Advice on low maintenance recording of large orchestra, please.

[timcmit]

I have just started recording classical music. To date I have been recording the orchestra I play in, and so I have certain requirements:
1. Solution must look after itself – user interaction mostly limited to start and stop
2. It needs set up from scratch on each occasion (easily transported to and from location with minimal setup time) and the members of the orchestra will not appreciate too much disruption (not lots of stands and trailing cables)
3. Many players will not appreciated spot miking
4. Good quality, but nothing too expensive

I have tried a Decca Tree using 3 Audio Design OM1 mics (no outriggers as didn’t seem to add enough to justify more stands) – good and clear coverage of orchestra, both string and wind and using a single stand. I want a bit more focus and so am going to mount two Audio Design CM4 on the tree, in a Gerzon array to achieve sufficient SRA, and using stereo shuffling in post to improve stereo separation. The stand will be placed just behind the conductor at 3.1m height. I intend to back one or other of the mic configurations by 6dB.

I have also tried Rode M5s in ORTF, with omni AB spaced at 50 degrees (in Faulkner array (?) style). Placed 5m behind the conductor at 2.5m height, this seems to have good coverage, but loses the woodwind a little and captures more of the hall acoustic, which is OK, but not the best. Again, likely to lead with the ORTF and reduce the Omnis by 6dB.

Using 2 Zoom F6 units, with 32bit float, there is little need for interaction, and mounting them on the stands means a modular solution with no trailing wires that can be largely prefabricated.

All post processing will be using Cohler Classical on Reaper, probably with a bit of large hall reverb, a bit of compression to reduce the dynamic range due to very loud passages and boost of frequencies above 5KHz (+6dB, to improve clarity of sound in general).

I wonder if anyone has advice like “I wouldn’t do that” or “ could be improved by...”?

[RichardT]

Hi Tim, welcome.

I can’t comment on the recording side as I know nothing about that, but I would say: don’t go into the post-recording phase with any pre-conceptions about what processing you need to do to the raw recordings. A 6dB boost above 5kHz is absolutely massive, for example.

It’s best to listen to what you have, identify if there are any issues, and fix them. Don’t apply any processing unless you need to.

It can be helpful to compare your own recordings to your favourite recordings by other people, or recordings of the same repertoire by other people. This will give you a good idea of what might need fixing.

[Hugh Robjohns]

The more mics you rig, the more phase issues you'll have, and shuffling inherently adds more phase complications... so, personally, I wouldn't combine multiple arrays in the way you describe if it could be avoided.

Much depends on what precisely you mean by 'focus' but if you're happy with the basic Decca Tree balance, and you've experimented with dimensions and mic angles, I'd try adding Acoustic Pressure Equalisers (spheres) to your omni mics (as per the M50s mostly used in pro trees). APEs might well do exactly what you need (and boost the HF in a natural way too).

I don't know what's available specifically for your OM1s, but there are 3D printed spheres available online for most SDC omni mics. 40mm is the standard sphere diameter.

So I'd try that first best option.

After that, I'd probably try an OCCO array instead of the Tree, possibly with hypercardioids instead of cardioids if you need more reach.

[Bob Bickerton]

I’d support the concept of trying to make this work with one array.

I tend to use either an OCCO array (which appears to be a new name for what was previously known as a Faulkner array) consisting of an ORTF pair with Omni outriggers spaced at around 600mm or a Gerzon array, which gives not dissimilar options for manipulation in post using shuffling. Working with one array affords focus on mic positioning.

I’d also question working to a set formula in post as any processing will surely depend on mic positioning and room acoustics.

Finally, I tend not to spot mic unless necessary, but find that it can be most useful for concerto soloists to gives options for balance and definition.

Bob

[MarkOne]

I don't know what's available specifically for your OM1s, but there are 3D printed spheres available online for most SDC omni mics. 40mm is the standard sphere diameter.

https://www.shapeways.com/product/5KY2M ... arketplace

Every day is a school day.... I'd never heard of this before. What's the science Hugh?

[timcmit]

The more mics you rig, the more phase issues you'll have, and shuffling inherently adds more phase complications... so, personally, I wouldn't combine multiple arrays in the way you describe if it could be avoided.

