Taperssection.com
Gear / Technical Help => Microphones & Setup => Topic started by: Billy Mumphrey on October 19, 2022, 04:05:20 PM
-
What kind of distances between mics can you guys recommend for a 3 mic mix?
-it will be a LCR (left, center, right) on a stereo bar. (2) Line Audio cm4’s (wide) cardioid on the outside and a single at853 hyper as the center channel.
-I’m really starting to prefer PAS method and I will be doing this with the outside pair. The on-axis response, when pointing your mics forward more, is much more preferable to my ears than leaving the mics pointed outward in a standard ORTF and getting more off-axis sound hitting your capsules. I understand the more you point (directional) mics forward, you need to increase the distance of the mics to maintain stereo qualities.
I have exchanged an email and polite request with Scott from SRS (Shapeways) about taking his swivel ORTF CM4 bar:
https://www.ebay.com/itm/133336035095
and simply making it wider. Then potentially adding a clip dead center to hold the at853 hyper if he can. If he can’t I plan on finagling my own clip.
Knowing this, any recommendations on what the new width should be? If ORTF is 17cm and NOS is 30cm maybe somewhere in the middle at 25cm? Or perhaps just a NOS bar at 30cm? I hope to be running 3 mics most of the time, but in the rare occasion I only run the CM4’s, would 30cm be too far apart for a stereo pair even if they’re wide cardioids and I’m pointing them forward more?
I wish I could experiment myself but I don’t have my cm4’s yet and I wouldn’t mind having my mount ready by the time they arrive. Thanks again in advance for everyone who has indulged my hairbrained ideas as of late, I certainly welcome any critiques of my (probably) flawed logic so far.
-
No clear answer. If you haven't played with the Neumann Recording Tools app or the Shoeps Image Assistant, they'll help visualize what happens with at least the LR pair.
-
No clear answer. If you haven't played with the Neumann Recording Tools app or the Shoeps Image Assistant, they'll help visualize what happens with at least the LR pair.
oooh very cool, i didn’t know those resources existed. I will check those out after work, thanks.
I know there’s a reference table of some kind that shows the recommended distances with angle change (or something like that)) but I can’t find it and I don’t know if TS search is broken but my search results are off and it doesn’t allow the “3 mic mix” search. I’ll figure it out.
-
No clear answer. If you haven't played with the Neumann Recording Tools app or the Shoeps Image Assistant, they'll help visualize what happens with at least the LR pair.
oooh very cool, i didn’t know those resources existed. I will check those out after work, thanks.
I know there’s a reference table of some kind that shows the recommended distances with angle change (or something like that)) but I can’t find it and I don’t know if TS search is broken but my search results are off and it doesn’t allow the “3 mic mix” search. I’ll figure it out.
http://www.sengpielaudio.com/Visualization-AB60-E.htm
-
Great question. The TL;DR is to go as wide as you are able to in practical terms. Two feet is reasonable if you can push the center mic forward somewhat in front of the other two. The wider you can go the less the center one needs to be pushed forward, which is helpful in multiple ways.
If an extended answer is welcome, read on..
I understand the more you point (directional) mics forward, you need to increase the distance of the mics to maintain stereo qualities.
^
This angle / spacing relationship (in combination with directional pattern) is the core stereo relationship between a stereo pair of mics.. and by extension it applies to arrays of more than two microphones as well.
Very generally, a minimum of twice the spacing appropriate for an L/R pair is the starting point for L/C/R triplet spacing. That kind of makes sense as that makes the distance between any two adjacent microphones the same as it was before. Problem is the angle between those two adjacent mics is halved, so ideally more spacing is required to compensate for the narrower angle. The other thing driving the need for increased spacing is the angle between mics that you need to achieve PAS.
I’m really starting to prefer PAS method and I will be doing this with the outside pair. The on-axis response, when pointing your mics forward more, is much more preferable to my ears than leaving the mics pointed outward in a standard ORTF and getting more off-axis sound hitting your capsules.
^
This, in my experience is spot on. Problem is that from a typical recording position, PAS typically equates to a rather narrow angle between mics, which ideally benefits from increased spacing between them to compensate. Because of that things get wide fast. Check out the >>Improved PAS table<< (https://taperssection.com/index.php?topic=167549.msg2087409#msg2087409), also linked in my signature line. It's useful as an easy way to determine the optimal spacing between a pair of microphones when Pointed At Stacks, based on the PAS angle from the recording position and mic pattern. Yet that's two mics, not three. Three takes more spacing, but it illustrates the trend.
The Neumann Recording Tools app and Schoeps Image Assistant EmRR mentions are useful tools and work for analyzing a stereo pair of mics. These tools are primarily about predicting image width and stereo image position accuracy, which while nice to have is not the most critical thing to a good recording. But its a good start. These apps can be a bit wonky and tricky to interpret, so don't feel like mastering or fully understanding them is necessary. The Image Assistant [http://ima.schoeps.de/ (http://ima.schoeps.de/)] is able to analyze a 3-mic array in addition to a stereo pair (not sure about the Neumann app). To do that it will be easiest to start with the Decca tree preset and modify the mic patterns to emulate those of your mics, then modify mic angle (labeled as Epsilon, equating to half the total angle between L/R mics), horizontal spacing, and center forward spacing to reasonable values while watching the graph display. Read the total image width (the recording angle within which imaging should be accurate) across the bottom, where the diagonal lines curve back to horizontal at lower left and upper right. Best optimization will form a straight diagonal line running from lower left to upper right across the display, meeting up across the center without too much variation. Play with it and try to get the two diagonals to line up (usually takes adjusting forward spacing as well as L/R spacing).
Other resources:
I've attached a couple PDFs. The first covers 3-microphone configurations, except using omnis instead of subcards. You can use a bit less spacing with subcards in place of the omnis if you angle them. The second covers 4-microphone configurations using an X/Y or M/S pair in the center in place of the single mic. Consider doing that if you can. Page two shows the most compact version, which works in challenging rooms and replaces the omnis with supercardioids. That can work with subcards too. Space them a bit more if you can, but don't worry too much if you can't. (These are new revised versions of the old OMT booklet linked in my signature which needs updating)
Michael William's MMAD (Multi Microphone Array Design) website. He's the "Stereo Zoom" author, and much of the stuff above is based on his work. Start here: https://www.mmad.info/MAD/Ch_n_cov.htm (https://www.mmad.info/MAD/Ch_n_cov.htm). Similar to the apps, that website all about achieving seamless imaging between microphone pairs, in this case the multiple mic pairs of an array. It's a hyperlinked decision matrix which links to printable PDFs showing different microphone array setups. The 3-mic configuration section was never completed, but the 5 mic section works and starts with the L/C/R front portion, so just choose any configuration angle for the rear-facing mics, and it won't matter if you choose Front Sound Stage Coverage or Surround Sound. It only uses a single mic pattern for all mic positions however, so choose either cardioid or hypocardioid (subcard). Then compare choose between approximate PAS angles. It can be useful to see these known-good arrangements even if you can't achieve them simply as reference points.
-
The extended answer is welcome. :)
I have my homework tonight! It appears my initial idea for an LCR is off, at least in terms of stereo imaging, but I need to read more. I have recorded a couple LCR’s in the last year and I simply enjoyed them, especially the blues trio where I ran the mics onstage. Perhaps this means I need to do some field research and go try more 3 mic mixes with my current gear.
-
here is a 1 minute sample of a blues trio i recorded last march in a small theater:
https://www.dropbox.com/s/zb2ajplhisfmlfb/mic%20onstage%20blues%20trio%20sample%202448.flac?dl=0
3 mics LCR onstage, stage lip.
(2) C4 cards ~25cm apart, around 40 degrees
(1) at853rx hyper center
no editing except adding gain to the C4 stereo file (I added a bit more to the right channel to help the mix sound better to my ears).
