String quartet - 3 condensers, some 57s\58s

[forumuser840717]

A few thoughts:

Everyone seems to be answering this as if what's needed is a way to record a 'conventional' string quartet in conventional layout, playing a classical or pseudo-classical piece that works acoustically. And assuming that the quartet can balance itself appropriately.

Is the genre a 'normal' classical/performance piece where everybody aims to produce a balanced quartet sound, acoustically, in a real space, or something needing individual control over levels and perspectives or wanting to allow different FX or non-real panning to be applied to each part?

Or where a drier sound is wanted either as a starting point in a mix or for a more intimate perspective, e.g as string parts in a pop/rock/whatever track?

If it's the former, then my normal starting point for a session type recording with no restrictions on positioning of players and mics in a room that's appropriate for the programme, genre, and desired sound would probably be a pair of omnis on a Jecklin Disc with the option of a spot mic or a stage riser for the cello as a way to adjust any small balance issues. Or, if a tighter stereo image is wanted, either a Blumlein 90 degree fig-8 pair or, more likely a Blumlein style MS pair of fig-8s, again with the option of a cello spot or riser. But only as a starting point; ultimately anything is possible from a single point stereo (or mono!) mic to multiple things scattered all around the room.

If going with a pair of cardioid SDCs as the main pickup, I'd probably check out something like NOS or one of the others which has a smaller separation angle between mics, or the Gerzon shuffled close cardioids.

I've often found that, depending on the mics and how the room influences the pair position, ORTF can emphasize level/clarity differences on the outside players at the expense of levels/definition/imaging on the already slightly physically recessed inner players. I know that the inner parts/players being further away in reality is something for the quartet to understand and accommodate in their playing, and that convention would have the outer players perhaps slightly more 'up front' but recording very rarely resolves such details completely as they really are, instead exaggerating or recessing differences and perhaps introducing other issues.

Cardioids with really good, smooth, off axis performance are more important in ORTF than for techniques which place the middle less far off axis to the mics. On some quartets, programmes and rooms this is more of an issue than others.

I don't know the se08s but if their off axis performance is less than examplary it might be worth trying something less angle dependent than ORTF.

Of course, in a session scenario, there's always the option (within a limited window) of adjusting the relative positions of the players to fix small balance issues.

If a reasonably well optimised MS pair is available, I'd probably try a version of that as it reduces potential issues with on or off axis frequency/phase responses and perhaps matching between the cardioids. On the other hand, if the only way to do MS is to bodge a Fig-8 from two separate cardioids, while possible, it's far from ideal and I wouldn't bother.

[Hugh Robjohns]

... the JS Mid-Side Encoder and JS Mid-Side Decoder built into Reaper....

I'm intrigued now!

I have and use Reaper, but I've never used those JS plugs just because I'm familiar with MSED. I will investigate but I'm not near it right now...

What's the difference between the JS encoder and JS decoder? The equations are the same for both operations, so you just need a single matrix. If MS goes in, LR comes out, and if LR goes in MS comes out, so why separate encode/decode plugs?

[James Perrett]

Reaper offers a few more options on the decoder like swapping channels as well as adjusting the position and level of the centre signal.

[Hugh Robjohns]

Okay, thanks James.

[Carmelious]

THE RECORDING WENT NICELY as such:

ortf se08s at 90*
+cello close mic sm57
+just because it was there - far center nt1

will update with the outcome. THank you!

[Arpangel]

What track do I put the decoder on? the master?
Or just the mic track? I take it the figure eight goes on just one channel?
I’m just confusing out of phase etc, as in the hardware set-up maybe.
Do I have to use the JS Encoder and Decoder together?

[James Perrett]

There are two ways of using M-S. You can either have an M-S source as we are talking about here with microphones or you can have a normal stereo source which you want to process with M-S. The encoder is needed for the second scenario where you want to convert from normal stereo to M-S.

If you already have an M-S recording you just need the decoder but, as Hugh says, the encoding and decoding processes are identical if the decoder settings are left at their defaults so you could actually use the encoder to do the decoding.

The Reaper M-S decoder expects both the M and S signal in the same audio stream. This can either be in the same file or, if recorded to separate files, you could combine the two tracks, putting them on opposite channels in a folder track and put the decoder into the folder processing.

If you want to process everything as normal stereo then put the decoder at the start of the processing chain but any processing that you want to do in M-S needs to go before the decoder.

[Hugh Robjohns]

What track do I put the decoder on? the master?
Or just the mic track?

Stereo mic channel.