Much depends on what precisely you mean by 'focus' but if you're happy with the basic Decca Tree balance, and you've experimented with dimensions and mic angles, I'd try adding Acoustic Pressure Equalisers (spheres) to your omni mics (as per the M50s mostly used in pro trees). APEs might well do exactly what you need (and boost the HF in a natural way too).

I don't know what's available specifically for your OM1s, but there are 3D printed spheres available online for most SDC omni mics. 40mm is the standard sphere diameter.

So I'd try that first best option.

After that, I'd probably try an OCCO array instead of the Tree, possibly with hypercardioids instead of cardioids if you need more reach.

Thank you - that has been very helpful.

I was really taken (and quite amazed) with the quality of the Decca Tree (2m between left and right, centre 0.5m forward). The sounds of the instruments were vivid: the oboe every bit as clear as the strings. However, it didn't sound quite like I expect an orchestra to sound like; maybe with the position of the mics, this is what the conductor hears?

The Faulkner Array was more as I would expect an orchestra to sound, but I lost some of the vividness and clarity of the instruments (the oboe now sounded distant) - perhaps as it needed to be mounted further away. I was hoping to get the best bits of both, but realise this is simplistic and, even in my limited experience I have experienced problems combining different arrays. You suggest an OCCO with hypercardiods - do you think this would be more effective if it was closer to the orchestra (perhaps compromising the SRA of the cardiod pair, falling back on the SRA of spaced omni pair?).

I will try to get hold of the APEs as you suggest, as I feel the Decca tree really has a lot to offer.

I also fancy exploring the Gerzon array, not now to combine with the Decca tree, but hope that with the wider SRA and being able to mount it nearer to the orchestra, that it will provide better balance of the woodwind versus the strings. If I use the Gerzon array on its own, is stereo shuffling likely to give acceptable results?

Thanks again, I think you saved me a lot of experimentation!

[timcmit]

Hi Tim, welcome.

I can’t comment on the recording side as I know nothing about that, but I would say: don’t go into the post-recording phase with any pre-conceptions about what processing you need to do to the raw recordings. A 6dB boost above 5kHz is absolutely massive, for example.

It’s best to listen to what you have, identify if there are any issues, and fix them. Don’t apply any processing unless you need to.

It can be helpful to compare your own recordings to your favourite recordings by other people, or recordings of the same repertoire by other people. This will give you a good idea of what might need fixing.

Thank you. I take your point on the treble boost, but mine was a frustration with my ORTF recording which lacked the vividness of the Decca Tree. The high frequencies seemed to put some of it back, but realise it would be better to capture it in the first place.

I have listened to others - a bit disheartening! I get the feeling that many more modern recording employ spot miking to achieve better clarity.

I'll be careful in post processing!

[timcmit]

I’d support the concept of trying to make this work with one array.

I tend to use either an OCCO array (which appears to be a new name for what was previously known as a Faulkner array) consisting of an ORTF pair with Omni outriggers spaced at around 600mm or a Gerzon array, which gives not dissimilar options for manipulation in post using shuffling. Working with one array affords focus on mic positioning.

I’d also question working to a set formula in post as any processing will surely depend on mic positioning and room acoustics.

Finally, I tend not to spot mic unless necessary, but find that it can be most useful for concerto soloists to gives options for balance and definition.

Bob

Thank you Bob. We're you suggesting that the Gerzon, unlike the ORTF, wouldn't need flanking with Omnis (perhaps the Stereo shuffling provides low frequency boost?)

[Bob Bickerton]

I’d support the concept of trying to make this work with one array.

I tend to use either an OCCO array (which appears to be a new name for what was previously known as a Faulkner array) consisting of an ORTF pair with Omni outriggers spaced at around 600mm or a Gerzon array, which gives not dissimilar options for manipulation in post using shuffling. Working with one array affords focus on mic positioning.

I’d also question working to a set formula in post as any processing will surely depend on mic positioning and room acoustics.

Finally, I tend not to spot mic unless necessary, but find that it can be most useful for concerto soloists to gives options for balance and definition.