I did NOT position the center mic forward more (so sound is hitting it first) as I didn't know about the potential benefits at that time. I plan to experiment with this, maybe run it ahead for one set and on an even plane for another.
Is that enough stereo imaging? Yes, for me. But probably because of the onstage setup of the trio. 3 instruments all aligned at the back of the stage about 15-20ft from my mics will do that. You can probably guess the location of the players.
If I'm in a noisy club and I'm pointed at the PA in the back of the room, I question if this setup would make that much more of a difference. The sweet spot up front could potentially catch some stage action. I'll try it out and report my findings.
-
My work headphones broke yesterday, but will try and give your sample a listen later. L/C/R works great on stage. And in that scenario you can often angle the L/R mics wider and push the center mic forward, both of which reduce the need for as much spacing between L/R.
Here's an example of such an on stage setup for a jazz trio-
(https://taperssection.com/index.php?action=dlattach;topic=1625.0;attach=109854;image)
Based on this MMAD setup-
(https://taperssection.com/index.php?action=dlattach;topic=1625.0;attach=109866;image)
Similar MMAD setup using cardioids instead of supercards (note the wider angle between mics)
(https://taperssection.com/index.php?action=dlattach;topic=1625.0;attach=109868;image)
-
To run a L/C/R without much spacing, ideally the L/R mics should be highly directional and angled widely. That works up close or on stage, but not so much from a distance. From an audience recording locations the L/C/R triangle grows increasingly wider and less deep. That trend continues the farther back the recording position is and as the angle between L&R becomes narrower. From far back the L/R spread is wide and the forward center spacing minimal.
I've found no better way to deal with a crappy room from an audience position than the "tough room' setup on page 2 of the OMT4 PDF, which uses four channels of directional mics pointed at stacks.
Advantages other than imaging-
I think for tapers the primary advantage of using three mics instead of two is increased flexibility over balance. It gives you level control over three inputs for L/R balance, center/sides balance, Left/Center and Right/Center balances. Much easier to even things out as needed to get a good balanced sound. Beyond that, the center can be EQ'd differently, which in combination with level control is a super powerful tool. If open to EQ'ing, try reducing the low frequency content in the center mic channel and EQing higher up for good clarity and presence. Or try it the opposite way by doing that to the L/R pair. You can sort of play one off the other to some degree, with lots of options to play around with. You can do more and push this kind of thing further than you could with a stereo pair because everything gets blended together in a balanced way.
Secondly, the wide spacings or wide mic angles provide a better portrayal of the diffuse reverberent part of the recording - the part that conveys the "sounds like you are there in the room" sensation - while the center mic anchors the middle so there is no hole in the middle problem.
-
I think for tapers the primary advantage of using three mics instead of two is increased flexibility over balance. It gives you level control over three inputs for L/R balance, center/sides balance, Left/Center and Right/Center balances. Much easier to even things out as needed to get a good balanced sound. Beyond that, the center can be EQ'd differently, which in combination with level control is a super powerful tool.
This. There’s an extra channnel on my recorder so why not.
I want to experiment with how far in front I like the center mic. I always assumed having your mics on an even plane in relation to the sound source was a given. Apparently the small time delay difference that comes from moving a centered microphone forward has desireable results. I must test this theory.
-
Theory? We don't need no steenkin' theory ...
The three-mike idea comes from the old Bell Labs concept of stereophonic sound in which there would be as many loudspeakers in the playback system as there had been microphones used in the pickup arrangement. When record companies such as Decca, RCA got hold of it for LP recording, home stereo setups were already defined as having two loudspeakers. The center track was thus a utility track, often used for solo elements which could be EQ'ed and balanced (as a whole) in post-production against what the L and R microphones were picking up (as a whole). Think spot miking the soloist in an instrumental concerto or an opera aria--or in a recording of a complete opera, all the singers as a group in relation to the orchestra, where the singers were placed away from the orchestra during the session, so that their closer pickup in the C track would overwhelm their accidental pickup in L and R. This relieved the producer and engineers from having to set those balances during the live session, where they would then become unalterable.
Many Decca recordings had the center microphone placed forward of the line between L and R, but it's a myth that there is one definitive "Decca Tree" geometry; they tried many different arrangements over the years, never settling on any one of them, and often using baffles between/among the microphones to increase the separateness of their pickup from one another. However, in no event did they follow Gutbucket's idea that L and R should be spaced at all widely apart. The entire European conception of spaced-omni miking for stereo is based on a smaller distance between L and R than has ever been traditional in the United States; there are incoherencies in the stereo image (ambiguous / unstable localization of sound sources) that occur with widely-spaced omnis that their ears can't stand (and having learned to hear them myself, neither can I).
RCA, on the other hand, did space their L and R microphones apart more widely, with C deployed either in the true center or else closer to the soloist(s) depending on how the track was intended to be used--as a spot mike, or as a general fill-in. The engineers always ran the center track as hot as possible below the overload limits of the tape, then in the 3->2 mixdown, (a) if C was spot-miked soloists of course the levels were adjusted by ear and from section to section of the music, while (b) if it was a general center fill, then its level was always reduced about 6 dB on average so that it added clarity and anchored the stereo image but didn't collapse everything into mono, as it would do if mixed in at anything like equal gain to L and R.
The Mercury "Living Presence" recordings on the third hand so to speak, if I'm not mistaken, came closer to the idea of three microphones all in a row. The Fines (husband and wife team who recorded the series) didn't always use the same mike for all three positions (actually, six since for safety they ran two separate sets of mikes into two separate three-track recorders), but that was mainly because the mike that they wanted to use for all six positions had already been discontinued before they started recording in stereo, and it was difficult to scrounge up six of them--it took them a few years to do so, but from then on they used that one type of microphone--the Schoeps M 201 in its omni setting--for all their stereo recordings. That's a roughly diffuse-field-equalized pressure transducer with a collar around the diaphragm (apart from when the mike was set to its cardioid position, but they never did that).
Decca and RCA also used microphones with rather special characteristics for these recordings: small-diaphragm pressure transducers (single-diaphragm omnis) where the capsule was embedded in a 40 cm sphere, which gives the microphone added sensitivity in the "presence" region in the front, and causes the on-axis rise in the treble to start at somewhat lower frequencies than it otherwise would for a small diaphragm construction. (The collar around the omni capsule in the Schoeps M 201 had a somewhat similar effect.) Their preferred mike was the Neumann M 50 although both labels also used the M 49 at times--a weird substitution since the capsule construction and sound are quite different. It may have been some situations in which the M 50 wasn't available, or some odd acoustic in some recording venues that required them to use the M 49s as directional microphones, an option which the M 50 didn't offer.
In addition to the three main microphones, Decca and RCA used spot mikes sometimes, and/or "outriggers" (additional, quite widely-spaced L and R microphones whose signals were mixed in to the L and R tape channels at lower levels), but Mercury as far as I'm aware never did so. In all cases there was a separate control room setup with loudspeaker monitors, and test takes were always made at the start of the session, then played back and listened to, so that adjustments could be made before starting the "real" recording.
--best regards
-
Good historical references, thanks. Past informs the present.
To clarify, when I suggest the L/R pair of an L/C/R microphone array be "spaced widely apart", that's relative to the "near-spaced" arrangements such as ORTF, NOS, (DIN), etc. concert tapers commonly use. Relative to the old RCA and Mercury recordings these spacing are not anywhere close to as wide. The methodology is quite different.
Backing up a bit- As I see it, the most common problem concert tapers encounter when attempting to introduce additional microphone channels (that is to say, using 3 or four microphones on the same microphone stand, with the intention of mixing them all down to two channel stereo) is placing too many microphones in close proximity / not angled far enough apart from each other to avoid problems. And this is exactly what I read as being asked about in the original post in this thread. The well known two-channel near-spaced stereo microphone pair configurations represent optimized combinations of spacing, pattern, and angle.
