Topic: Recording the audience

[marcb]

Does anyone have experience running a SBD mix with mics pointed at the audience to pick up the “crowd”. At the big broadcast shows I sometimes see a mic(s) on stage pointed toward the audience and/or a mic (s) at the FOH mix. If so, I’d think the on stage mics are some kind of directional and the ones by FOH is an omni? For anyone with knowledge, how are these tracks handled in post, are their levels typically increased (decreased) between (during) songs?

[BlueSky71]

I think they are generally shotguns, so that they do not pick up anything super close, they are aimed up and over the front rows. I think you may want to invert the phase on them in post. As for levels, someone else will chime in.

[rocksuitcase]

The basic answer to your post-processing question is YES. The mastering engineer will use those tracks to "enhance" the audience sound at their discretion.

depending on PA company, and FOH engineer, the single mic you see at the mix position *might* be a measurement mic for their SIM or SMAART system. SIM systems will assist the FOH to "balance" out the PA with (often automated) parametric equalization.

https://timobeckmangeluid.wordpress.com/2013/07/21/meyer-sound-sim3-vs-rational-acoustics-smaart7/

https://www.av-iq.com/avcat/ctl1642/index.cfm?manufacturer=meyer-sound-laboratories&product=sim-3

[jlava]

might help
https://www.prosoundweb.com/capturing-the-energy-of-live-shows/

[morst]

I've set up multiple shotguns pointed out to the crowd on a stadium show.
One set was for recording, and one was for in-ear monitors, but I don't recall which was which of the short or longer shotguns...

[roffels]

Late last year, I ran some cardioids on stage facing the audience to capture the audience and room sound. I really enjoyed how that one came out, but the venue was very much an integral part of part of the performance and I loved the ability to mix in the reverb to my heart's content. I also was working with stems (not proper multitracks - vocals from multiple performers were on one track for example) so I had a little more flexibility in the balance than I would have from a typical board feed+audience capture.

If the performance was critical to me, I'd capture board, something FOB (if I knew the room well enough, omnis, if not, both cards and omnis), and something from on stage. My FOB cards are usually just a backup should something else go wrong.

[goodcooker]

I've set up multiple shotguns pointed out to the crowd on a stadium show.
One set was for recording, and one was for in-ear monitors, but I don't recall which was which of the short or longer shotguns...

This is exactly it for side stage shotguns - ambience/audience reaction for recording or in-ears or both.

The mic you see by itself at FOH (if it's a skinny pointed omni mic) is for gathering info for measurement for the sound system.
If you see more than one it's probably a band doing what we do - capturing the room for recording.

[marcb]

Thanks to all for the very helpful info!

[Scooter123]

When I transferred the 2000 or so Westwood One shows a couple years ago, most tapes had a track which was just crowd noise.  When I listened to it in detail, it was two segments, one for the mix when music played and one between tracks, each about 30 seconds long and each looped.  Biff Dawes was the engineer for most of those, and since each track was mixed really low down into the mix, 30 seconds was fine. 

[Twenty8]

Recently I taped Phish at Orion in Huntsville, AL.  While I don't really recall the setup for on-stage (mics on each side of the stage, 8 feet or so, unable to recall if it was on mic per stand or two), there were additional mics set up above the board in the camera well.  2 sets of mics on each side of the well, pattern unknown (short shotguns or cards would be my guess as they had Movo windscreens on that blocked further visual confirmation and I was below them on the floor), spacing seemed to be wide ORTF or NOS.  These particular mics were not level with each other.  What I mean is, while both sets of mics were at about 7 feet, they were off-set in height as if on Shure a27m mounts.  There was also a dummy head next to the sound engineer, Garry Brown.

I only wish these aud mics were blended nearly the entire show, if only very low in the mix, as I consider Phish official sbds to be super dry and not sound like the show in a way that other band's do with the same intent on matrix blending.

[Gutbucket]

I run an additional audience facing microphone or pair for every recording I make.  Here is what I've learned-

The most important thing regarding a microphone pair specifically dedicated to audience reaction and/or room sound is minimizing pickup of the PA and on-stage sound in that pair.  The less PA and on-stage sound that makes its way into that pair, the more useful it will be in the mix.

The second most important thing is for that pair to be time-aligned with your main pair, or just slightly behind it. 

If you place this pair on the same stand as your main pair except facing rearward, that automatically achieves good time-alignment.  This works well if your recording position is relatively close such that you are near the center of the audience or closer.  It works great on-stage or at the stage-lip. Use cards or supercards (supercards provide a bit better attenuation across the full rear hemisphere, especially if the config includes some angle) with just a bit of angle between them.  Too much angle lets more sound leak in from the front, especially if using cardioids.  Use spacing more than angle between this pair to generate the desired stereo width. If your primary recording is already wide sounding, you need not space them very wide. About 12" of spacing between them works well for me.  If your primary recording is not that wide sounding to begin with, spacing the audience pair more widely can provide good decorrelation and spread the audience out to the sides.  Think of them more like AB spaced omnis, except using cardioids or supercards facing rearward.