If you were recording normal Left-Right stereo, you'd bring the two mic signals into reaper on a stereo track.

You do the same with Mid-Sides — bring the mics into a stereo channel — but you need to convert to Left-Right to mix with other signals, so you drop an MS decoder in as the very first plugin on that stereo channel so that your MS mics are immediately converted to left-right and can be used in the normal way.

The decoder pluhin will have controls to adjust stereo width and, possibly, other parameters too.

I take it the figure eight goes on just one channel?

Yes, channel 2, with the mid mic on channel 1.

I’m just confusing out of phase etc, as in the hardware set-up maybe.

That's the beauty of DAW plugins. All that complicated channel duplication, polarity inversion, panning and routing all goes away. The plugin does it all for you. MS goes in, LR comes out.

Do I have to use the JS Encoder and Decoder together?

No, just the decoder. The signals are already encoded as MS because of your mic array, so you just need to decode back to LR.

The encoder is for when you have a normal LR stereo track that you want to process as MS (and then you'd use the decoder after the processor to return to the LR format again.

[forumuser840717]

THE RECORDING WENT NICELY as such:

ortf se08s at 90*
....

Good that it went well but what's ortf ____ at 90*?

[Wonks]

Indeed. ORTF is a very well defined mic arrangement, with the mics splayed out at 110° with the capsules 17cm apart. Anything else isn't ORTF as you aren't mimicking the way our ears hear things (without the use of a dummy head).

[Hugh Robjohns]

Good that it went well but what's ortf ____ at 90*?

You and me both, eh?

For the benefit of others, the French ORTF mic array has a very precise specification: 170mm capsule spacing with a 110 degree mutual angle.

If the stereo mic array doesn't have that exact spacing and that exact mutual angle, it ain't ORTF... it's simply a near-spaced stereo pair of your own design.

For reference, there are several other precisely defined near-spaced mic arrays, such as the Dutch NOS (300mm and 90 degrees), or the German DIN (200mm, 90 degrees)... etc

Varying the capsule spacing and mutual angle alters the stereo acceptance angle and the imaging, and altering these parameters along with the distance from the source helps to find the optimum sound stage and aural perspective. So there's nothing wrong with altering the mutual angle or capsule spacing of an array to achieve the desired sound... just don't call it a defined array when it isn't because it causes unnecessary confusion (and sometimes mirth )!

[Arpangel]

Good that it went well but what's ortf ____ at 90*?

You and me both, eh?

For the benefit of others, the French ORTF mic array has a very precise specification: 170mm capsule spacing with a 110 degree mutual angle.

If the stereo mic array doesn't have that exact spacing and that exact mutual angle, it ain't ORTF... it's simply a near-spaced stereo pair of your own design.

For reference, there are several other precisely defined near-spaced mic arrays, such as the Dutch NOS (300mm and 90 degrees), or the German DIN (200mm, 90 degrees)... etc

Varying the capsule spacing and mutual angle alters the stereo acceptance angle and the imaging, and altering these parameters along with the distance from the source helps to find the optimum sound stage and aural perspective. So there's nothing wrong with altering the mutual angle or capsule spacing of an array to achieve the desired sound... just don't call it a defined array when it isn't because it causes unnecessary confusion (and sometimes mirth )!

I’ve used the ORTF system and even got the measuring tape out! and it’s worked fine, but I’m wondering, how much does it have to be out so you’d hear a difference?

[Hugh Robjohns]

Not much... Try it!

[Aural Reject]

Good that it went well but what's ortf ____ at 90*?

You and me both, eh?

For the benefit of others, the French ORTF mic array has a very precise specification: 170mm capsule spacing with a 110 degree mutual angle.

If the stereo mic array doesn't have that exact spacing and that exact mutual angle, it ain't ORTF... it's simply a near-spaced stereo pair of your own design.

For reference, there are several other precisely defined near-spaced mic arrays, such as the Dutch NOS (300mm and 90 degrees), or the German DIN (200mm, 90 degrees)... etc

Varying the capsule spacing and mutual angle alters the stereo acceptance angle and the imaging, and altering these parameters along with the distance from the source helps to find the optimum sound stage and aural perspective. So there's nothing wrong with altering the mutual angle or capsule spacing of an array to achieve the desired sound... just don't call it a defined array when it isn't because it causes unnecessary confusion (and sometimes mirth )!

I’ve used the ORTF system and even got the measuring tape out! and it’s worked fine, but I’m wondering, how much does it have to be out so you’d hear a difference?