Introducing a third channel throws that optimization out of whack and an entirely different combination of spacing, pattern, and angle is required. Simply adding an additional center microphone channel to a standard two-channel near-spaced stereo pair configuration may seem attractive in practical setup terms, yet is in my opinion not the way to go.
There are essentially two good choices:
1) Stick with two channels and abandon the idea of using additional microphones. This is arguably "correct", certainly more simple to implement and is likely the most appropriate choice for most folks. Use any extra channels for recording the soundboard feed, spot mics.. or to record a second stereo pair arrangement for comparison purposes, choosing one or the other, yet not with the primary intention of mixing them together.
2) Re-arrange things so that a microphone array which includes additional microphone channels it setup to work correctly without problems. This is what I've been playing around with for years and what I'm attempting to help Chanher with in his exploration of this path. I've found it engaging and worthwhile for a number of reasons, including the pragmatic aspect of gaining some degree of increased control in situations where tapers otherwise have very little of it, and also in terms of the quality of the resulting recordings when it is done right and it all comes together correctly.
-
I have played around in the past with wider "in a line" spacings somewhat more similar to what Mercury "Living Presence" was using, perhaps RCA, some of Owsley Stanley's Old and In the Way recordings, and more recently described by David Griesinger. Mostly at the stage-lip, using three, four, or five relatively wide-spaced omnis arranged in a line across the front of the stage. It worked and was fun to do but isn't really practical to do regularly. Good practical balance control, good resulting recordings, and a fun taper option for some things, yet very different from what most tapers will want to do and what we are discussing here.
A better reference for the spacings I'm suggesting is the more modern 3-channel arrays dating back not 6 decades but more like 2, intended for L/C/R speaker playback (think surround system playback, excluding the surround speakers.. or not). This is were I initially started with this exploration of recording more than two microphone channels and how best to arrange things for that approach, and I was somewhat surprised and pleased that what worked well for multichannel playback translated extremely well to two-channel stereo playback as well while providing the very welcome increase in practical flexibility mentioned previously.
Am I right or wrong? Do I care?
The suggestions I'm making here are based on my empirical explorations of this stuff, informed by review of historical methods of recording, the psychoacoustics of human hearing, lots of experimentation and listening, and working within practical constraints of the situations in which tapers find themselves doing our thing. In doing so I've tried to determine what the underlying core relationships are that seem to be most important, and I try to describe things in those terms on this forum rather than simply saying, "just set things up this way". It works really well for me, and I'd like to give back to the community by helping other tapers who are interested in exploring recording in similar ways.
That said, my recommendation to most folks remains to stick with two channels!
Fair enough?
-
Specifically on the thing about forward spacing of the center microphone- Don't worry about it too much.
Achieving sufficient spacing/angle between adjacent mics is the main thing (something typically near-spaced-like is about right, more as the angle between adjacent microphones grows narrower). The primary reason for pushing the center mic forward is a way to achieve that if/when you can't otherwise get the L/R pair spaced widely enough, which is to say about twice the spacing of the standard near-spaced two-channel configs). I'd rather have the L/R pair wide enough that the center doesn't need to be forward by much, but having it forward a little bit is good.
^This is not primarily about imaging. Those imaging apps have the center microphone pushed forward to get the image distribution to link up nicely across the center, and playing around with adjusting the forward spacing parameter in them appears to produce major implications. Don't worry about that too much. Technically accurate source position imaging is nice to achieve, yet is well down on the hierarchy of what is most important. You'll get plenty of good sounding width and imaging regardless of center-channel forward spacing.
Here's a reasonable practical take on optimizing it- assuming you can achieve sufficient L/R spacing to begin with, push the center microphone forward just enough that the Left-Center microphone pair is perpendicular to an imaginary line pointing at the left edge of the ensemble or left PA stack, and the Center-Right microphone pair is perpendicular to an imaginary line pointing at the right edge of the ensemble or right PA stack. How far forward that is will be determined by how far back the recording position is.
-
Great discussion, as always. I am typically running an OMT8 with 4ft spread for omnis, but that is with 3 other pairs of directional mics in between them. 2ft should be plenty for two cards and a hyper. The Smallrig 15mm camera rails make a great modular mic bar with the appropriate hardware. Not too obtrusive, light, and sturdy. I have my most directional pair, short shotguns, on a separate bar out front maybe 6 inches or so. I am just trying to get as much direct signal as possible at the show, and I think in that regard the additional directional mics do make a difference.
-
^You're fully into the optimized multichannel thing.. exploring the logical extreme of the thing much as I do. I hesitate to suggest most folks go that far. Its fun though!
One more general thought-
Like most things, the key is gravitating to the "just right" mid-point between extremes. Not too much spacing or angle, yet also not too little. Not too many channels, but neither too few.
The multi-microphone array I recommend to most folks wanting to try using more than two microphones in combination is four channels total: a coincident center pair placed in between a "mid"-spaced pair using spacing twice as wide as a typical near-spaced pair used on its own. That represents a practical, not overly complex arrangement not difficult for tapers to achieve. The better it sounds in the room at the recording position, the more open of a pickup pattern can be used in the twice-near spaced pair. And the more open the pair the wider it should be. Not that complicated.
The coincident pair does a nice job with imaging and center solidarity that might otherwise be weak in the twice-wide pair. The twice-wide pair does a nice job providing an open ambient feel a coincident pair often tends to lack. Because both in combination serve to cover the weaknesses of the other, each of them need not be as perfectly configured as they otherwise would need to be when used on their own in isolation.
Same guidelines for spacing as the 3-microphone configurations discussed above, but the coincident pair in the center instead of a single microphone is a big advantage in my experience. For those suspicious of arrays of more than 2 microphones, this is what I'd suggest trying. It works really well in practical terms for concert taper scenarios. Try it and see if you like it. It's perfectly okay not to like it, but try it first in a few taper scenarios, give mixing the two pairs a bit of listening effort, and then decide if its something you like and worthwhile or not.
-
A lot of excellent information and history in this thread, thanks guys.
My preferred approach is to take in as much info and theory as you can, then throw it out the window and go tape. set up however you want. If you keep taping and learning you'll have good results.
I'm planning a low-profile homemade 3 mic bar in my head, will post pics.
-
I recorded Fleece, a fun 4 piece from Montreal the other week and the stars aligned for a 3 mic LCR in the "sweet spot" + mono soundboard. I thought I'd share some samples so everyone can hear and compare a stereo pair, a 3 mic mix, and a 3 mic mix + a soundboard feed.
This was at Lost Lake in Denver and it is a smaller dive venue. The room is a bit odd as it's more wide than long, BUT that means the "sweet spot" is much more accessible as it is literally right where the soundboard is (slightly LOC but only 25-30 ft from the stage/PA). Chris the soundguy let me put my mic stand up on the actual table (legs closed) and taped to the corner which pretty much puts it pretty much DFC. Perfect spot.
CM4's are ~28cm apart at ~40 degrees and I ran them into the MP2
C4 hyper as center mic (bass roll off switch engaged), about an inch ahead of the CM4's. Straight into the dr70d
I have 4 different samples:
01 stereo CM4 pair (normalized)
02 mono C4 hyper with bass roll off switch engaged (normalized)
03 3 mic LCR with some slight EQing and normalized
04 3 mic LCR mix + mono soundboard mastered, EQ'd, and normalized
https://www.dropbox.com/scl/fo/akf2ez14mg7vthk1fdfv5/h?dl=0&rlkey=6zctnokzb1aylf65qv3g8potm
Sorry for the crappy pics, my phone's camera is all screwed up and it took me forever to just to get these crappy pics. All my money has gone to gear so no new phone haha.
