Hey, gang! Ah, never a shortage of new challenges to master, right? The symphony I work with will be doing the Abe marimba concerto, and while I've recorded lots of mallet instruments in solo and other configs, and many concertos with more "traditional' instruments, never marimba with orchestra. The two videos I found with clear views of the mics were radically different. One had a pair of Schoeps about seven feet up and splitting the instrument in thirds, while the other had a single mic I couldn't identify dead center and inches from the bars. (Possibly an XY or Blumlein capsule?) My first inclination would be to go with a pair of wide cardioids from above, in intentional stereo rather than the usual closely spaced parallel config. But that could be weird against the orchestra imaging, and obviously wouldn't isolate the marimba well. And close miking, while the isolation might be useful, doesn't make a lot of sense to me on such a beast of an instrument.

Of course, in my experience, too much soloist from my main array is usually a bigger problem than not enough, but one never knows...

Anyway, as usual, I welcome any and all experiences and opinions from all y'all.

i've successfully been recording (and amplifying) marimbas with close mics in a l/c/r+lfe configuration. 3x mk-4 and a blm-3g, the latter with a steep lpf at ca. 80-100hz.

I'll bet that sounds really cool for a solo marimba. They're certainly large enough and deep enough to justify a 3.1 recording. Do you think you'd use the same setup with an orchestra close by?

i couldn't find any reason to change the setup in a orchestra setting - well, except that in the studio (and for solo marimba), i added a pair of rear ambis (for 5.1).

Of course. How close, and at what angle are you placing the close mics?

If I wanted isolation, here’s where I would try an A/B pair of fig-8s over the marimba (pointed down). Ribbons and a bit of proximity effect might bring out the low-end naturally, and the nulls will reject the orchestra (unless there’s a low reflector/shell over the ensemble).

depends mostly on the music stand(s)/sheets but if given the choice, height can vary from 125cm to 175cm, angle ca. 45 degrees, mics pointing towards the center of the instrument; blm either centrally positioned underneath or towards the low register.
on some occasions, i had to go closer (and lower) and could use but a stereo pair (plus a blm) for which i used a pair of mk-21's in wide a/b, spaced ca. 1m iirc.

Sean - considering the orchestral context, you will want the marimba to occupy a specific location in the stereo image, and not try to be some huge LR spread. I would suggest a relatively narrow AB pair of cardioids, sort of like what you might use for a vocalist, but a bit wider.

Some intriguing ideas I hope I get to try in a more controlled environment someday. For this concert, we'll have about two minutes to place mics between pieces, so combined with the issue of sensible stereo imaging, I'll probably go with the usual parallel, spaced cardioids (or wide cards). This will be fast, and will allow me to place the stands at the ends so as not to be in front of the performer.

A marimba is so wide, if you get quite close, the edges will sound distant. The same goes for the CDEFGAB notes, they can sound further away if you cannot get above the instrument.

If you need to get closer for isolation, especially with fig8 you need at least three mics. Please note this goes completely against my normal micing practices, but a marimba is a special case.

So, if you get close, a narrow AB will highlight the middle of the instrument. This can work if the music does not utilise the full range of the marimba.

I assume we are talking about a bass marimba ?

Agreed, it's an unusually wide instrument, so I'll space the mics quite a bit further apart than the usual very narrow spacing I use for string or wind instruments. I'm not overly concerned with isolation, as I'm sure I'll get plenty of it in the main array. It's a full-sized marimba, and a virtuosic piece that uses every inch of its physical range and every dB of its dynamic range. The spot mics might come into play when our guest soloist goes down to ppppppp. Here's the piece if anyone's interested:

https://www.youtube.com/watch?v=zWxvz6_swdk

Think of it as a miniature choir or wind section of an orch - Best start is to place two cards at the 1/3 and 2/3 points of the width of the instrument, and place no closer than the distance between them.

Three mics would only be needed if it is necessary to place them closer; in that case, still place each no closer than the distance between them.

Again, these are starting points.

In this situation you’d want to pay attention to potential comb-filtering ie the 3:1 rule…..or ignore, if you prefer to sidestep inconvenient truths

https://www.dpamicrophones.com/mic-dictionary/3-1-rule

https://www.dpamicrophones.com/mic-u...ow-to-avoid-it

There's theory, and then there's practice - and just plain listening to results.

Taken literally, if you 'do the math' - many things described in the Decca book violates the 3:1 rule. Three to four mics across a choir violates it, the Tree itself violates it!

Amazing all those records are even listenable.

And how does two mics at the 1/3 and 2/3 width of the instrument violate the 3:1 rule?; the mics are not "panned to the same position", to quote the DPA page.

The same source also notes that the effect is likely to be ameliorated if directional mics are used, as they are not strictly sampling the same piece of air (or section of instrument), and combing itself (particularly if mic levels are not equivalent) is often difficult to detect...short of a thinning of timbre, or 'body' of instrumental (or vocal) sound.

It's arguably part of the 'charm' of the Decca Tree that mic in-panning and spacing gives rise to ambiguous and cloudy phase relationships...but mic spacing is of more relevance to the instrument under discussion....it having somewhat comparable dimensions to a grand piano. As outlined below, not being panned to the same position doesn't negate out of phase relationships.