Bob

Thank you Bob. We're you suggesting that the Gerzon, unlike the ORTF, wouldn't need flanking with Omnis (perhaps the Stereo shuffling provides low frequency boost?)

The shuffling has the effect of ‘opening up’ the sound, which to my ears sounds like more generous stereo imaging and lower frequencies, but you really need to try it to see if it suits you. Using high-quality cardioid mics with a good off-axis response is critical. I use MKH 8040s.

Re-reading your first post I see you used Rode M5s for the ORTF configuration, not the best mics in the world for this application. My approach with the Faulkner array is to balance in post between ORTF and OMNIS - it’s not a matter of pre-calculating the outcome, more a question of using ears! You’ll make different decisions based on programme, venue acoustics, etc.

Bob

[resistorman]

However, it didn't sound quite like I expect an orchestra to sound like; maybe with the position of the mics, this is what the conductor hears?

Exactly. If the hall is great sounding you might want to capture its sound either live or via convolution. It is not a sin to add artificial reverb either

[jimjazzdad]

Here's a comment from the peanut gallery. ORTF will be more "low maintenance" and will produce consistent results in a variety of sessions in different spaces. Decca trees and omni mics have a lovely sound in a good space but will always tell the naked truth in a less than ideal acoustic. A really good ORTF - Schoeps MSTC ideally - on a light-weight stand is easy to move around to find the sweet spot, unlike a Decca tree. The directional nature will filter out audience noise and the less than perfect diffuse sound from a less than perfect hall. If you wish, couple the ORTF with one of those whiz 32 bit multi converter field recorders and forget about setting levels. That's my advice for a low maintenance recording rig and its worth every penny (pence) you paid for it.

[Hugh Robjohns]

Every day is a school day.... I'd never heard of this before. What's the science Hugh?

Well found. Shapeway aren't cheap, but they have an amazing range to pick from (some designs are much better than others). I've used them for a few things myself.

The APE sphere does what it says — alters the acoustic pressure on the diaphragm. The net result is an HF boost and a narrowing of the polar pattern at HF. Its the former that brings clarity and 'focus' in a Decca Tree, while the latter ntroduces bigger inter-channel level differences at high frequencies to enhance the stereo effect.

APEs vary in size with 30mm diameter being the smallest and 50mm the biggest. Neumann's classic M50 mic used a 40mm sphere and the majority are of that size.

A 30mm APE introduces a roughly 2dB peak two octaves wide centred at 5kHz, and creates a cardioid-like response above about 10kHz.

A 40mm sphere does a little bit more — slightly stronger and wider peak, and cardioid-like from a slightly lower frequency. And the 50mm does an even more heavy-handed version...

[Hugh Robjohns]

I was really taken (and quite amazed) with the quality of the Decca Tree (2m between left and right, centre 0.5m forward).

That's quite a stunted tree!

If the side mics are 2m apart (which is about the maximum width for a Tree), I'd expect the centre mic to be more like 1.5 metres forward of them.

That front distance is important in defining the stereo image and minimising LF correlation.

However, it didn't sound quite like I expect an orchestra to sound like...

Most modern orchestral recordings make use of a lot of accent mics and have an exaggerated clarity that you will never get with a Tree on its own.

The Faulkner Array was more as I would expect an orchestra to sound, but I lost some of the vividness and clarity of the instruments (the oboe now sounded distant)...

I'm not sure what to make of that... how do you expect an orchestra to sound?

However, placing an array right behind the conductor will inherently exaggerate the depth of the orchestra, and so winds and percussion will naturally sound more distant than the strings. Increasing the array height can help to even things out, ad can moving the array further back, but both will also increase the ambience/reverb.

You suggest an OCCO with hypercardiods - do you think this would be more effective if it was closer to the orchestra...

It might be... If you've experimented with different cardioid spacings/angles and you're not getting what you need, then it's an alternative option that might help. Try it and see...

I will try to get hold of the APEs as you suggest, as I feel the Decca tree really has a lot to offer.

I also fancy exploring the Gerzon array, not now to combine with the Decca tree, but hope that with the wider SRA and being able to mount it nearer to the orchestra, that it will provide better balance of the woodwind versus the strings.

As I said above, getting closer to the strings is unlikely to improve the balance of distant woodwinds. Think about their relative distances and the inverse square law of sound propagation.

