Broadcast Mix Part 2

Here’s the process we now take to create the mix…

We now use three sets of ambience/audience mics, each with a different purpose for capturing the room.  Inspired by some great discussion on the Digidesign User Forum, my implementation is…

• Mics #1 are a set of shotgun mics (purchased for this project) on the outside corners of our stage, mounted just under our side screens, pointed out into the room.  These are the primary audience pickup source because so much of the audience’s energy is directed towards the stage that these do a great job of capturing without adding too much PA to the mix.  
• Mics #2 are a set of Crown PCCs placed on the lip of the stage at the 1/3 and 2/3 lines from left to right.  These combine with the shotguns to provide some presence and immediacy to the signal since another set of shotguns wouldn’t be acceptable aesthetically in these positions.  
• Mics #3 are a pair of Shure SM81s hung from the first catwalk, almost at the half way point of the room, also on the 1/3 and 2/3 lines to the stage.  These mics serve almost entirely the purpose of room ambience – making the mix sound like it is happening in the auditorium rather than happening in a dead studio.
• We may still look to add an additional set or two but I’m stuck because adding them will mean needing to deal with delay times between those mics and the rest currently in place.  Two of my current three sets are directly in line with the PA so there’s little issue there.  The delay induced by the third set certainly does some phase stuff to the mix but the added space it puts in the mix when the speaker is up is worth the compromise to me.  So we’ve dodged the bullet to this point and I’d like to avoid having to go there with more mics.  This is one of the only limitations of creating the broadcast mix inside the Venue rather than in an outboard mixer/processor.  There is no way to have the degree of time alignment control that would be needed in the broadcast feed without making other more significant sacrifices.

All channels on the console are split up into one of three main bus paths – master L/R is music, center is speech, and subgroup 7/8 for standalone playback sources.  These three paths are then combined in the matrixes to feed the PA and outside world.  Our PA processing is also set up for speech and music, so there is a stereo music matrix (with playback included in it) and a mono speech matrix that feed the PA.  On the video side, music, speech, and playback are combined with the three sets of ambience/audience mics in a stereo matrix.  A nice Massey mastering limiter, phoenix cranesong, and 7 band EQ for any trouble spots are put over that mix as a whole and the finished result is what you hear below.

According to Smaart, music averages at 88-95 dB in the auditorium, speech and videos run 68-75 dB. That equals an average 20 dB natural difference that feels right in the main auditorium. In the processing of audio for video, I am actually adding back 6 dB cumulatively to speech, making the speech/music difference 14 dB rather than 20 dB and this is really important.  Too little added back makes speech too low compared to music, so you have to turn it up.  But too much added does the reverse – turn it down when the speaker comes up.

That is why playback has to be split up from music in the broadcast matrix.  If playback is lumped in with the band, it comes out too low on video compared to speech because playback is in that lower dynamic range in the room.  But if playback were lumped in with speech directly instead, it would be in mono and receive all kinds of undesirable processing in the speech world that isn’t desired.  So…on the PA side, playback and music are each sent to the music PA matrix at equal volume.  But on the video side, playback is sent to the matrix at the same level as speech but through its own path.  So if speech and playback were sent to the matrix at 100% level, music is sent at 90% level.

Here’s a sample of the finished result of worship from our New Community service last night…

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

In my next post I’ll outline the final steps necessary to fine tune for building feeds vs. broadcast.