OneMic recordings, hear what the band heard in the studio!

[ChrisG]

I found the vocal reverb detracted from the one-mic listening experience. If nothing else, IMO it should have been a subtle addition.

 

[mjgraves]

One mic, or coincident techniques, can be magical. Recall the Cowboy Junkies "Trinity Sessions" which was recorded in a church in downtown Toronto. The band was arranged around a single Calrec Soundfield microphone.

 

[goat76]

I found the vocal reverb detracted from the one-mic listening experience. If nothing else, IMO it should have been a subtle addition.

Is it a particular one-mic recording you are referring to?

I’m sure the amount of reverb Cuniberti applies to the vocals will differ from recording to recording, and some of them will likely not have any reverb at all depending on the style of the songs.

 

[ChrisG]

Ah yes, sorry.
2nd video in the first post. The rest of the instruments are in a natural acoustic space from the instruments+room capture that you get with these mic techniques. The voice has a load of extra reverb pasted on top, putting it in a totally different "space" to the rest.


I like the approach of minimalist micing*, and typically have a similar approach with my own work (when possible/appropriate/etc), and I found the vocal reverb on that video quite jarring. I can imagine there might have been someone else in the studio, asking for more more more FX to get a "modern" sound, but IMO it didn't fit that production.


* For example, a couple of live shows I've done recently involved a singer with a guitar, and another with a bazouki. I used a single Austrian Audio OC818 on a stand at around chest height, played with the polar pattern to minimise feedback from the PA, and added a touch of reverb. NB - since the mic was capturing voice + instrument, any effects applied would apply to both sources equally, so the impression is of a matching acoustic space.

 

[KMSA]

The microphone the Sound Liaison guys use is a 3 capsule mic, which I guess explains why those One Mic recordings are so superior sounding compared to the rest. These does live up to their claim of sound made visual.

 

[ChrisG]

2x fig-8 plus 1x omni does open up lots of opportunities. They could be summed to XY in any polar pattern, as well as M/S stuff (NB - M/S and XY end up being equivalent, but switching to M/S processing can be useful on occasion).
FWIW, though, 3x capsules doesn't necessarily mean a recording will be "better", it just keeps some options open.

An approach with similar results would be a Schoeps M/S setup, with a selection of different capsules for the M mic.

Chris

 

[KMSA]

the latest 2 one mic recordings from Sound Liaison are completely different types of music. Cloud Song is at times very loud almost rock inspired jazz , while Down To The Downtown is intimate Americana songs extremely beautiful played by this guitar and bass duo. The problem I have with these albums is that all my other albums sounds artificial after hearing these.

 

[KMSA]

This is going to be an interesting comparison!

A One Mic Recording. This recording is part of an experiment demonstrating the sonic difference between a One Mic recording, done with the Josephson c700s microphone, and a more traditional multi-mic recording. The sound in this version is coming from the C700s microphone in front of the band. The other microphones are not being heard in this video. For more info on our one mic recordings visit: soundliaison.com/one-mic-recordings Carmen Gomes - vocal Folker Tettero - guitars Peter Bjørnild - double-bass Bert Kamsteeg - drums, Audio recording by Frans de Rond for Sound Liaison. Video recording by Wolfgang Maiwald

 

[KMSA]

2x fig-8 plus 1x omni does open up lots of opportunities. They could be summed to XY in any polar pattern, as well as M/S stuff (NB - M/S and XY end up being equivalent, but switching to M/S processing can be useful on occasion).
FWIW, though, 3x capsules doesn't necessarily mean a recording will be "better", it just keeps some options open.

An approach with similar results would be a Schoeps M/S setup, with a selection of different capsules for the M mic.

Chris

Thankyou for this very interesting info. Can you recommend recordings done in this way?

 

[ChrisG]

The mic layout I described is the Josephson C700S, used by OneMic. https://www.josephson.com/pdf/srs7.pdf

I don't think they disclose how they've processed the mic outputs.


Whichever way it's processed/mixed, though, it's still a coincident mic technique. ie, the playback-position of a source is entirely due to level differences from L to R.
Other mic techniques space the microphones by some amount, so that there are both level- and time-based differences between the L/R channels on playback.

Near-coincident techniques space the mics by up to around 30cm. Wider than that, and I'd say you're in "spaced mics" territory.


For my most recent recording of a choir (a good example of a spatially-distributed source), I decided to record a selection of mic techniques simultaneously so I could pick and choose later on. From centre to outside, I had:
- 1x Austrian Audio OC818 + Schoeps MK8 for a mid-side capture. I set the OC818 for a subcardioid pickup pattern.
- 2x Schoeps MK4 in ORTF
- 2x OC818 spaced at around 55cm. I took separate feeds from the front and rear capsules (back-to-back cardioids) so I could play with the polar pattern later on.