If recording from farther back in a good acoustic space this can provide good room sound but the audience reaction back there will tend to be far more lack-luster and prone to distracted conversation.  Better to have them on-stage or at stage-lip in that case and delay or time-align them as needed. The best audience reaction is up front.  One trick I've found with them at the stage lip is to hang them under the stage facing out at the audience.  This uses the stage-lip itself as baffle to additionally attenuate pickup of on-stage and PA sound more than just relying on the directional pattern of the mics themselves.

[Gutbucket]

Late last year, I ran some cardioids on stage facing the audience to capture the audience and room sound. I really enjoyed how that one came out, but the venue was very much an integral part of part of the performance and I loved the ability to mix in the reverb to my heart's content. I also was working with stems (not proper multitracks - vocals from multiple performers were on one track for example) so I had a little more flexibility in the balance than I would have from a typical board feed+audience capture.

If the performance was critical to me, I'd capture board, something FOB (if I knew the room well enough, omnis, if not, both cards and omnis), and something from on stage. My FOB cards are usually just a backup should something else go wrong.

^This.  The comment about getting control over room 'verb and how much audience reaction is emphasized is spot on.  One of the reasons I love taping using multichannel microphone arrays is that I gain far more control than I would've otherwise had. EQ and parallel compression applied to this pair further increases its utility.  The burden of this blessing is deciding how complicated to get in the mix, it can be difficult to decide how much time to put into it once you get a taste of what it can do.  Done well it can sound very "you are there" immersive with the audience reaction seemingly coming in back, way off to the sides, and all around you.

[Scooter123]

Biff Dawes used a single shotgun at the stage lip pointing back to the sound board.  Very little music entered that track.  Alternatively, he also used a single shotgun mike at the soundboard, again facing back, with better results of eliminating any music.  He used the same or similar audience tracks without any music on them in a loop, which ranged from about 30 seconds to a minute.  No need to time align. 

[Gutbucket]

Yes, a single channel can work much better than folks may assume.  I employ a single rear facing channel in my 4 channel non-open recording rig to quite good effect.

And yeah, there is a trade off between placing rearward facing audience mics farther back in the room verses up at the stage- better in that there is less PA/stage level to reject, worse because there is less engaged audience enthusiasm behind that location, in contrast to up front where the PA/stage sound is louder and more difficult to minimize pickup of to the same degree, yet with much more engaged audience reaction.

The idea of looping the audience sound is kind of funny.  I get it for use as an ambient bed.  Yet that will produce no correlation with the rise and fall of audience reaction to the music itself and the acoustic interaction of the room with the performance, both of which are a big part of the magic a rear-facing channel or pair provides when it works well and is appropriately incorporated into the resulting recording.

Parallel compression on the audience-facing channels is useful because it effectively brings up the acoustic environment bed level details during the quiet parts such as between songs, yet smoothly reduces the relative contribution of these channels as the music gets louder.  That keeps everything cleaner and avoids it getting muddy and overly reverberant during the energetic parts.  It's also an effective way to achieve a bit of reduction of overall dynamic range in a way that is easily managed and sounds particularly natural, as the dynamics of the primary channels are not being effected at all.

[Edit- I've not tried side-chaining the parallel compression applied to the audience/room channel(s) such that it is being triggered by the level of the primary channels instead of the level of the audience/room channel(s), but that would seem to be a good approach.  That way the level of the music itself is the control rather than the compression reacting to any loud sound originating from behind. Might try that next time.]

[Scooter123]

I am not advocating the Biff Dawes technique, merely advising interested parties of how one very famous engineer of live music handled the audience feed in the Westwood One radio shows.  His focus was on the performance, not trying to correlate simultaneous audience reaction to a particular event occurring on the stage. 

If that is the goal, here is what Biff Dawes did in one example that I transcribed--the Doobie Brothers at Universal.  A very good show but for the encore, Kenny Loggins came out.  What Biff did was basically increase the gain of what I perceived to more of pre-recorded audience noise (screams, hooping and hollering) and then he brought it down.  One minute later it seemed to me that the audience track was the same scream, hoop and holler loop, but at a way lower gain.  Indeed, he seemed to have about a dozen pre-recorded loops for his mixes. 

Much of Westwood One's stuff was cherry picked and sampled to bring the best recording out the listener.  For example, he always seemed to record multi-date shows at a single venue, Universal Amphitheater was one, and an arena in Concord were two favorites of his, and he would stitch together a single radio show from the best of two or three performances at that same venue, and indeed the track sheets noted that the artist had the legal right to pick and choose what tracks went on the final radio show, and some songs were omitted altogether for unknown reasons.  Then he would stitch in the voice over and audience sound, the latter being very deep in the mix.  Those shows rarely presented the whole single performance and while his technique might seem "fake" to some, the management apparently thought the listener would rather hear the music (and the best versions at that venue) rather than the audience sound.