-
some notes:
-I didn't put a sample of the mono soundboard feed, if anyone wants to hear that lmk. I was expecting the usual vocal-heavy mix, but it was actually very keyboard heavy. I did my best to even it out with compression and EQ but I'm still somewhat inexperienced with these tools. As many of you know, mixing in uneven soundboard feeds is a little dangerous; it very often has a negative effect on the actual MIX of the instruments. We naturally want to improve our recordings and the clarity of a soundboard feed technically does that but at what cost? If it upsets the balance of instruments and vocals (some instruments suddenly are clear and upfront while the others are distant and quieter) is that actually improving the recording? I'm still undecided if the sample with the sbd is listenable as I do feel the keyboard is a bit louder than the other instruments. Opinions and critiques are welcome.
-I should definitely mention that even though I had the center mono hyper ~1 inch AHEAD of the CM4's (as recommended), I ended up visually lining up snare hits on ALL sources/feeds. I did some quick listening to the 3 mic mix with the center mic left alone and then with it visually aligned (the actual difference was miniscule) and if I'm being honest I didn't hear any difference so I left it lined up. I know Gutbucket talked about moving the center mic forward more in relation to the distance of the outside (left right) stereo pair and perhaps in future recordings I can experiment with that.
-
I did some quick listening to the 3 mic mix with the center mic left alone and then with it visually aligned (the actual difference was miniscule) and if I'm being honest I didn't hear any difference so I left it lined up.
Sound moves about one foot per millisecond (it varies a bit due to a bunch of factors). So one inch is only a twelfth of a millisecond. Reports vary, but that is still below the lowest estimate of audibility that I recall seeing. I would be very surprised if you could hear that. I would still line them up, though, if for no other reason than "because why not"...
-
Thanks for the samples, I'll try and give them a listen maybe tonight or this weekend.
I don't expect the exact forward spacing of the center mic to be super-critical here, and it looks like a reasonable arrangement in your photos. In general, use a bit of forward spacing for the center microphone and don't worry too much about it. In one sense it may be advantageous to arrange things so the Left/Center mic pair is perpendicular to an imaginary line pointing at the left PA speaker and the Center/Right mic pair is perpendicular to an imaginary line pointing at the right PA speaker, which is easy enough to visualize and apply at setup.
May not be necessary, but if you'd like to play with slight modifications to this arrangement the next time out, I'd try increasing the spacing between Left and Right microphones when the center mic is in play. One way to know when such a 3-microphone arrangement is close to optimized is when the Left/Right pair sounds a bit over-wide or a weak across the center when listened to in isolation with the center channel muted, but the 3-channel mix is well balanced with the center channel unmuted. That's not critical to making a good recording which benefits from the center microphone, but will tend to make the best of the combination of the 3-microphones.
More technically-
In terms of imaging, forward spacing of the center microphone is mostly relevant when playing back the 3 recorded channels discretely over 3 speakers across the front, in regards to getting a precisely accurate image distribution across the boundary between the left-center speaker sector and the center-right sector.
In terms of achieving a low value of diffuse field correlation (which is what makes the ambiance and reverberation in a recording sound natural, open and enveloping) significantly greater spacing is required (a few feet, rather than a few inches), which when applied to how far forward the center microphone position would likely require application of delay or visual alignment to bring the direct-arriving sound from the stage and PA back into proper coherence across all three channels. Increasing spacing between the Left and Right microphones won't require additional delay or visual alignment to keep the sound arriving from the stage/PO coherent while likewise decreasing diffuse field correlation, and even a slight increase helps achieve better diffuse decorrelation, up until it gets too wide such that the center microphone starts sounding more like a separate island of sound and the good smooth, solid image distribution between speakers begins to suffer.
-
I would still line them up, though, if for no other reason than "because why not"...
I gotta line em up. Gutbucket is right, I shouldn't worry about it too much, but I can't help it.
I'd try increasing the spacing between Left and Right microphones when the center mic is in play. One way to know when such a 3-microphone arrangement is close to optimized is when the Left/Right pair sounds a bit over-wide or a weak across the center when listened to in isolation with the center channel muted, but the 3-channel mix is well balanced with the center channel unmuted. That's not critical to making a good recording which benefits from the center microphone, but will tend to make the best of the combination of the 3-microphones.
Yes, I was limited by that particular stereo bar. But I'm sooo close to finishing the final version of my homemade adjustable stereo bar that will allow widths up to 47cm for the outer pair. I used an early prototype for a 3-mic trial recording last month with the cm4's at 47cm and at that width the cm4's by themselves are (obviously) much too far apart. But when I added the at853 hyper to the mix it was much more balanced and full. Even for a 3 mic mix my initial impression is that 47cm (for the outer pair) is a bit too far, fortunately I'll be able to play around with that width and I'll probably try around 40cm's when the bar is completed. That room (and mix) sounded like shit so I didn't wanna put the time into mixing and mastering that particular recording.
I've been thinking about setting up a video camera right behind my mics at a show and slowly (and smoothly) sliding the outer mics to different widths and angles WHILE recording a band. Then line up the audio with the video in post and we'll be able to HEAR how different widths and angles affect the sound while watching the mics being moved. I'm a visual learner and for some reason this kind of experiment really appeals to me. I need to finish my stereo bar first and then find a show in a decent sounding room where I have the flexibility to pull this off, the stand can't be raised really high unless I wanna bring a ladder (not happening).
You mentioned previously in this thread that a 4-mic mix is actually the best sounding option for anyone wanting to go beyond 2-mic stereo recordings and I agree; I'm just limited by my 4-track DR70d and I tend to have the option of a mono sbd feed at a lot of shows I tape. It just seemed logical to me to try to use the 3 available channels (after getting a mono sbd feed) in the form of a LCR 3-mic mix. Hopefully these experiments and samples demonstrate just how beneficial the 3rd mic can be, I'm certainly open to the possibility that it's not worth the effort. It's definitely a lot of fun trying it out though! When I have some shows with no sbd option I'm excited to try (2) at853 hypers in XY between either some omni's or the CM4 wide cardiods.
These Tascam DR70d's are going so cheap now and IMO the current firmware is stable and reliable. I'd sure love to upgrade to the new Mixpre's/Zoom's but I plan on keeping this for awhile as I only paid $179 on ebay and the Mlady batteries on Amazon are $22 for a 2 pack and 1 of those will power the DR70d for 10+ hours. For around $200 you have a 4 track recording setup, unheard of in the early 2000's when I started experimenting with aud + sbd matrices. Side note someone buy Dan's Mixpre 3ii in the yardsale before I do haha.
-
I would still line them up, though, if for no other reason than "because why not"...
I gotta line em up. Gutbucket is right, I shouldn't worry about it too much, but I can't help it.
I am right there with you! It isn't hard to do and it would nag at me if I didn't align them (even if I couldn't hear a difference).
-
Note on lining up the waveforms in the DAW- Don't get flustered when aligning files from a non-coincident microphone configuration only to find that some of the peaks line up perfectly while others don't. Only sound sources located directly in front and behind the array should line up exactly. Any source located over to the left or right will naturally be somewhat misaligned as the wavefront will reach one microphone slightly before the other, more so the farther off to one side or the other the source is and the wider the spacing between microphones. By contrast, with coincident microphone arrays all peaks and valleys should line up regardless of source position. Extra credit for anyone pointing out the special case detail with certain coincident arrays.
The idea of videoing the setup while adjusting the spacing is brilliant. You can really get a good feel for dialing it in, for how much things change and by how much, by doing it that way. Just keep in mind that you are likely to end up with different spacing preferences depending on things like how far away the recording position is and what angle you are using between the left/right pair.
I did something similar when playing around with the spacing between omnis years ago, but did so by writing down the spacings and the recording time-point for each change on a note-pad. Lots of scribbling. Videoing it sounds much easier. If not too much hassle, share the video here. I'd love to check it out and bet others would too.