Thanks again, I think you saved me a lot of experimentation!

Oh dear... that's the last thing I wanted to do! Recording is all about experimentation and critical listening, because every situation is different.

[timcmit]

Here's a comment from the peanut gallery. ORTF will be more "low maintenance" and will produce consistent results in a variety of sessions in different spaces. Decca trees and omni mics have a lovely sound in a good space but will always tell the naked truth in a less than ideal acoustic. A really good ORTF - Schoeps MSTC ideally - on a light-weight stand is easy to move around to find the sweet spot, unlike a Decca tree. The directional nature will filter out audience noise and the less than perfect diffuse sound from a less than perfect hall. If you wish, couple the ORTF with one of those whiz 32 bit multi converter field recorders and forget about setting levels. That's my advice for a low maintenance recording rig and its worth every penny (pence) you paid for it.

Thanks. I have a 32bit float recorder and it's a boon for me as I should be playing rather than setting levels! I think this sounds good advice and I will be having ORTF as a backup in any event.

I'm interested in the mic choice. When I thought many mics were necessary, I purchased a number of cheaper ones - I now realise that I dont need many and could have focused on just a few... I rely on Line Audio OM1s and CM4s for my 'better quality'. How much difference would better mics make given the volume of the orchestra and the acoustic of any hall?

[jimjazzdad]

Thanks. I have a 32bit float recorder and it's a boon for me as I should be playing rather than setting levels! I think this sounds good advice and I will be having ORTF as a backup in any event.

I'm interested in the mic choice. When I thought many mics were necessary, I purchased a number of cheaper ones - I now realise that I dont need many and could have focused on just a few... I rely on Line Audio OM1s and CM4s for my 'better quality'. How much difference would better mics make given the volume of the orchestra and the acoustic of any hall?

I don't own any Line Audio mics (I really should get some) but I have listened to them critically and they are excellent; you would have to spend a lot more - e.g. Schoeps - to see any worthwhile increase in quality. The self noise is a little higher, but you may not notice any difference as the mic's noise is likely to be below the noise floor in almost any concert hall, especially if there is an audience present. The OM1 (Omni1) mics are not really tailored for use at a distance. Their response is ruler-flat, so they may need a little HF EQ if used as room mics (or the APE spheres Hugh mentioned). The CM3/4 are slightly wide-cardioid but should work very well in ORTF or in a slightly tighter NOS pattern. But, as always, experiment and let your ears guide you. I recommended ORTF to you because you said in the OP that you are a musician and are looking for a "low maintenance" rig. I too am still a 'schoolboy', having been recording orchestras and ensembles for about a dozen years since I retired from my other working life...I'm still learning.

[timcmit]

That's quite a stunted tree!

Sorry, my centre mike is 1m forward (not 0.5m as I said), but I will extend this for the benefits you mention.

I'm not sure what to make of that... how do you expect an orchestra to sound?

I meant the sound you might expect sitting in an audience; more in front of, rather than around the listener that the Decca Tree seems to provide. The ORTF array seems to give more of this - I like the best bits of both methods!

As I said above, getting closer to the strings is unlikely to improve the balance of distant woodwinds. Think about their relative distances and the inverse square law of sound propagation.

Of course. At least the woodwind are raised higher than the strings, and with the 3m Tree height, they are perhaps a similar distance to the rear desks of all of the string sections. I worry that as you get further from the orchestra, the geometry suggests that all of the desks of the 1st Violins and Cellos in particular, start to become relatively closer to the array...time for some experimentation!

When I referred to less experimentation; I will always experiment, but at least now I can avoid doing so in areas that are destined to be fruitless! Thank you.

[timcmit]

OM1 (Omni1) mics are not really tailored for use at a distance. Their response is ruler-flat, so they may need a little HF EQ if used as room mics (or the APE spheres Hugh mentioned).

They are on order!

I recommended ORTF to you because you said in the OP that you are a musician and are looking for a "low maintenance" rig. I too am still a 'schoolboy', having been recording orchestras and ensembles for about a dozen years since I retired from my other working life...I'm still learning.