Here the recent Decca Book states:

"Another important aspect of the microphone spacing is the impact it has on the amount of out-of-phase components that are present in the sound, which is always a potential problem with spaced microphone techniques. The different path lengths taken by soundwaves to reach each microphone will result in some phase differences between the left and right signals, and their significance depends on the wavelengths involved.

For example, the wavelength of a 1 kHz signal is approximately 34 cm (13″), and this is the kind of distance by which we are separating the microphones. The wavelength of 100 Hz is around 3.4 m, so a microphone spacing of 30 cm (12″) is less significant for these frequencies.

Phase differences for individual notes can be seen quite clearly if you get a pianist to play individual notes one at a time and look at the results on a phase scope. Some notes will be in phase (predominantly vertical display on the scope), some will experience an intermediate phase shift (generally circular display on the scope), and others will be distinctly out of phase (predominantly horizontal display on the scope). It is these latter ones which are potentially a problem.

With spaced microphone techniques you can never get rid of them altogether, but if there are too many then it will start to affect the overall feel of the sound, and it is a good idea to reduce (or at least control) the number of notes that are distinctly out of phase.

In a test performed by Mark Rogers, all 88 notes on a Steinway D were played individually and recorded simultaneously at three different microphone spacings: 20 cm (8″), 40 cm (16″), and 60 cm (24″). The results are shown in Figure 5.11, where it can clearly be seen that the number of notes that are distinctly out of phase increases along with the distance between the microphones, while the number of notes that are clearly in phase decreases. The out-of-phase notes tend to be at the higher end of the frequency range, as the lower notes have wavelengths that are too long to produce this effect.

Having a few out-of-phase notes does not adversely affect the listening experience, and again this suggests that a spacing of around 12″ (30 cm) is most likely to give a satisfactory width without too much in the way of out-of-phase effects"

If the stage placement resembles that of this video: https://youtu.be/ssUZdLuLsqg
...you'll presumably be running a main pair plus outriggers (flanking mics around 8-12 feet either side of main pair)...which would give you a very wide, predominanly left-heavy ...yet 'cinematic' and reasonably focussed main-mics sound already ? In this setup you may not even need spot mics on the instrument at all...

Without flank-riggers you'll have an even more left-heavy balance, with less detail...putting more emphasis on your spot miking.

Something like this represents typical spot miking for solo marimba in a concerto setting: https://youtu.be/zWxvz6_swdk ....although you could also get away with a single solo spot in the centre of the instrument (to provide the transient attack/localization information) if using sufficiently close flankers plus main pair... to contruct the width aspect of it

Tell us what you have planned as your main/overall front-line miking for this concert ?

Thanks. Do you have a reference for this test ? For me this illustrates brilliantly why I do not like spaced techniques on grand piano.

op can only hope that - as for any 'percussion-heavy' score - no spaced main mics will get used...

(...and some comments here make wonder if some of the posters have ever at least seen a marimba, let alone heard, recorded or even played?!)

I would argue noone here has ever succesfully recorded a bass marimba.

It is quite impossible.

For starters, the damn instrument always sounds dirty and out of tune (very strong fourth harmonic which clashes madly in eg. minor chords). But only when recorded, standing next to the instrument it suddenly sounds clean !

Also, the sheer size and weight & volume are almost impossible to playback, even when the recording is successful.

No direct reference, it's quoted in the piano section (5.3.3) of the Dunkerley et al book on Classical recording techniques, which is an excellent document if you don't yet have it, and Mark Rogers himself is a co-author !

https://www.routledge.com/Classical-.../9780367312800

However you could probably contact him directly to track down the source of that research...presumably it's been published: http://mark-rogers.com/index.php/biography

...only to the point that it is impossible to record any instrument 'accurately' but a marimba indeed has highly distinctive (if not unique) characteristics.

Thanks, I may have to order that book. I also have to re-order some books that I have lent out and never got back ...

There should be a special place in hell reserved for book borrowers who don't return books to their owners.

I hadn't seen that version of it. Interestingly, I couldn't see any mics in the video at all, and definitely no marimba spots.

We've been making an effort to reduce the number of stands on stage while still giving the recording more of a live feeling than an intimate "studio" sound, so the past few concerts I've opted for a kind of modified "Other Faulkner" array, with a pair of ORTF 8040's in the center and CMC65's in omni at 67cm on the outside of a single bar behind the conductor at about 12' high. I've been pretty happy with this so far. The 8040's have great reach and focus for the winds, brass and center strings, while the slightly wider Schoeps (wider than several of the other four-mic arrays) give the desired width and ambience with some emphasis on the first violins and cellos. So yes, the marimba will be prominent in the front array, and left-heavy—which is okay since that's where she'll be. Again, I don't expect to need much from the marimba spots, maybe riding them a bit during the extremely quite parts of the cadenzas. I always use some parallel compression in the mixes to pull up the soft sections anyway, since these recordings often end up broadcast on PBS affiliates.