Topic: Acoustic Recording Techniques

[al w.]

I put a couple of samples up here if anyone is interested.

https://samply.app/p/FqcMuOjiLELXeu9AHMPo

Looking forward to listening, but right now it looks like the player is set to private 

[Thelonious]

Should be fixed. My bad, technology is not a personal strength. 

[aaronji]

^ Taping might help you with that (to some extent, at least)! Lots of software, hardware, and technical stuff to ponder...

[Gutbucket]

Nice! Will give a listen later.

In general, to tame DPA 4060s when necessary try a parametric EQ centered around 12kHz with Q and cut adjusted to suit.  That tends to correct for the peak in response without throwing as much of a blanket over everything. Might need to do that in combination with a shelf filter further adjusting general treble response.  Other thing you can do is to use them without any grid installed in situations where you think they might sound bright.

I suspect you probably don't need the omnis in the mix, but you never know for sure until afterward, and having them there makes for a good for comparison.   Setup like that with the same spacing as the CMC112 pair, the 4060's look to be nicely protected, and it provides a good opportunity to compare those two pairs in isolation, as well as compare X/Y center + omnis against X/Y center + CMC122 pair.

Ordinarily if I were intending to use the omnis in the mix with the other two other pairs I'd either space the omnis considerably farther apart - say 60" or so, which helps when sources are widely distributed across the stage or moving around like the trumpet - or I'd tape them to the vertical surface just under the stage-lip facing out at the audience so that they stage lip acts as a baffle between them and the sound coming from on stage. The second option provides a greater degree of mixdown control over direct/reverberant balance and some control over the band/audience balance, due to the directional mics above the lip providing maximal pickup of the band and reduced pickup of the audience and room, while the omnis under the lip do the opposite.  Also, when used that way under the lip the omnis don't need to be spaced as far apart as they do when up on stage with the other mics.

[Thelonious]

^ Taping might help you with that (to some extent, at least)! Lots of software, hardware, and technical stuff to ponder...

One can only hope. But seriously, I'm pretty good with hardware, it's apps I tend to need to spend more time learning.

Nice! Will give a listen later.

In general, to tame DPA 4060s when necessary try a parametric EQ centered around 12kHz with Q and cut adjusted to suit.  That tends to correct for the peak in response without throwing as much of a blanket over everything. Might need to do that in combination with a shelf filter further adjusting general treble response.  Other thing you can do is to use them without any grid installed in situations where you think they might sound bright.

I suspect you probably don't need the omnis in the mix, but you never know for sure until afterward, and having them there makes for a good for comparison.   Setup like that with the same spacing as the CMC112 pair, the 4060's look to be nicely protected, and it provides a good opportunity to compare those two pairs in isolation, as well as compare X/Y center + omnis against X/Y center + CMC122 pair.

Ordinarily if I were intending to use the omnis in the mix with the other two other pairs I'd either space the omnis considerably farther apart - say 60" or so, which helps when sources are widely distributed across the stage or moving around like the trumpet - or I'd tape them to the vertical surface just under the stage-lip facing out at the audience so that they stage lip acts as a baffle between them and the sound coming from on stage. The second option provides a greater degree of mixdown control over direct/reverberant balance and some control over the band/audience balance, due to the directional mics above the lip providing maximal pickup of the band and reduced pickup of the audience and room, while the omnis under the lip do the opposite.  Also, when used that way under the lip the omnis don't need to be spaced as far apart as they do when up on stage with the other mics.

Thank you for the tips, I will investigate the EQ suggestions. It's the next frontier of learning for me.

The reason the DPAs were located directly under the other mics was protection. The pic doesn't show it but there were chairs in the audience all the way up to the right side of the mic bar and I had to move the chairs on the left side so the band could load around it. This place holds like 30 people and that was pretty much the full footprint I could afford and I still worried every time someone went on or off stage.

I did, in fact, listen to the all the pair individually and tried a mix with the DPAs and the centre pair and one with all 3. The DPAs had the horns more prominent in the mix and I may go back and re-EQ the DPA pair and try that again both alone and with the MK22s. Between the difficult treble and what I assume are phase issues, the 22/41 pair won but only after listening to each mix on multiple systems. The treble issue is funny because I didn't notice it on the bass tilted headphones I used to mix and they sound great on a bluetooth speaker but the second I played it on something with more treble it made me fill slightly ill and I abandoned that pair from the mix for now.