To clarify, I think 4-mic arrays using a coincident center pair (which is still only 3 LCR microphone positions in space) represents a good/reasonable sweet spot for most tapers wanting to play around with using more than two microphone channels. Best sounding might be just two channels.. or six, or whatever. Too many variables to predict best sounding, and a lot of it is how you set things up and how you use whatever you've got. Mostly I think using more than two microphone channels can help to stack the deck in your favor, making it more likely you can produce a great sounding recording given all the real-world constraints under which tapers record.
-
Note on lining up the waveforms in the DAW- Don't get flustered when aligning files from a non-coincident microphone configuration only to find that some of the peaks line up perfectly while others don't. Only sound sources located directly in front and behind the array should line up exactly. Any source located over to the left or right will naturally be somewhat misaligned as the wavefront will reach one microphone slightly before the other, more so the farther off to one side or the other the source is and the wider the spacing between microphones. By contrast, with coincident microphone arrays all peaks and valleys should line up regardless of source position. Extra credit for anyone pointing out the special case detail with certain coincident arrays.
The idea of videoing the setup while adjusting the spacing is brilliant. You can really get a good feel for dialing it in, for how much things change and by how much, by doing it that way. Just keep in mind that you are likely to end up with different spacing preferences depending on things like how far away the recording position is and what angle you are using between the left/right pair.
I did something similar when playing around with the spacing between omnis years ago, but did so by writing down the spacings and the recording time-point for each change on a note-pad. Lots of scribbling. Videoing it sounds much easier. If not too much hassle, share the video here. I'd love to check it out and bet others would too.
To clarify, I think 4-mic arrays using a coincident center pair (which is still only 3 LCR microphone positions in space) represents a good/reasonable sweet spot for most tapers wanting to play around with using more than two microphone channels. Best sounding might be just two channels.. or six, or whatever. Too many variables to predict best sounding, and a lot of it is how you set things up and how you use whatever you've got. Mostly I think using more than two microphone channels can help to stack the deck in your favor, making it more likely you can produce a great sounding recording given all the real-world constraints under which tapers record.
Once again you are correct, peaks may not line up within a stereo file as the sounds (from the sides) are hitting the mics at (slightly) different time intervals. that's what creates stereo sound! how quickly I forget, perhaps I'll leave the center track alone.
I shouldn't have used "best sounding" when describing 4mic vs 3mic. Describing 4mics (with 2 coincident mics in the center) as the "sweet spot" when experimenting with more-than-2-mics is a good way of putting it.
It might be a while before I can get to the video experiment but hopefully sooner than later and I'll certainly post the results here. Thanks again for the guidance.
-
I've done some three mic setups, though I don't claim any expertise.
https://archive.org/details/spafford2018-10-27.AT4031.AT853sc
https://archive.org/details/Spyscraper2019-02-02.e614.AT4031
https://archive.org/details/goose2019-02-02.OCT2.e614.AT4031
https://archive.org/details/cycles2018-10-27.AT4031.AT853sc
-
Agree DR-70D is a great box for the money. If you want to expand on a budget the DR-680 (mki or ii) can be had for cheap and you can get up to 8 channels (with digi-in).
-
Following up on this discussion..
Here's a link to the Schoeps image assistant (available in on-line version and phone app) capable of assessing 3-microphone setups in addition to 2-channel ones, which allows for choosing a different pickup pattern for the center microphone position, and can be fun and informative to play around with- https://schoeps.de/en/knowledge/image-assistant.html
I thought this was posted early in the thread but it appears not.
-
This app looks great. I could have used this just a week ago. But what does the Epsilon value mean here? It doesn’t seem to alter the visualization at all.
-
It's the angle of the L/R mics, measured as half of the total angle between them (90 degrees = fully side facing, 180 degrees apart).
If altering its value does nothing, its likely that mk2 omnis are selected for both the left and right microphone positions, which I think is the default. Switch them to a directional pattern and you'll see any changes in Epsilon reflected in both the top-down view of the array in the upper left and the visualization curves.
-
Best sounding might be just two channels.. or six, or whatever. Too many variables to predict best sounding, and a lot of it is how you set things up and how you use whatever you've got. Mostly I think using more than two microphone channels can help to stack the deck in your favor, making it more likely you can produce a great sounding recording given all the real-world constraints under which tapers record.
I wanted to note an experience this weekend that applies to anyone thinking of adding more mics or currently doing it. I showed up early to a local jazz jam, hoping to get some good 3-4 mic onstage action. Due to multiple circumstances I made the smart decision to ditch all the new gear and just run a simple stereo setup in the sweet spot of the room. Very glad I did. I could just tell the onstage mix would be skewed; they mic'd the quieter instruments through a small PA so out in the room would get the best mix of all instruments and thus the best recording. Ended up with an excellent recording that is very listenable.
My point is don't be afraid to master the art of 2-mic stereo recording as well. I thought of doing 3-mic LCR in the audience, with the center channel picking up some sweet sound direct from the stage, but 2 discrete mics just felt better than a larger 3 mic setup as I was right in front of the majority of the viewing audience. Real-world constraints. I ran at853rx cards PAS straight into the XLR's of the warm mod pmd661 and I couldn't be more pleased.
Side note I'm getting fatigued with the at853 hyper capsules, I think I'm gonna save them for the shittiest of venues and/or extra chatty situations. This is my first time using the warm mod 661 and I think it is just as smooth but more detailed than the mp2. I personally find these mods to be subtle yet still adding pleasing frequency bumps in the right areas. They (usually) make raw files more listenable, skipping tedious post work IMHO. I can't wait to try the CM4's into it.
Also that Schoeps app is VERY cool.
-
-I should definitely mention that even though I had the center mono hyper ~1 inch AHEAD of the CM4's (as recommended), I ended up visually lining up snare hits on ALL sources/feeds. I did some quick listening to the 3 mic mix with the center mic left alone and then with it visually aligned (the actual difference was miniscule) and if I'm being honest I didn't hear any difference so I left it lined up. I know Gutbucket talked about moving the center mic forward more in relation to the distance of the outside (left right) stereo pair and perhaps in future recordings I can experiment with that.
Saturday I installed smaller Movo windscreens in place of the big Shures in my recording rig for a day of recording Sunday. While doing that I was checking the 3-microphone L/C/R portion of the array against what is suggested by the Schoeps Visual Assistant for this combination of pattern, spacing and angle. In it, playing around with the center forward spacing mostly affects how correctly the imaging links up between the L/C and C/R segments of the array. I found that given the pattern, spacing and angle I'm using between the L/R microphones (supercards, 24" [61cm], 45 degrees), Visual Assistant suggests more center forward spacing than the ~6.5" I am using currently. Alternately or in combination with that, I could push L/R farther apart and/or angle them wider. Shifting L/R a bit farther apart is what I think I'll do when I rewire everything at some point, as that's when I have the opportunity to adjust wire-lengths and move things around more freely. The current L/R spacing was originally setup for use with a wider L/R angle, but I found I like the L/R pair more on-axis with the PA so I now angle them +/-45 degrees, and should increase the spacing to compensate for that change in angle. Ideally I'd like to figure out a way to easily adjust the angle of the L/R pair to anything between, say, +/-30 deg and +/-90 deg when setting up in the field, then adjust spacing to fit the angle. Would be super slick to build the bar-clamp mechanism in such a way that things are mechanically linked, so the angle would change simultaneously as spacing is adjusted, and all incorrect combinations are automatically avoided, but that's more mechanical complication than necessary.
The only 3-mic arrangements in Visual Assistant that work with no center spacing at all are super-wide arrangements of omnis (typically) in a line. For relatively-narrow arrangements with L/R spacings closer to that of typical 2-channel near-spaced arrangements, the only solution via Visual Assistant is to push the center microphone position much farther forward, something like 1 to 2 meters or more (in plan view the LCR triangle becomes very tall with a narrow base, rather than wide at the base with relatively minimal height), which then requires the application of delay or manual time-position shifting of the center signal in the DAW such that an impulse from the stage is time-aligned across all three channels. That's not a practical taper arrangement at all though, as front back spacing is usually more difficult to achieve in taper situations than left/right spacing.