Some tasting notes:
- The M/S capture sounded fine, if a little 1-dimensional. The flexibility of M/S processing is always a bonus, though: I always find it cool to bring up the fader of the Side mic and hear the audio transform from mono to stereo.
- The ORTF sounded great, but perhaps a little dry. Moving the mic stand further back into the room would have helped, I suspect, but it was a live recording with an audience. Keeping the stand out of the way was also a priority.
- The spaced mics presented a rather diffuse stereo image, which tended to gather at the speakers instead of filling the space between them. There was some centre-image, but it was weak. Changing the polar pattern altered the amount of ambient/reverb capture, but the stereo effect stayed similar.
- Taking the spaced mics and adding the Mid of the M/S back in allowed a stronger centre image. Result: similar to spaced, except the sound has a third, central point to gather at.

My mix ended up being the ORTF pair, plus a little bit of the rear-facing capsules of the spaced pair. Adding those allowed me to add a bit more of the room reverb as it was captured, in a diffuse non-obvious kind of way. I think the results sounded rather good, and have sent the files off to the choirmaster for evaluation.


If I was to repeat the experiment, I would add my last pair of OC818s in a narrower spaced position, and take both outputs of the Mid mic. That's another five channels of recording, though.

Finally, I have wondered if it'd be worth using the spaced/rear feeds and routing those to surround channels. Might be fun, but multi-channel audio has become very complex. A whole different discussion...

Chris

 

[goat76]

@ChrisG, It's interesting to read about your recording and the different stereo microphone techniques you used. I understand if you can't share your recordings, but it would have been cool to hear them all separately and play with them in a DAW.

For a choir recording as you made, I wouldn't expect a Mid/Side recording to be the best-suited recording technique as such microphone configuration probably works best when there is a solo vocalist or a solo instrument "smacked in the middle" with a higher importance than the rest of the singers/musicians.

 

[Blumlein 88]

Something I've tried which works well is an ORTF and the spaced rears. Only the spaced rears are cardioid pointed up towards the ceiling. It's something from one multi channel technique, but it works nicely to mix into a stereo front pair for ambience with slightly reduced pickup of audience noises.

 

[Buster]

the latest 2 one mic recordings from Sound Liaison are completely different types of music. Cloud Song is at times very loud almost rock inspired jazz , while Down To The Downtown is intimate Americana songs extremely beautiful played by this guitar and bass duo. The problem I have with these albums is that all my other albums sounds artificial after hearing these.

Wow! The Down To The Down Town album is simply the best recording of an acoustic guitar in existence.

 

[KMSA]

2x fig-8 plus 1x omni does open up lots of opportunities. They could be summed to XY in any polar pattern, as well as M/S stuff (NB - M/S and XY end up being equivalent, but switching to M/S processing can be useful on occasion).
FWIW, though, 3x capsules doesn't necessarily mean a recording will be "better", it just keeps some options open.

An approach with similar results would be a Schoeps M/S setup, with a selection of different capsules for the M mic.

Chris

Interesting. I have some reading to do understanding this stuff, but anyhow these various One Mic recordings are a revelation.

 

[ChrisG]

Something I've tried which works well is an ORTF and the spaced rears. Only the spaced rears are cardioid pointed up towards the ceiling. It's something from one multi channel technique, but it works nicely to mix into a stereo front pair for ambience with slightly reduced pickup of audience noises.

This is approximately what I ended up with, except the rears were aimed approximately at the top of the back wall, rather than straight up.

Interesting. I have some reading to do understanding this stuff, but anyhow these various One Mic recordings are a revelation.

The OneMic recordings are pretty good, and I admire the work that's required to pull it off. The usual caveats about coincident mic techniques apply, though:

When we, as humans, hear something off to one side, there's both a level difference between our ears, and a time difference.
Coincident mic techniques of any kind will only ever produce level differences between the speakers that are playing the signal back.

ie, if something is picked up to the left of a coincident mic setup, upon playback ,the left and right speaker will play that signal at the same time, but the left speaker will play it louder.

Spaced mic techniques (60cm-spaced omnidirectional mics, for example, when used in a classical setting) will typically have lots of time difference between L/R signals, but not much level difference.
Upon playback, then, each speaker will play approximately the same level, but one of them will play later than the other, giving our hearing system timing cues about the sound's original location.

Following this, we can see why various near-coincident (ie, smallish spacing, typically 30cm or less) techniques were derived. ORTF, NOS, etc etc. Those use directional mics aimed in different directions to provide some level difference from left-to-right, but the spacing introduces some timing differences as well.

The placement and choice of mic array is a frequent discussion in recording circles.


Chris

 

[mjgraves]

When we, as humans, hear something off to one side, there's both a level difference between our ears, and a time difference.
Coincident mic techniques of any kind will only ever produce level differences between the speakers that are playing the signal back.

That's not strictly true. Ambisonic mics capture the entirety of the soundfield. Ambisonic processing can deliver the acoustic perspective of anywhere within the soundfield, including both level and timing cues.