Thanks again to everyone for the help, I'm super happy with how this turned out in what I consider to be a somewhat challenging environment.

[Thelonious]

I recorded another acoustic show and thought I'd share samples. As this was a Jazz Vespers show (part of a service) I only set up a single set of mics, off centre, but was happy with the outcome given the circumstances. For perspective, it was about 40C (104F) here yesterday so the church had a portable AC unit and a moving fan going. I was also set up to one side to allow the reverend to come out between songs and do his thing. I'm happy with the output, however, it would have been great to get another mic up to cover the other side of the church as the band covered the width of the building.

I have a lot of respect for those who record in a church now. The acoustics are not that easy to work with given the extreme reverb if you move too far from the source.

MK22s (21cm, 110 degrees)>Nbob>riotbox>zoomF3(96/24). Dithered to 48/24 normalized, compressed two very loud trumpet peaks but otherwise no post processing.

https://samply.app/p/37iClYfaU7HrQ4fsXpA2

[al w.]

Nice work! Sounds lovely! I feel like I'm there.

[Thelonious]

Thanks Al. I am very happy with the recording given the constraints.

[Gutbucket]

Returning to this thread feeling a bit philosophical today..

I have a lot of respect for those who record in a church now. The acoustics are not that easy to work with given the extreme reverb if you move too far from the source.

Challenging yet rewarding. The combination of clean dynamic sound directly from the instruments, supported by a deep lush reverberance can be really fantastic.  Recording acoustic instruments in that kind of environment really makes the differentiation between direct and reverberant sound clearly apparent, and very much rewards thinking in terms of that relationship.  It takes both those things working well in combination for the magic to really happen.  Attention to how both are handled and combined in the resulting recording becomes especially rewarding IME.  Especially since really optimizing for both can sometimes place things at odds.  Makes for a fun challenge.

While arranging the recoding setup in such a way to achieve good clean direct sound pickup with a properly balanced representation of sources involves all the things we tend to discuss at TS such as stereo mic setup, stereo recording angle, the resulting image distribution and all that, details about how the reverberant sound is picked up and translated into the recording is right up there with it in these situations.  I think valuing and really optimizing for that part is often a bit foreign to tapers simply because so much of practical PA-reinforced concert taping rewards maximizing good quality pickup of direct sound and minimizing pickup of audience and reverberant room sound to the greatest practical extent.  "Direct good, reverberant bad".  That's a gross simplification but I think rings true.  Of course tapers recognize and value a good reverberant qualities and "know it when they hear it" in a well made recording, but its easy to not be actively concerned with optimizing the quality of reverberant pickup other than simply minimizing it.  Even when mixing in an AUD to add bit of live feel to a dry SBD recording, that AUD was usually recorded in a way that primarily optimized for direct sound pickup (even though that's what the SBD is doing a better job of providing) and is rarely optimized specifically to provide the elements the SBD feed lacks.  It pretty much contains and provides the reverberant/room/audience-reaction qualities simply because its an AUD.  I'm not casting shade on taper methods here, it works, the priorities are not misplaced - it all makes sense given the challenges of PA-reinforced concert recording.  It's just that this style of acoustic recording really rewards nuanced ways of thinking about and approaching how everything other than the direct sound emanating from the instruments and PA is handled as well.

I find recording non PA-reinforced acoustic music in these kind of situations is a good reminder of what PA-reinforcement can and can't do well.  It's amazing how far a good PA is able to push the critical radius of reverberation out into the audience.  But also amazing how much it radically changes the nature of the sound in doing so, forcing it to be less dimensional with considerably less dynamic extension, both in terms of micro and macro dynamics, even when the PA is exceptional and has been setup and tuned outstandingly well. Much of that is easily overlooked when one has never or rarely records in these environments.

In addition to making great recordings of exceptional performances, recording in these types of scenarios makes for a great opportunity to think about all this for those so inclined.  Big thanks to all the folks contributing to this thread and sharing their experiences.  Make the most of it. 

[voltronic]

I recorded another acoustic show and thought I'd share samples. As this was a Jazz Vespers show (part of a service) I only set up a single set of mics, off centre, but was happy with the outcome given the circumstances. For perspective, it was about 40C (104F) here yesterday so the church had a portable AC unit and a moving fan going. I was also set up to one side to allow the reverend to come out between songs and do his thing. I'm happy with the output, however, it would have been great to get another mic up to cover the other side of the church as the band covered the width of the building.