As mentioned earlier in this or in another thread, if free to adjust pattern, spacing and angle, the sweet spot for tapers with regard to forward spacing of the center microphone position might be to use whatever arrangement achieves good image linking while placing the baseline of the L/C pair perpendicular to a line to the left PA stack, and vice-versa for the C/R pair and the right PA stack. Sort of PAS with regard to the each baseline on either side of center. That should optimize imaging as well as transient arrival from each stack to each side. It will also tend to produce a triangle arrangement that is significantly wider at the base (L/R) with relatively minimal forward spacing of the center.
-
My very crude visual depiction of gutbuckets “triangle” explanation of how to place the forward Center mic in an 3-mic LCR arrangement. As I understand it, you want to make an imaginary triangle (math people plz tell us the type of triangle) with the Left mic, Center mic, and left PA speaker, and then a second triangle with the Right mic, Center mic, and right PA speaker. I plan on trying this the next chance I get.
Please correct or add anything.
-
Yes, that's what I'm proposing as a reasonable, relatively easy to apply rule of thumb that aims to get time-alignment right natively without needing to do anything in the DAW, while also achieving about the right forward-center spacing needed for good image linking.
-
Below is a screen-shot of the L/C/R microphone arrangement currently comprising the center portion of in my rig, simulated in Image Assistant after measuring it up last weekend. The angle of the L/R microphones in the rig are currently constrained to +/-45 degrees. From the Image Assistant curves it appears this arrangement would benefit from a bit more center forward spacing so as to achieve optimal image linking.
So in the second image, I've shifted the center mic 5cm farther forward, which visually improves linking in the graph, but also widens the SRA a slightly by doing so.
BTW, it sounds good as is, without using any time alignment - and the forward spacing is pretty close to achieving the geometric alignment Chanher and I discuss above.
-
Following up, I made a handful of 3-mic LCR recordings in 2024. Following Gut's advice, and against my own instincts, I placed the center mic around 4-5 inches ahead of the flanking Left-Right pair. Pics below. Long story short, I REALLY enjoyed the results. I'll post samples soon.
(2) Line Audio CM4 55cm apart, PAS
(1) Studio Projects C4 MKII hypercardioid ~5 inches ahead
-
Right on. I'm happy to hear it worked out to your satisfaction.
BTW you may be aware of this- Since we last talked about all this here, I've updated the Improved PAS technique to extend it from two mics to three (or four if using a coincident pair in the center), and it now also includes adjustments for closer verses more distant recording positions for some of the configurations. It might be of interest to you even if only as a general guide. If you are happy with the setups you've been using then there is no need to modify them. A link to the thread with the revise PDF covering both 2 and 3 position PAS and a bit about what it does and why is in my signature.
-
Following up, I made a handful of 3-mic LCR recordings in 2024. Following Gut's advice, and against my own instincts, I placed the center mic around 4-5 inches ahead of the flanking Left-Right pair. Pics below. Long story short, I REALLY enjoyed the results. I'll post samples soon.
If you have the mounts and a four-channel recorder (not to mention the time and inclination), it would be interesting to mount one hyper 5" ahead and a second in-line with the CM4s. The time of arrival distance is so slight (about a third of a millisecond, I think) that I (and maybe others) would be curious to hear the difference.
-
^ I'd certainly be interested in hearing that.
I'd expect the difference in such a test to be subtle, mostly likely heard in the smoothness of the imaging handoff between the mics across the playback stage and possibly a difference in the perception of depth. If you do decide to make the comparison Aaron suggests, the closest data point from Improved 3-mic position PAS diagrams is this one:
- PAS angle = 100° (outside mic pair angled +/-50°)
- If close to the source, say <4m away- Outside pair spaced 54cm (21") apart and center mic or pair 17cm (6.5") forward
- If far from source, say >6m away- Outside pair spaced 58cm (23") apart and center mic or pair 20cm (8") forward.
The setup you mention with the outside pair spaced 55cm (22") apart and the center mic pushed 5" forward is already close to that. If you do decide to make such a comparison recording, I'd increase the forward spacing of the center mic a bit more to 7" or 8" if easily doable. Partly because that's what the calculations suggest is best, but also because doing that would be expected to make any audible difference between the arrangement with the center mic positioned forward verses the one with all mics positioned along the same line more apparent.
However.. If the musical performance is important to you and you have a second hyper and an extra recording channel available for it, the stronger motivation for me would be to place that extra mic coincident with the other center mic to form and X/Y pair in the center. Doing that will almost certainly make for a more dramatic difference, which in my experience makes for an even better recording.. even if it doesn't help us confirm the forward center positioning thing.
That said, the initial attempt at trying an X/Y center pair and/or making this comparison test is probably best reserved for something you don't care quite as much about and are willing to sacrifice the channel currently dedicated to mono SBD. If recording something important to you, probably best not to take chances and make the best recording you can using an arrangement you already know and trust.
/not volunteering you, but if you find yourself motivated to make such a comparison please report back!
-
I also expect the difference to be subtle, and possibly non-existent (to my ears). One of my first thoughts before even trying this was that I should do a control, just like you said, where I use NO forward spacing. But after starting to wrap my head around Gut's triangles explanation (that I posted a [crude] drawing of earlier in this thread) I really wanted to try it!
If I have time, I'll happily try to get a 4th hyper aligned with the LR mics. May not happen soon though.
I am also of the opinion that an XY pair in the center would increase the enjoyment for me. Do you place that pair forward as well?
In the samples below, I was fortunate to be close to the stage for pretty much all of them. I thought that allowed the center hyper to actually get great onstage sound from the actual amps and drums. Suuuper fun to play with in post. LOTS of ability to manipulate the mix etc. These may not be great representations of LCR recordings from the section, but it demonstrates some cool results with only 3 mics. Would love to one day replace that C4 hyper with a single mk41 cap > active cable.
https://samply.app/project/BHDbKcmpsQCtDSXnq5O3/
Track 01 is at a brewery, but it's not as bad sounding as most breweries. I ran at a front row table, about 20-25 ft out, slightly ROC but pretty close to the sweet spot.
Track 02 is the same venue but different group (except keyboard player) but they did this super quiet but juicy opening track I wanted to share. Pretty much only keys are coming from the PA, so the CM4's are PAS getting great keyboard sound, while the hyper is pointed straight at the stage, getting drums, stand-up bass amp, and horns.
Track 03 is at a small stage in an enclosed back patio and is essentially indoors; but there's a tree literally in the sweet spot (pic below). It was an improv hosted by a popular drummer, so I decided to angle the hyper right at the drums (they were off to the left) as I wanted to feature him a bit in the mix.
Track 04 was outdoors, like 25-30 ft from the stage. bonobeats got a great onstage recording that's on the archive.
These were all recorded with a Zoom F6. The files are essentially raw except for normalization and dithering to 24bit (from 32fp). And mixing the amount of center mic of course. I did some mastered versions for the artists, but I also enjoy raw files, especially when discussing LCR technique.
-
Thinking about shows that I go to that I want to record, last night I actually came up with a game plan for tomorrow night which is so similar to this. I'll use the x/y built in mics from my recorder clamped in the center, slightly forward, with a cardioid pair pas. I figure if it all fits within 1.5' 20' back from the stage and 30' high from the "balcony", I might get a good recording in a very very crappy place to record. We shall see. Or I'll do my normal which is boom some pas and add in the sbd, but I want to go sbd free on this one just to test my theory.
-
Yes, when substituting an X/Y or M/S pair for the single Center mic, just place that pair in the same position as the single mic. Unlike using an X/Y pair by itself, it will be safest to use a relatively narrow X/Y angle, say 90-deg or even less, rather than the 120-deg or so I'd usually prefer for a non-close-mic'd X/Y pair used on its own. It will be adding additional "stereo-ness" to what you are already getting from the wide-spaced pair. I like using a M/S pair partly because the single Center mic remains the same, and its more compact and easier to setup. I can point it directly at the snare drum or whatever, and add as much Side "width" afterward as I like, including none at all. But I almost always use some of it. Requires a fig-8 though, or a dedicated M/S stereo mic.