That might including rendering to binaural format for headphones. Or rendering for full ambisonic playback array, with timing and level optimal for each speaker position.

To do anything less would result in an incoherent mess.

 

[Blumlein 88]

That's not strictly true. Ambisonic mics capture the entirety of the soundfield. Ambisonic processing can deliver the acoustic perspective of anywhere within the soundfield, including both level and timing cues.

That might including rendering to binaural format for headphones. Or rendering for full ambisonic playback array, with timing and level optimal for each speaker position.

To do anything less would result in an incoherent mess.

How will a Soundfield microphone deliver timing differences? Most are in a tetrahedron meant to function like a coincident mic even though they are very slightly not coincident.

 

[mjgraves]

I understand the principle, but I'm not sure I can convey it.

With first order ambisonics the signals from the capsules (A-format) are processed via a matrix into X, Y, Z & W signals (B-format). The phase relationships are precisely maintained. From these signals an ambisonic rendering engine can derive a playback signal for any arbitrary point in space. Timing differences between the capsules form the basis for that transformation.

An ambisonic rendering into binaural delivers two distict signals. Both level and timing are handled. Ambi Explorer is a free Ambisonic playback app for mobile devices. It allows you to experiment with B-format files set, playing back into binaural. You can adjust the head position and hear the resulting change in the presentation.

 

[Blumlein 88]

I understand the principle, but I'm not sure I can convey it.

With first order ambisonics the signals from the capsules (A-format) are processed via a matrix into X, Y, Z & W signals (B-format). The phase relationships are precisely maintained. From these signals an ambisonic rendering engine can derive a playback signal for any arbitrary point in space. Timing differences between the capsules form the basis for that transformation.

An ambisonic rendering into binaural delivers two distict signals. Both level and timing are handled. Ambi Explorer is a free Ambisonic playback app for mobile devices. It allows you to experiment with B-format files set, playing back into binaural. You can adjust the head position and hear the resulting change in the presentation.

Don't think so. What the full B format is equivalent to is three coincident figure 8s and a coincident omni. They can give the equivalent result in any direction or pattern, but not in a different location.

Don't have access to the full paper, but here is an abstract where they investigate using a pair of soundfield microphones to overcome that limitation.

https://www.semanticscholar.org/paper/Spaced-A-B-placements-of-higher-order-Ambisonics-Grond-Kelly/71f43f005a8a1b9de45163c0d686beb5e4902b8e

Here is the patent from Gerzon and Craven.

US4042779A - Coincident microphone simulation covering three dimensional space and yielding various directional outputs - Google Patents

A microphone assembly for simulating a plurality of coincident microphones has at least four microphone units mutually disposed at the integration points of an integration rule for the surface of a sphere. In one embodiment, four such microphone units are disposed on respective faces of a...

patents.google.com

Final sentence in the claim ends with:

........the same frequency response to sounds from all directions up to a limiting frequency and are substantially identical to the outputs obtained from a plurality of coincident microphones.

 

[mjgraves]

Don't think so. What the full B format is equivalent to is three coincident figure 8s and a coincident omni. They can give the equivalent result in any direction or pattern, but not in a different location.

Don't have access to the full paper, but here is an abstract where they investigate using a pair of soundfield microphones to overcome that limitation.

https://www.semanticscholar.org/paper/Spaced-A-B-placements-of-higher-order-Ambisonics-Grond-Kelly/71f43f005a8a1b9de45163c0d686beb5e4902b8e

Here is the patent from Gerzon and Craven.

US4042779A - Coincident microphone simulation covering three dimensional space and yielding various directional outputs - Google Patents

A microphone assembly for simulating a plurality of coincident microphones has at least four microphone units mutually disposed at the integration points of an integration rule for the surface of a sphere. In one embodiment, four such microphone units are disposed on respective faces of a...

patents.google.com

Final sentence in the claim ends with:

........the same frequency response to sounds from all directions up to a limiting frequency and are substantially identical to the outputs obtained from a plurality of coincident microphones.

Core Sound just posted something interesting about this on their FB page.

--

Which Sounds Better: Coincident Microphone Array or Spaced Microphone Array?

We recorded a 60-voice choir performing in a great sounding room. We recorded the performance out in the room rather than with spot mics on the stage.

Does recording with a coincident array using a single higher-order ambisonic microphone better recording with a spaced array using two higher-order ambisonic microphones?

The coincident array was decoded to Blumlein (virtual coincident array of two crossed figure-8 mics rotated 45-degrees).

The spaced array was decoded to LCR130 (three virtual hypercardioids spaced at 14 cm and angled at -65, 0 and +65 degrees).

You decide!

Blumlein

Half Acre - Blumlein.wav - Shared with pCloud

Store, share and listen to music with pCloud. Access on any device. Create a free account now!

u.pcloud.link

LCR130

Half Acre - LCR130.wav - Shared with pCloud

Store, share and listen to music with pCloud. Access on any device. Create a free account now!

u.pcloud.link