I have a lot of respect for those who record in a church now. The acoustics are not that easy to work with given the extreme reverb if you move too far from the source.

MK22s (21cm, 110 degrees)>Nbob>riotbox>zoomF3(96/24). Dithered to 48/24 normalized, compressed two very loud trumpet peaks but otherwise no post processing.

https://samply.app/p/37iClYfaU7HrQ4fsXpA2

I'm getting to this really late, but this sounds really nice and unhyped, just like the chill playing.

[Gutbucket]

Went back to listen to Thelonious's jazz in church recording again this morning..

I agree with Voltronic. That church reverberance is handled nicely, suiting the horns the and relatively sparce mid-tempo arrangements.  I think the reverberance helps the horns imaging far left and piano on the right to sound less disconnected and off to the edge by themselves, particularly the sax - the extra verb and its more distant sounding placement works fine. Like'n the nicely centered drums and bass.  Good translation of the acoustic bass and piano, both of which can be tricky at times in a reverberant space. I see a GK amp behind the bassist facing the recording position which makes that one easier.  Without an amp, it can be difficult to get sufficient level and detail from the bass compared to the other instruments, in which case it helps to position the rig somewhat close to the bass. Piano is the Roland through an amp I assume, which can make recording that instrument in good balance somewhat easier too.  Standard non-upright acoustic piano in an ensemble setup in this kind of reverberant space can be challenging since the soundboard is horizontal and radiates up/down.  When the mics are in about the same plane as the soundboard, almost all sound picked up from the piano will to be early reflections off the floor and lid along with reverberant sound, rather than direct sound.  Yes that's how pianos naturally sound for folks in the audience and players on the other side of the stage, but getting clear piano that's well balanced with the other instruments radiating more directly toward the recording position can sometimes be tricky.

The size of the ensemble and repertoire strikes me as an appropriate for that space. A jazz ensemble that's larger or more amplified might not work as well in that room, particularly if playing energetically.  If things were to get much louder and denser, the reverb that sounds quite nice on this recording might instead get swamped, cluttered, muddy and over dense.  That happens frequently in a local music college hall here which features acoustics appropriate for classical music when used for jazz ensemble performances.  Small "mostly acoustic" jazz ensembles work better in that rather live room than larger ensembles due to the long reverb time, especially in the bass region. At some point as the ensemble gets larger and louder, they need to shift to close mic'ing through the PA and stage monitors to better manage the acoustics.  The PA then pushes the critical radius of reverberation out much farther so it doesn't just sound like mud.

[Thelonious]

Thank you for all the feedback and suggestions to my earlier posts. They are really helpful as I try to hone my approach to non PA recording.

I recorded the Ed Lister Group again, as part of their monthly residency, at a small bar last night. The group had a local 15 year old piano player sitting in, which was a change from having a second horn as a lead instrument.

I thought I would share some samples of this one as it builds on the learning from my previous matrix attempt a few months back (last month they were shooting a promo video in the bar so I reverted to a small two channel set up for that occasion).

Building on the previous matrix approach, and some of the feedback, I created a matrix using:
1) Mk122s - 90 degrees, ~60cm
2) MK641s - XY, 90 degrees (brough in to ~+/-30% spread in post)
3) 4060s - ~60 cm, boundary mounts (Just enough to fill out the bass and bring in some room

Sources 1 and 2 are roughly equal in the mix while source 3 was brought in just enough to fill out the bass and add a bit of room to the sound. My sense is this makes the imaging slightly less precise, but also makes it more cohesive and less dry as you can hear a bit more of the audience.

I think the overall sense of space is improved (potentially with a slight reduction in image precision). I had no issues with the treble "glare", which I had in the previous recording, after a 3db roll off above 12Khz to attenuate the extra highs (I had short grids on the mics). This may be due to the use of the boundary mounts as last time I had them mounted in the mounts I use to mount to clothing which had them slightly off the floor.

I am very pleased with the recording overall, and really appreciate the extra "depth" the introduction of the omnis brings.

Samples can be found here. https://samply.app/p/YQ8XlA8i0zl92iZtequX

[Gutbucket]

I look forward to listening. Without doing so yet I'll say that what you describe very much reflects my thoughts when balancing a mix made with that kind of 3-part microphone arrangement - so advantageous to gain the luxury of being able to balance "sharp dry imaging precision" against "open atmosphere audience and room" afterward without having to compromise more fundamental balance concerns. And just feels really good to work the balance and blend of a lush engaging sense of "being there" against clear placement details. 