What you say about recording from on or near stage about picking up the direct sound from the amps and drums many tapers will relate to - love that pure, direct, real and dynamic sound, and the wider, well-defined instrument placement in the resulting stereo image. And I certainly relate to what you say about playing around with it afterward to find and the best sound. The increased flexibility that these kinds of properly arranged multiple-mic recording methods provide is not only a lot of fun, but is also valuable in taper situations where we need to be able to pretty much just walk in and set up, under a lot of constraints, assess things by eye and past experience, and are unable to really sound check things properly and reconfigure based on that if needed before recording.
Thanks for the photos. I love a good tree in the sweet spot!
I look forward to giving your samples a listen, hopefully this weekend.
VOF-
Grab the SBD too if you can and not a hassle, just in case. Feels really good when no SBD is needed, but also good to hedge the bet.
-
Someone said Samply is making them sign up before viewing the files, I did NOT know that (I hate that shit) so here's a dropbox link:
https://www.dropbox.com/scl/fo/w3p6up6lcfoczub7g810l/ABbAdHD5ZuoKx4-PajXb6Vo?rlkey=9qesfo54wqclgcx39z0kz3a7f&st=zjclqpk8&dl=0
Samply is pretty cool though, I personally recommend it as it has a setting for lossless playback.
Some other random slightly off-topic thoughts on 3-mic setups:
-I absolutely plan on doing more of this setup, but following along with what was being discussed in the OMT thread, sometimes it's just not practical to get all these mics up in the air. ESPECIALLY in regards to pissing people off. Gotta respect the artists, staff, and fans. I recently had a show where I brought this setup and upon looking at it, the owner of the venue expressed the slightest hesitation at it's size and I immediately switched to my low pro u853r setup (more on this in another thread). They were pumped that I was taping so it was a no-brainer to scale it down. A cardioid stereo setup in the sweet spot still sounds great.
-
-I absolutely plan on doing more of this setup, but following along with what was being discussed in the OMT thread, sometimes it's just not practical to get all these mics up in the air. ESPECIALLY in regards to pissing people off. Gotta respect the artists, staff, and fans. I recently had a show where I brought this setup and upon looking at it, the owner of the venue expressed the slightest hesitation at it's size and I immediately switched to my low pro u853r setup (more on this in another thread). They were pumped that I was taping so it was a no-brainer to scale it down. A cardioid stereo setup in the sweet spot still sounds great.
Spot on. The OMT setups are an exploration of advantageous multichannel setups that be used to great advantage when not in people's way or upsetting anyone. Perhaps useful to a majority of tapers simply as a data point of what might be done when pushing the edge of excess. A practical concession in applying it is how to achieve those things while making the setup as minimally imposing as possible, and how to further reel in its visual/spatial foot print by varying degrees without overly compromising what it sets out to do when necessary.
Didn't get a chance to check out the Samly link over the weekend, thanks for the drobox link. Hope to get a chance over the next couple days.
-
Thanks for all the advice and theory.
One day, when I win the lotto, I have visions of a couple 4060's, spread 2-3' with coat hanger wire, and an MK41 + active cable in the center for the ultimate low-pro 3-mic rig.
-
A bit off topic because this is a 4-mic question:
My current set up is one straight home-made aluminum bar; two KM143 in NOS position in the center and two DPA4090 in AB position on the outside (distance 130 cm).
Would I benefit from putting the NOS pair a bit ahead of the AB pair to link the mics better?
I have the Schoeps tool, but that is limited to three mics maximum.
-
You might. Worth a try. Maybe push the NOS pair forward by some amount, or the omnis back by some amount, and compare that to having them all in a line just to determine if doing something like that seems to be moving in the right direction or not.
Michael Williams has specific suggestions for doing so in a way such that the imaging of each adjacent mic-pair segment perceptually links up with that of the next without overlaps or gaps. Other than the Schoeps Visual Assistant which AFAIK is based on the same data but only covers 2 and 3 microphone positions, the easiest way to access those arrangements is through the interactive hyper-linked MMAD (multi-microphone array design) section of his website (https://www.williamsmmad.com/), however it currently only works for some 2 channel (https://www.williamsmmad.com/MAD/2ch/2ch.htm) and 5 channel arrangements (https://www.williamsmmad.com/MAD/5Ch/5ch_SRND_FSSC_1st.html). The others are "In Preparation". In addition to Image Assistant, I use that interactive MMAD site to inform my 3 position and 5 position arrangements (for 3 just use the 5 channel arrangements and just ignore the outermost or rear-facing pair).
However, even though the interactive website section for 4 mics is not operational, the data can be found in his AES papers. Here's one which includes 4 channels across the front stage:
2004 -117th AES Convention, San Francisco, USA - AES Preprint 6230
« Multichannel Sound Recording using 3,4 and 5 channel arrays for front Sound Stage Coverage »
https://www.williamsmmad.com/Papers/6230%20San%20Francisco%202004%20(57%20pages).pdf (https://www.williamsmmad.com/Papers/6230%20San%20Francisco%202004%20(57%20pages).pdf)
The arrangement shown below is close to your current spacings, which I pulled from page 46 of that paper.
Caveats-
1) Currently, all the MMAD arrays use the same microphone pattern for all microphone positions (using different pattern mics in various positions is one thing I intend to ask Mr Williams about). I choose hypocardioid as being somewhat close to the km143 pattern of your near-spaced pair in selecting the setup shown below. Just substitute your 4090 omnis in the wide-spaced position in place of the hypocardioid pair.
2) These setups all assume playback using same number of speakers as microphones. In my past experimentation with multichannel playback speaker arrays these setups have worked really well for that. However I also found that they "fold down" to 2-channel stereo very nicely. That conclusion is my own finding and reflects how I use these most of the time these days (mixing them down to 2ch stereo), but it is not Michael William's intention as far as I know (this is another thing I intend to ask him about). I just want to be completely clear about that.
-
As mentioned, the arrangement above is for four hypocardioids (somewhat approximating widecardioid). Below are the closest ones to your current spacings for four omnis, and for four cardioids.
I've selected all of these simply because they are the closest 4ch MMAD arrangements to the mic-spacings you are currently using. Notice that all of these indicate a forward spacing for the center pair that is pretty significant! Although all MMAD configurations use some amount of forward spacing, there are others which do not use nearly as much forward spacing and instead have the mics closer to being arranged along a line, however they use lateral spacings between the mics that are different from your current spacings. Those other configurations will produce a different stereo recording angle for the center pair segment coverage and the lateral segments. It's all a juggling of parameters required to get the linking to hand-off smoothly from one segment to the next. In all of the examples I've attached here the SRA segment coverage remains the same despite the different pickup patterns: 90 degrees for the center pair segment and 40 degrees out to of either side of that for the lateral segments.
If you go and find these particular setups in the appendix of the paper linked above, scroll up and down a few pages and take a look at how the dimensions of the setup changes as the coverage angles are altered. You might want to try a setup that keeps the mics more in a line but will require using a different spacing between mics. Just keep in mind that doing so will alter the stereo recording angles.
