Miking a string quartet live

[Junior8]

I simply gave up X/Y coincident pairs long ago - I (and many of my clients) just hate the flat-earth sound I get from them, whatever mics and rooms I've recorded in. It's just too... mono.

 

For this purpose is that such a disadvantage? Back in the sixties I recall the late Ted Nurse of Gloucester regularly recording choirs with a single AKG on a very high stand, straight on to a Nagra IV monitoring it over headphones. Provided you selected the right room and did a few trials with positioning the results were ideal for this sort of purpose.

[Oldradiohand]

For home/hifi playback, for me it's ORTF every time these days on things like this in good-sounding rooms. Bad-sounding rooms, especially when time is short, simply get the ORTF mic placed higher-and-closer than I'd like for an "ideal" room to minimise capture of the stuff I don't want to hear, often making the room itself mostly inaudible if the mic is close enough. Sometimes I might even use a small foam panel (usually the one from the box the mics came in) somewhere near the mics to screen off the reflections from the area of the room I least want to hear. Yes, the proximity to the source means the imaging is wider than I'd like, but that can be fixed later, if needed. Then I can apply careful amounts of reverb in post - usually based on impulse measurements of rooms I *do* like for the kind of material. If mono-compatibility is an absolute concern in the brief I might go for M/S techniques instead. Taking careful note of the approximate (or even actual) dimensions of the room and number of people present at the time can help with this, so that the reverb applied in post can be tailored to better match what was there on the day.

 

And if the room's *really* bad, I'll go ORTF AND close-mic. I can usually salvage something from the room that way.

 

I simply gave up X/Y coincident pairs long ago - I (and many of my clients) just hate the flat-earth sound I get from them, whatever mics and rooms I've recorded in. It's just too... mono. Push the angles further apart to compensate and you start picking up more crud you don't need, and opening up a hole in the middle. Use hypercardioid mics, and you start picking up stuff from behind. And woe betide you if you have a soloist who then deigns to move during the piece. Imaging collapses as the original positioning (therefore timing, therefore comb-filtering) relationship changes. If you catch it in a rehearsal you pop a spot-mic on them so you can blend it in. You then think you can time-align, except you then realise they're moving in three dimensions and you now have three sources to keep track of, so you adjust the timings again and they move again, and... gaaaaagh...

 

By contrast, I find ORTF gets me the placement cues without the extra work of other techniques. To my ear it seems the distance between the capsules forces the comb filtering to occur at frequencies that are less distracting. I say "to my ear" because while I *could* (and know how) I've not done the math to prove it yet, so I might be looking at the wrong reason for what I'm finding. But practically I find that if my "lead" moves around, the image shifts in a more realistic way, and doesn't mess up the treble as they do so. In practice I get plenty of options to fiddle with imaging in post if I want to, and this combination gets me lots of happy clients. And it even seems to work well in a surprising number of live reinforcement gigs, too. Collapse it down to mono with a simple downmix and I find a simple EQ tweak is often enough to get the balance back on most mixes. So much so that I now have a simple M/S split EQ setup in my mastering software rig to deal with that automatically.

 

MS gets me plenty of technically-correct options to play in post, and results in lots of happy technicians (especially those with backgrounds or dayjobs in broadcast) due to mono-compatibility. Strangely however, by the time the work has been through post, the creatives always end up happier with the pure or post-enhanced ORTF approach even when I've set up the MS processing to mimic (again in post, by going back to the maths and physics) what the ORTF actually would have done on the day. Perhaps something is lacking in my modelling because I'm not able to model the frequency-response differences are of typical cardioid capsules vs Omni vs Fig-8?

 

The Superlux S502 is hitting me as a great way to do ORTF on a budget without making it look like something that came out of Dr Frankenstein's lab. Feed that into a mic-level recorder of choice (I happen to have a few laying around that give phantom) and you're away.

 

Finally, for anything that's going to be played back a reverberant space, close-mice'd multitrack might be a way forward, so you can fix the balance later, either at playback, or at least on mixdown away from the live recording environment.