An advanced mixing technique which is probably not required for this recording yet is an cool option to be aware of, is to vary the level of the "roomier" microphone channels along with the loudness of the program.  Between pieces and during quieter portions and slower movements, more level from those channels opens things up nicely.  Conversely, when things get loud and energetic, reducing the contribution from those channels can help clean things up.  That's is a general trend that applies broadly to pretty much all recording situations, but tends to apply especially strongly to performances in smaller reverberant spaces where the acoustics can get over dense and somewhat muddled when it really gets going.

In this case that would entail varying the level of the omni pair somewhat, you just can't vary it so much that it becomes obvious.  I most often do that on a rear-facing mic pair that's primarily dedicated to the pick of room and audience, or sometimes just the Side channel of the center Mid/Side pair that I use instead of an X/Y pair, sometimes both, since I can vary those more than the omni pair without upsetting the overall mix.  Can do that manually using level automation points in the DAW which provides the most control although it takes more time to do, or can arrange carefully tuned compression on those channels so that their relative contribution compared to the "sharp dry imaging precision" channels decreases as the overall SPL increases.  A cool tweak that helps when doing it with compression is to trigger the compression on those channels via a sidechain input from the "sharp dry imaging precision" channels.  That way the contribution from the more ambient channels is being managed by what is happening in the dry direct sound channels, and the audience reaction between songs isn't squashed just because it happens gets energetic, but only gets reduced when the level in the"clear and direct" channels becomes more energetic.  It serves as sort of a form of more advanced parallel compression - the direct sound portion retains its full dynamics and automatically effects how much compression is applied to the reverberant/ambient/audience contribution to the mix. 

It's a subtle but rather cool option that further leverages the use of more than one pair arranged so as to work together, with each contributing something slightly different to the mix.

[Gutbucket]

Thinking a bit more on this, if you ever find yourself so inclined you might also try what I'm sometimes doing with my M/S center-pair Side channel, but with your X/Y center-pair instead.  Would be as an alternate to, or in addition to varying the level of the omni pair described above.  Several ways to do it, each effectively altering the  "~+/-30% spread" of the X/Y pair as the overall SPL level rises and falls.

One is to convert the X/Y pair to Mid/Side and vary the level of Side. Another would be to automate the panning-width of the X/Y pair directly.  Because of the presence of the other stereo pairs also contributing to the mix, doing this tends to be less about simply altering stereo image width than it would be if listening to that pair in isolation and instead somewhat more about altering the sense of "openness and fore-aft depth" - at least to my ear and for lack of a better description.  Just seeks to improve things by allowing the depth and ambience to open up a bit more when the recording is able to support it, while tightening and flattening up more when things get loud, dense and more messy.


Applied generally, any straight 2-channel stereo recording or already mixed multi-pair recording which suffers from becoming overly dense and reverberant during the louder sections might also by be improved by converting to Mid/Side and subtlety tying M/S ratio to SPL in this way.  However, since it will then be affecting the width of the entire stereo recording, its likely less will be able to be applied without the altercation of stereo width becoming apparent.  As SPL increases the sense of stereo width will be slowly reduced in addition to the recording sounding less reverberant and dimensionally deep.

[Thelonious]

Thank you very much for the suggestion on compression of the omni channels linked to SPL of the "main channels". In this case the recording is quite sparse, and the mics are quite close to all of the sources so I don't find it gets too congested, but I can imagine many cases where this will be useful so I will investigate the use of automation in my DAW, which I have never explored to date. I can certainly picture that in one of Ed's other bands where there can be up to 12 musicians, including 8 horns, in a relatively small space.

I kept the omnis at ~2' spacing due to the proximity to the audience (there is a couch stage left and the chatter is already more audible than I would prefer during the quiet sections). A +/- 3' spacing would put my left mic almost at their table.

I should have mounted below the stage lip, however, leaving the omnis up top at this distance provides a safety net in case I screw something up with one of the other channels. That would leave me with a pair of ~2' spaced omni pair that is very listenable in the worst case scenario.

Thanks again for the tips. If you had told me 13 months ago, when I first asked on TS about suggestions for setting up my pair of CM4s to record Julian Lage in what was my first time open recording a live concert that I would make a recording like this and be looking at further optimizing it, I wouldn't have believed it. A very fun journey and I am super grateful for all of the learnings from the group.