphones are obviously better than dual-diaphragm microphones in this respect; on the lower end of the spectrum it's the general acoustical design and not the size that matters most. Over the decades, engineers have learned to work with the shortcomings of dual-diaphragm capsules at both ends of the spectrum for spot/solo recording (using isolation and proximity effect, and choosing suitable room acoustics), but two-microphone stereo is a whole other thing, requiring (relative) uniformity of polar patterns across the spectrum.

-- Yes, for front-arriving sound the treble will be "beamier" due to the larger diaphragm. But beware the fallacies of imagining that a microphone has a narrower pickup pattern than a cardioid pattern actually is (it's really more like an omni with a hole in the back), and especially that the first-arrival sound will be what the microphone mainly registers--an illusion that our brains create for our benefit as we use our ears, that we grow up with and habitually take to be the reality of our sonic environment. That second illusion can be especially tenacious.

Most microphones don't know how to pay attention to the main, direct sound component and relegate everything else to being part of the ambience or character of the space the way we humans do unconsciously. I have to say "most" because a few fancy microphones with DSP built in can actually do something like that. But for all other microphones, everything that they receive is effectively "foreground". When you're not recording close-up, and you're in a reverberant space, sound arriving from more or less all angles at once will be the great majority of what reaches a microphone. In the microphone's output signal, the sound will be substantially altered by its particular, peculiar response(s) at all those other angles. That's why off-axis frequency response matters so much for any recording that includes the space that the acoustical events are happening in.

Polar patterns, if you make the effort to read them, reveal that information in a way that most manufacturers don't give out directly. They're like a secret code that only professionals know, or amateurs who've taken the time and trouble to learn how to decode it. That's why I posted my messages above with the table and all. The manufacturers could give you that information directly, but it might undermine their sales of certain products.

-- The original generation of large-diaphragm condenser microphones, back in the 1920s and into the 1930s, were pressure transducers (not cardioids!) with a range only to about 8 kHz, but because of their size they smoothly and regularly rolled off the treble of any sound that arrived off-axis. This includes the Western Electric condensers that launched the era of electrical phonograph recording and motion picture sound almost exactly 100 years ago, as well as Neumann's first capsules from the later 1920s, which were based on the Western Electric design but using metallized plastic diaphragms rather than stainless steel.

That approach gave way in the early 1950s to using a small-diaphragm omni embedded in a sphere (typically 40 mm diameter), such that the diaphragm of the capsule is flush with the surface of the sphere. That's a really useful technique for when you can't (or don't want to) be close to your sound sources; I'm surprised that more people here don't use it. The Neumann M 50 is the best-known historical example--the classic "Decca Tree" microphone. They still had some of those at RCA Studios in Manhattan when I was working there, and it was amazing how they could be placed 20 feet from a group of musicians and convey as much detail as if they were only maybe four or five feet away, while keeping the nice room sound of a big, well-designed studio such as our late, lamented Studio "A".

[Gutbucket]

^ Very much liking these descriptions-

[snip] "..beware the fallacies of imagining that a microphone has a narrower pickup pattern than a cardioid pattern actually is (it's really more like an omni with a hole in the back)"

"..and especially that the first-arrival sound will be what the microphone mainly registers--an illusion that our brains create for our benefit as we use our ears, that we grow up with and habitually take to be the reality of our sonic environment.

Also particularly-

"Most microphones don't know how to pay attention to the main, direct sound component and relegate everything else to being part of the ambience or character of the space the way we humans do unconsciously. [snip] ..everything that they receive is effectively "foreground". When you're not recording close-up, and you're in a reverberant space, sound arriving from more or less all angles at once will be the great majority of what reaches a microphone. In the microphone's output signal, the sound will be substantially altered by its particular, peculiar response(s) at all those other angles. That's why off-axis frequency response matters so much for any recording that includes the space that the acoustical events are happening in."

That strikes me as the most on-point comment made yet in terms of explaining the underlying "yes, but" of the question.  In addition to describing how a single microphone behaves (snipped exceptions excluded), it accurately describes the environment in which concert tapers are operating, and why microphones that have more consistent, well-behaved polar patterns tend to be favored by concert tapers.   The microphones which most often satisfy those requirements tend to be high-quality small diaphragm condensor microphones. 

Further, it extends to the stereo microphone configuration in which the microphones are used.  The stereo mic'ing configuration similarly "flattens" the very complex multidimensional acoustic situation occurring at the recording position into an output that is significantly less multidimensional.  In choosing which stereo configuration to use (which particular microphones, polar pattern, angle, spacing) we exercise a limited but very useful degree of control over that "flattening", including how much unflattened dimensionality remains and what forms it takes.

[kindms]

Buy teds AKG 426 you may prefer the "smaller" foot print of a stereo LD mic. it makes coincident setups easy.

i had 414s and found setup / footprint to be challenging at times. however being able to run splits was nice of course