 

I understand what you mean about bad rooms and I would advocate closer miking to overcome this - it would probably be an idea to record a backing track in a deader acoustic and then add some reverb if required depending on the playback venue.

 

Not sure why you're getting comb filtering on a coincident pair, a half wavelength cancellation at 16kHz is a centimetre or so - it should be easy enough to get the vertical coincidentally closer than this. At about 6" you'll start to get cancellations at about 1K which is one of the reasons you have to use cardioids to avoid common pickup .

How are you mixing ORTF and M&S? I can't see how a near spaced pair would work with processing designed for a coincident technique....

[Solstace]

How are you mixing ORTF and M&S? I can't see how a near spaced pair would work with processing designed for a coincident technique....

 

I'm not, except that all stereo signals can be boiled down electrically to a Mid (M) component and a Side (S) component. So I've started keeping a M/S split tool in my DAW signal flow as a way to process those mid and side components separately if I need to. Then recombine to L/R again afterward. It's an exceedingly quick way to both check mono compatibility and adjust stereo width without reaching all the way over to the master section; likely on another monitor or even another device in my regular rigs.

 

I did work out an approximate fudge to get a blended ORTF-like sound out of an M/S mic set, figuring it was easier to introduce new artefacts into a coincident sound than it was to pull them out of a non-coincident sound, but will need to refer to my notes on how I got there and why I even tried it. Think it involved something like overdubbing a church organ to an existing orchestral recording. Got most of the way there.

[Oldradiohand]

I see what you mean about expanding stereo width using M&S conversion - I've experimented myself with but wasn't convinced by the results unless the source material was coincident or pan potted stereo. I've only used MS to XY conversion in anger to convert MS recorded signals from coincident MS pairs.

[Bazz339]

I'd also have a look at the Tascam DR40.

 

Anyone have/use one?

 

They have mic in on XLR with 48V phantom and are not overly expensive although probably double the DR05.

[paulears]

I'm still not convinced that a CD style recording is the most appropriate as a track to play - most of the lovely details being discussed here apply to recordings for distribution, not show tracks - where very often they end up in mono because of the PA speaker placement. String quartets sound very, very strange one sided, and with 1st violin firmly left and cello right, most times you'd be reducing width, probably severely so!

[sam.spoons]

I think we all started from the premise that the recording was for a CD that would be listened to rather than a CD which would be used as music for a live performance. Obviously the two require different techniques to achieve the best result. As PaulEars suggests, mono compatibility is pretty crucial for the 'music for performance' result.

[Jivemaster]

So for a show music track would it be better to record in mono? One mic?

 

As the troupe goes out to lots of venues, what is the preferred primary music media, what are the essential secondary ones.

[azlan]

my preferred primary music media would be something like a laptop & an ESI gigaport, meaning that you could spit out a mono/stereo mixdown (possibly taken from a co-incident pair) where channel counts were limited, but also add spot mic'ing to the mix to make sure that you get the level of detail that you need, and to help control reverb if you are playing in a very live room (if you already have a fairly reverbarant recording, and play it back in another room which is adding more reverb, it could get very muddy very quickly), the playback would be triggered using something like multiplay or qlab and sent out of the various outputs into the desk for actual mixing with the live act.

 

 

although, I should say, this is me coming primarily from a rock&roll background, so my way of doing things may not give the most natural results for that style of music

[sam.spoons]

I'd say a CD player is the most likely playback device. So take a CD (and a spare) mixed/mastered fairly dry (at least no extra reverb added) and use a mono friendly miking technique (see above) when recording. I'd also take the same tracks on an iDevice or other .mp3 player and a collection of cables and connectors to cover all eventualities (mini jack, ¼" jack, phono/rca and XLRM as a minimum but be careful about getting 'balanced' and 'stereo' inputs mixed up). Finally, I'd have a backup playback system of some kind in the car, even if it's only a big ghetto blaster, that way the show can go ahead even if the venue PA is completely useless.