As a recently retired Scientist/Engineer I am looking forward to the challenge! Thanks for your advice.

[tea for two]

the sound you might expect sitting in an audience; more in front of, rather than around the listener that the Decca Tree seems to provide. The ORTF array seems to give more of this - I like the best bits of both methods!

I will always experiment, but at least now I can avoid doing so in areas that are destined to be fruitless! Thank you.

As a recently retired Scientist/Engineer I am looking forward to the challenge! Thanks for your advice.

Weeeell just to scratch that itch : sticking a mic in-between an ORTF : aboot 1m-1.5m infront.
You never know.
I'm sure others have tried this.
I haven't lol. Butt I would.

[Hugh Robjohns]

Weeeell just to scratch that itch : sticking a mic in-between an ORTF : aboot 1m-1.5m infront.
You never know.

Not if you don't understand the principles, no....

I'm sure others have tried this.

I doubt it. It would be a pointless and destructive addition to an ortf array....

[jimjazzdad]

While it is generally accepted that you don't position omni and cardioid mics at the same distance or together in an array, there have been a few engineers who hace had success with this sort of experiment - e.g. Tony Faulkner or Volker Straus. I have always meant to try a single omni with an ORTF pair, low pass filtered to just add 'bass beef' to the cardioids...the trick would be to just keep anything below say, 125 Hz, since most of the directional cues are much higher. But I guess if it really worked, all the hot classical engineers would be doing it.

[Drew Stephenson]

I guess it's easier just to use a little eq lift and avoid any potential phase issues.

[jimjazzdad]

I guess it's easier just to use a little eq lift and avoid any potential phase issues.

I can't imagine that there would be any phase issues with an omni capturing just the low frequencies at 8.5 cm from a cardioid - its decimal points of a wavelength and probably nanoseconds of time shift (I'm not great at the math but I have general sense of it). I guess I'll just have to try it next time I deploy an ORTF set up. But yes, EQ is easy.

[tea for two]

I'm sure others have tried this.

I doubt it. It would be a pointless and destructive addition to an ortf array....

Of course they would have to narrow down what worked what didn't.
Blumlein Faulkner Gerzon et al didn't just magically ooot the ether pull out their arrays, they would have experimented to reject various configurations.

Conventional wisdom exists for a reason : mic arrays being one such.
Still feel free to do our own experiments : we are not slaves to others concepts. No one's going to die doing such experiments.

Just seen Jim's post "had success with this sort of experiment - e.g. Tony Faulkner or Volker Straus."

::

While it is generally accepted that you don't position omni and cardioid mics at the same distance or together in an array, there have been a few engineers who hace had success with this sort of experiment - e.g. Tony Faulkner or Volker Straus.
I have always meant to try a single omni with an ORTF pair, low pass filtered to just add 'bass beef' to the cardioids...the trick would be to just keep anything below say, 125 Hz, since most of the directional cues are much higher. But I guess if it really worked, all the hot classical engineers would be doing it.

We just don't get the luxury to experiment. So we stick to what we know works.
Many of the hot classical engineers get to often work in fab rooms halls studios so the known arrays works best.

I'm just putting it out there as an experiment if we ever get the luxury to experiment.

[Bob Bickerton]

Conventional wisdom exists for a reason : mic arrays being one such.
Still feel free to do our own experiments : we are not slaves to others concepts. No one's going to die doing such experiments.

A few thoughts:

Those people who experimented did so with experience and in-depth knowledge of the principals involved

I'm a slave to things that work

You could die if the massive experimental array fell over on top of you

Bob

[Hugh Robjohns]

I have always meant to try a single omni with an ORTF pair, low pass filtered to just add 'bass beef' to the cardioids...the trick would be to just keep anything below say, 125 Hz, since most of the directional cues are much higher. But I guess if it really worked, all the hot classical engineers would be doing it

It would add bass beef.... but you'd also introduce some LF phasiness because of the LF phase shift inherent in the cardioids and potentially also in the low-pass filtering.

And you'd have mono bass with the reduced sense of spaciousness that entails.

One of the reasons classical engineers like spaced omni techniques is because of the desirable sense of spaciousness that comes from stereo spaced mics where the LF phase differences translate into LF directionality.