I suspect most of us will never have the good fortune to be asked to mic up such a wonderfully mahoosive instrument like this, but watching the incredible Nakibembe Xylophone Troupe playing the Embair at the Berta Berlin CTM Festival in 2020 it got me thinking this would be quite a challenge to mic up and record!

Here we see a basic staggered line of I think 5 or 6 SM57s covering the length of the instrument.

So as a bit of a fun thought-experiment because it's so unusual, what would you do, let's say if you were asked to engineer a whole album of this?



And enjoy the fantastic playing in any case - especially if you are a Steve Reich fan and can hear the obvious root influences on display here for pieces of his like Drumming & Six Marimbas etc.

More info about the instrument itself here: https://www.songlines.co.uk/features...box=1655110749

I work folk festivals where people show up with all sorts of strange African and Asian (and sometimes eastern European) instruments. I stick a finger in one ear and listen with the other, moving my head around and see if there is any place close up where it sounds like it does from a distance. Sometimes it's not possible and adding multiple mikes (with comb filtering!) is needed for larger instruments like giant gambangs. Sometimes you can pull back and mike the room or area! Sometimes the load-in took too long and there's no time to do anything other than set up a mike where it looks okay and hope for the best....
--scott

When we were kids, my brother and I created something we called Paul Bunyan's Xylophone one day when we were splitting firewood: we split a bunch of logs and laid them out in a 15-foot row from largest to smallest and ran up and down with sledgehammers knocking out tunes.

Check out the use of RE-20s in this thread: Oldschool ECM Recording - Making Of

Three mics spaced evenly across the instrument would be about 3'9" apart; for zero combing on the center mic, they would need to be about 15" from the instrument, to be just within the 3:1 rule.

Since the L and R mics would be over 7' feet apart, they could be panned in a bit to prevent an over-wide image.

In actuality, the center mic would probably not need to be mixed at same level as the outer two, so they could all be backed up a bit more.

Also, since very few people are familiar with the instrument, a tiny bit of comb would not be a problem.

i've been recording indian, african and european percussion ensembles for decades, starting with 'les percussion de strasbourg', conducted by pierre boulez ca. 1978.

my reference remains to be the large marimba as used in western/classical music; x/y, m/s or l/c/r from above - or then a pair (or trio) of blm/pzm's from underneath have been my choice.

imo there's no need for multiple dynamic mics unless one likes the slightly copressed sound, the lack of air band or one has to go very close to keep out nearby instruments.

Milking it would depend if it were part of an ensemble, or the solo instrument.

I'd spend time with closed-back headphones and a cardioid mic on a boompole, swinging it slowly from directly overhead to directly out in front, and same height off the floor as the instrument (ie 90 degree arc)
This will tell me if there's any sweet spot of propagation, or zones/nodes of dips or losses in response.

Once the sweet spot is found, I'd put an ORTF or XY main pair there to anchor the overall image and sound, at a suitable distance so that detail and room ambience were captured..., and if necessary add spot mics (perhaps including Omni or PZM mics on the floor underneath) to capture full frequency extension and 'excitement' of the instrument.

Hell, I'd add an AB Omni (or sub-cardy) pair overhead too...recording channels are cheap !

I agree this would be ideal.

There are so many variables here of course, starting with the orientation of the whole instrument - would it be parallel to the audience or at 90º? The article also mentions: "When played in the villages of Busoga, the keys are laid on banana stems and placed over a 50cm-deep hole dug into the ground. This trough provides a booming resonance that enhances the instrument’s bass keys, connecting the embaire to the land of its ancestors in a very literal way." So a decision on that would seem to be required as that might affect a mic placed underneath, or not as the case may be.

Of course, these would mostly be decisions for the hypothetical producer of our hypothetical solo Ebaire album!

Thinking of a 'conventional' vibes-type set-up on a stage... I think my immediate thoughts would be two things along the lines of what @ studer58 points out, i.e.:

1) Will a single stereo pair cover the whole instrument successfully, and if it does how to find the sweet spot with 6 equally important players spread out along the length of the instrument?

2) You have players on both sides of the instrument, so a main pair out front might end ups getting some dips as they move about. Again, one could either go with that being 'how the instrument should sound anyways' or take steps to mitigate it for the purposes of our album. Not sure.

I agree that AB overhead omnis might be a good starting point, unless there's such a thing as ABC omnis in which case maybe that!

---

Indeed. I thought of Daniel's vibes and his @ idee und klang clip there - if nothing else, a long line of 6 x RE-20s overhead would look pretty cool!

"ABC" - three spaced mics is exactly what was suggested in post #4 .

Three spaced omnis is certainly a thing - usually referred to as 'M3'; the technique used for the Mercury recordings, and the first few years of Telarc.

Curtain array, 3 wide cards or omnis depending on context, maybe 6-8’ total spread, 3-4ft out.

How about Blumlein angled down from above and 1.5 meters away? Should capture the entire instrument within a 90 deg spread.

That is too close for such a wide instrument. The front/middle of the instrument will always sound less distant as the lower/upper registers.

Either have more distance, or use 3, hell even 6 mics. The problem is, getting really close to be OK with the 1/3 rule is impossible, as there are players everywhere.

can't believe anyone is still even considering the 3:1 'rule' in the digital age...

While I am not a strict adopter of any rule, you just explained why so many vocal + ensemble recordings sound as if the vocalist is standing in a box, or in front of a big glass pane.

Or do you have any digital tricks to solve this ?

well, i can't stop some folks from sticking to a recording but then not learn to mix...

anyway, ANY spaced setup - not only mics of an a/b, l/c/r or 5.1 setup but also multiple mics used as spots, mains and ambis - yields frequency areas with low correlation; there is no single ratio that would prevent all issues (but one can use a specific spacing to its advantage - that applies mostly to lf though).

however, there's indeed a digital trick to correct all issues created by any spaced setup (which i keep mentioning to no avail it seems):

all-pass filters (aka apf's) - to adjust them and to find areas within which issues occur, you need a fft as well though.

Maybe you can educate me, but if you have a spaced main mic, with a soprano solo who is quite close to the mains, and has a spot which is needed in the mix, but her level is almost enough in the mains : what are you going to do with alpf if she moves just a few inches ?

This example goes for a lot of situations, especially with not so directional mics.

Personally I can solve most issues with good placement and directionality, let’s not call it the 1/3 rule, but it IS related.

I need zero filtering afterwards, so I also don’t need to worry about how they sound.

looks like you're not an experienced user...

the interesting part about apf's is not the delay but that the frequency area within which the correlation issues occur remains constant (also in the example you just mentioned, although i'd never position my main mics in a way that the soloist will be almost equally loud in the mains as in the spot!)...

...and the thing about applying any tool - apf's in our case - is that it gives you options but of course you can chose to neglect them and pretent that the best possible sound always stems from mic selection and positioning alone.



[funny btw that live sound (which many demonize as being inferior to location recording, specifically of classical music) has long adopted apf's as an invaluable tool for optimum performance - anyway, what works on the way out also works on the way in which is why i'm often using apf's when recording; not on all mics of course but quite frequently on the most important couple of mics - IF i think the recording/live mix or on-air mix can profit)]

No, you’re right, I am very inexperienced.

Did you ever record a powerful solo voice, live, with a rather large ensemble ?

Did you ask the soloist to stand off stage ?

inexperienced regarding the use of apf's...

anyway, you're not arguing with me but with the laws of physics; if you don't see any value in using apf's, fine - i wouldn't wanna miss them.