-
Here's an example of a 4 channel front-coverage MMAD arrangement of cardioids that uses much less forward spacing, but has a significantly wider center pair spacing. The wider center pair spacing reduces the Stereo Recording Angle of the center segment down to 50 degrees (rather than 90 degrees in all previous examples above), while the lateral segments now extend out 80 degrees to either side of that center segment (rather than 40 degrees in the previous examples)
-
To be clear, I'm not saying the above arrangements will be right or best for you. Its just a presentation of one method for arranging things which is worth consideration as it's based on a lot of past work on stereo-image linking, which has successfully informed the arrangements I use. But.. my setups that are based partly on this have all used minimal front/back spacings for a couple reasons: 1) The practical constraint of needing to support these things from a single mic-stand, along with the practical constraint of not overhanging the isles between seating rows in front or in back of the recording position. 2) I find I like to use spacing more along the L/R lateral axis and mic-pattern directionality more so along the front/back axis. That keeps transients that arrive front the stage in front aligned more tightly in all channels and minimizes front bleed into rear-facing mics. And it may be that those things are what help these configurations mix down to 2-channel stereo so nicely for me.
I'll leave it here for now. Happy to discuss further if you like.
-
Thanks Lee for this very useful and extended reply. Being both performer and taper I have the opportunity to record a piece played on the pipe organ, change mic positions, and then record the same piece again.
Compare at home by direct switching between the tracks using Audacity.
Because of your and all other good suggestions on this forum, my recordings improved a lot.
I am going to check out all this and try :D.
One question: I use my Zoom F6 both for recording and playback. The device mixes the tracks real time and there is also an opportunity to delay one pair of the tracks. Each miilisecond of delay represents about 30cm distance.
I tried the delay effect in my current set-up and it gives an increased sense of depth and space, but also the sound becomes kind of "hollow".
Would this electronic delay gives same result as a physical move of the mics?
For example the 4 mic set-up as per figure 73 as you pasted; use same locations but instead of the 95.2 cm front-back distance keep all the 4 mics in line and apply ~3.2 millisec delay.
-
If you take a look at the lower right corner of each of the configuration diagrams I posted above, you'll see it states- "No Electronic Delay needed".
Some MMAD arrays do require electronic delay, others don't. When they do, the electronic delay indicated isn't going to correspond directly to what would be needed to compensate for the offset distance between microphones if one were aligning the waveforms of all channels for a wavefront arriving from a certain direction (say directly from the front). Electronic delay is instead used as another tool in addition to the pickup-pattern, spacing and angle between microphones to "steer" the edges of each SRA segment.
If you peruse the MMAD papers or the hyperlinked MMAD section of the site for the 5-microphone arrays you''ll come across some arrays that do require delay applied to some of the channels for the image linking to work properly.
First I'd try setting up the mics using the indicated physical locations.
Then after listening to that, I'd probably play around with delaying the front channels of those recordings on the computer by various amounts just to see what that does.
Then you might try moving the mics into a different arrangement, say with the pairs spaced less far apart, and listen to that without delay and then with it to "sorta/kinda" emulate them being spaced farther apart.
BUT! Know that electronic delay is not the same as the use of physical spacing, especially in these arrays. That's the "sorta/kinda". Delay is one-dimensional. These mic arrangements are two-dimensional. And the room is three-dimensional. Application of delay can only line things up for wavefront arrival from a single direction, and with that achieved the arrival time difference for a wavefront arriving from any other direction is not going to be aligned. What's more, the difference in alignment will increase as arrival shifts farther and farther off that axis until reaching a maximum for a wavefront arriving from the opposite direction, in which case the run-time delay from the physical spacing and the electronic delay will add together, doubling the timing offset instead of cancelling it out.
That geometric reality applies for any sound source bouncing sound around any non-echoic room, but I expect it to be especially relevant for pipe organ where the room is essentially an extension of the instrument, with wavefronts arriving from all directions and strong modal nodes distributed in 3-dimensions.
I do want to repeat that these arrays are actually intended for playback over quad, 5-channel etc, playback systems where the number of playback speakers equals the number of mic channels. As mentioned, I've found the arrays that I've adapted to my own use work really well for 2-channel stereo mixdowns too, which is why I'm suggesting they might be worth a try. But it also means the more significant front/back spacing between pairs in these particular arrays, which tend to be larger than what I'm using, throws things up in the air a bit too, in addition to the application of delay.
All I can say for sure is try it, play around with it, and please let us know how it works.
Oh yeah, almost forgot to discuss one important particular of the F6's channel delay function.. The F8 I use has the same capability. I was initially excited to play around with that during playback when mixing on the recorder itself, but unfortunately the delay can only be applied during recording and not during playback. That makes it far less ideal for experimentation when figuring all this stuff out in comparison to applying the delay on the computer. You'd have to make lots of recordings with different delay settings instead of just changing the delay afterward as desired. It would be applicable if after playing around with delaying things on the computer you determine that you definitely want to use a certain delay all on certain channels of whatever array you settle upon using regularly - and it would only be applicable to that particular array. In that case applying the delay while recording would conveniently eliminate the need to do that afterward.
-
I have tried numerous configurations using four mics. Looking around on Taperssection and Gearspace, lots of people did the same, apparently with no conclusive outcome what is best. This gives a very useful tool to get more grip on the situation.
I am really curious about the mathematics behind all these figures.
About delay during playback on the F6: I can do it via the remote function on the Zoom app on my IPhone. Press the button "pfl" at the track you want to delay during playback and then select "Input Delay" (0...30 msec could be selected).
-
Hmmm. Good to know you can do that on F6. I'll have to try the Zoom app with the F8 and see if I can do the same. Tried the app early on, but haven't used it since then, and never tried it for playback.
Not overly surprised there isn't a solid consensus on the best way to use four mics for recording pipe organ. I think most play around with it until they find what works. Probably varies a good bit with the particular organ and room.
[snip..] For example the 4 mic set-up as per figure 73 as you pasted; use same locations but instead of the 95.2 cm front-back distance keep all the 4 mics in line and apply ~3.2 millisec delay.
^
My posts above weren't meant to dissuade you from trying this, but rather to explain what's going on with the Williams arrays. Yes sure, try that and play around with adjusting the delay during playback to best effect. Actually, what you suggest is more in line with what I do than the set-ups shown above which use significant front/back spacings.
That's because a general guideline of mine for spacing the microphones in my multichannel arrays (please forgive me for repeating myself as this has been stated in the OMT threads probably several times) is to rely on spacing along the left/right axis more so than along the font/back axis. I like using some amount of spacing along both axes, but rely more on microphone directionality along the front/back axis - pointing directional mics forward and backward, without spacing them that far apart. That's based on several things. One is that I find my rear-facing mic channels much more useful when they exclude as much front content as possible. I found this to be true regardless of whether I'm playing the recording back in multichannel surround or mixing it down to 2-channel stereo. Loud sounds arriving from the front otherwise obscure the quieter stuff arriving from the back which the rear facing mics are intended to pick up. Even if I'm not using much of that rear-arriving content, its much more useful when its relatively isolated to the sound arriving primarily from the rear, and achieving that requires the use of microphone directivity rather than spacing. Another is time alignment of transients, which is something that is arguably important for stereo mix-down and not multichannel playback. Using less front/back spacing keeps the transients from the front content that do leak into the rear-facing channels more closely aligned with the front facing channels, incurring less time domain smearing when mixed together. A third is simply practical - its less of a challenge for me to deploy an array that is spaced widely in the L/R axis than one which has significant spacing along the front/back axis. So I gravitate to the Williams arrays that have less front/back spacing to begin with, and end up modifying those to fit what I'm doing.
All that said, some of it isn't applicable to what you are doing! Even though sound is arriving from all directions, especially so in a pipe organ hall, all four mic channels of your array are intended to be fully sensitive to sound arriving directly from the pipe registers in front, as well as to sound arriving from other directions in the room. You aren't trying to exclude foreword arriving sound from rear-facing channels, but rather picking up sound from all directions in those wide-spaced omni channels. The time-alignment aspect is probably more applicable, although you probably have more leeway there than other stereo mix-down applions just because of the nature of the instrument - its not super percussion rich with lots of sharp transients but rather more timbral, tonal, spacious and room filling. Some time smear may actually be to your advantage.
Hope that helps more than confuses. My apologies to others reading who find these wordy posts something of a bother.