As Hugh has said, using your ears and fiddling on appropriate lab monkeys is the best way....but if you cba or don't have a home office licence....

http://www.sengpielaudio.com/HejiaE.htm

[Hugh Robjohns]

The sengpiel calculators are very good, but I mostly use the Neumann Recording Tool app on my phone to visualise mic arrays tweaks as it's interactive and quicker to use (and easier to do when on location).

As a general rule, though, reducing either the mutual angle or the capsule spacing of a stereo array reduces the source width as heard over stereo speakers (and vice versa).

This is because reducing the mutual angle (or spacing) means the interchannel level (or timing) difference for an off-axis source also reduces. Smaller level /timing differences between channels produce images closer to the centre — hence a narrower source image.

Thinking of it on a different way, if you reduce the mic angle/spacing then to appear fully in one or other speaker the off-axis source has to go even more off-axis.

As a guiding technique to optimal mic array placement, I'd suggest rigging what you think might work — ORTF or whatever — switch the monitoring to mono, and move the array nearer or further from the source until the perspective (direct/reverb balance,) is just on the dry side of ideal.

Then revert to the monitoring to stereo and notice the slightly more reverberant character. Repeat adjustments until happy with the mono/stereo compromise.

Next, consider the stereo imaging. Does the source fill the soundstage appropriately?

If it's too narrow, try increasing the mutual angle and/or capsule spacing slightly (5 degree and 1cm increments) — being mindful that more direct signal is potentially being captured off-axis and with more risk of comb-filtering, so generally with more colouration. Check mono compatibility again.

If the source is overly wide then consider reducing the mutual angle and/or capsule spacing.

Also, be aware that changes to the mutual angle and capsule spacing not only alter the stereo recording angle (and thus source image width), but also the image linearity or 'spatial distortion' — the relationship between the incidence angle in the recording environment and the perceived angle in the stereo image portrayed by the speakers.

Spatial distortion can be significant in some configurations, which is why we have different standard arrays optimised for different stereo recording angles.

For more on that, see Michael Williams' books and articles.

[Aural Reject]

The sengpiel calculators are very good, but I mostly use the Neumann Recording Tool app on my phone to visualise mic arrays tweaks as it's interactive and quicker to use (and easier to do when on location).

Totally agree…but I’m, er, cough on a coffee break in the office and had that bookmarked

[Drew Stephenson]

As a general rule, though, reducing either the mutual angle or the capsule spacing of a stereo array reduces the source width as heard over stereo speakers (and vice versa).

I do find this weirdly counter-intuitive. My primitive brain says that if you narrow the mutual angle, the actual source takes up more of the space between the on-axis lines of the microphones, therefore when you play it back it should come out as a wider sounding source. A bit like zooming in with a camera lens.
I understand why this isn't the case, from your answer below, but my brain is really not happy about it!

[Hugh Robjohns]

I do find this weirdly counter-intuitive.

You're not alone. Most students of the subject struggle too.

My primitive brain says that if you narrow the mutual angle, the actual source takes up more of the space between the on-axis lines of the microphones, therefore when you play it back it should come out as a wider sounding source.

What matters to the speaker image is the interchannel level difference. More width needs more difference... but a narrower mutual angle gives less difference — and when the angle is zero, no difference and thus dual mono.

[Wonks]

I often find things are often easier to understand if you take them to one extreme or another. in this case, assume that both mics are parallel to each other and close together. It's easy to see that the stereo image width would be small/almost non-existent and you'd get close to a mono image.

Moving the mics apart, or widening the angle between them (within reason), can then only give you a wider stereo image.

[Drew Stephenson]

(still don't like it though!)

[Arpangel]

(still don't like it though!)

I’ve tried a textbook ORTF pair on our piano and it didn’t cut it, I’m using a wide, spaced pair of cardioids about 18 inches off, and it sounds like our piano, perfect.

[tea for two]

Zoom H2n has M/S facility which I wasn't aware of until now.

Reviewed 2012
https://www.soundonsound.com/reviews/zoom-h2n
"Initially I wondered if the M/S facility would work properly or be a marketing gimmick, given that this is a budget product. I'm happy to report that I found it extremely effective. Although the parameters can be adjusted on-board, in RAW mode, I found leaving this until the mixing stage was the best option, and the M/S decoder was joyously simple to use.

The quality of the mics is about the same as those on similarly-priced products I've tested: sound quality is reasonably good, and certainly usable, but not exceptional. Importantly, handling noise is very low — all the more impressive when the entire recorder is about the price of a typical budget condenser mic!"

So now I'm interested in acquiring H2n 2nd hand so as to checking out its M/S facility.