Surround Advice – Recording In 5.1


Copyright Mike Sokol 2001 – All Rights Reserved.

For the last few months, I’ve been working with a number of broadcast sound designers that are trying to make 5.1 surround mixes out of regular stereo program material.Admittedly, some of the surround mixes they’ve come up with are quite remarkable, considering the tracks they started out with. By using a combination of joystick panning and multiple reverberation effects, these “surround” mixes have a lot of interesting movement and texture.

However, the problem is that they really don’t sound “live” to my ears. This is especially true of live multi-track concert recordings for DVD’s or broadcast, where the audience in the surround speakers sounds canned, and there are lots of impossible sonic movements in the mix that don’t match up with what the camera is seeing.

Now don’t get me wrong, I personally like a music mix with lots of pans and moves. Back when dinosaurs ruled the earth (according to my kids), I played in various Rock bands where I jumped through fireballs on stage. I was weaned on Beatles in the 60’s and Pink Floyd in the 70’s. But when the audio is in a supporting role for the picture on the screen, the sound designer must be careful nt to build a soundstage that overshadows the picture on the screen, Hi-Definition or otherwise. And this is even more true for 5.1 surround mixes, since there’s the temptation to make the sound spin around the room, whether it’s appropriate to the picture or not.

While at a recent panel discussion in Washington D.C. on5.1 surround mixing for television, I posed a simple question to the panelists. Had any of them ever mixed an A/V program which had been planned for 5.1 surround from the start of the production? Interestingly, none of them had. The original microphone tracking wasn’t designed with surround in mind, and no special surround microphone techniques were used. It seems that someone else always does the field recording, and then the tracks are passed on to the sound designers for processing and mixing. I noted that trying to make a realistic surround mix of a live performance without the tracking mics being in the right places was the hard way of doing it. With the proper placement of a few extra mics, they could have tracks that would practically mix themselves in surround, and they wouldn’t have to resort to multiple reverbs to build an artificial space.

With these thoughts in mind, here is a basic microphone technique that will be useful for recording any type of live production with an audience that will be mixed to surround. Note that on the stage all the standard close miking techniques are in use. If there’s a professional sound company working the gig, they will typically use Shure SM-58s and other mics of that level for the main sound system. You can then take a split from the stage box and go to your own multi-track recorders, typically Tascam DA-88 decks or one of the cool new 24-track hard drive recorders.

Whenever possible, use one microphone per recording track, as pre-mixing channels to fewer tracks will mean less options for a surround mix. Also, remember that for nearly every video production, recording at 48 kHz is the proper sample rate since all DVD’s and HD-TV broadcasts are at that rate. Converting from 48 kHz down to 44.1 kHz is better than trying to go the other way. To my ears, one of the things that makes surround sound of a live concert so wonderful is the ability of the engineer to put the audience reaction “behind” the listeners in their living room. This means that you can simply mix all the individual stage mics split from the PA system across the Left, Center, and Right speakers in front. Then if you had something proper to put in the rear “surround” speakers your mix would be complete. Unfortunately, without properly placed audience response mics, you’ll have no simple options beyond artificial reverb to recreate the room. It seems that many live sound engineers are confused about what to do for microphone
placement. I’ve heard mixes where there were 6 or 8 audience response mics summed together in mono and prerecorded on one track. Also, I’ve seen live shows where these audience mics were placed in the back of the room and pointed towards the stage. Both placements are wrong and will yield rear speaker mixes that do nothing but echo or phase cancel.

Here’s a simple soluon that works beautifully for these situations, and here’s how I came up with it. In home theatre systems, the surround speakers are placed to the sides, almost like big headphones in relation to the listener’s ears. So I hypothesized that a binaural-type microphone mixed to the surround speakers would provide the listener with enough phase and temporal clues that it would sound very realistic. This in fact does seem to be the case.

FritzNow, I have done a fair amount of recording with a pure binaural head of my own design similar to the Sennheiser “Fritz” which resembles a human head with microphones in its ears. And while it does work quite well, there are a few problems with it for general usage.

One: it’s a pretty expensive toy. A Sennheiser Fritz head costs upwards of $5,000.

Two: its profile tends to draw a lot of attention from the crowd and makes people want to shout at it and throw things which is a real problem when recording.

And Three: a pure binaural stereo signal doesn’t work as well with speakers as with headphones, since the crossmixing of the right speaker to the left ear (and vice versa) spoils the binaural effect to a large extent.

SASS-P_SurroundYes, you can make a binaural to stereo encoder, which makes pure binaural tracks work over loudspeakers. However, there’s a simpler, cheaper, more elegant method. I’ve been using a Crown SASS-PMKII microphone, which is a semi-binaural microphone designed for standard stereo playback. It was originally designed as a way to record direct to stereo with live sound sources like symphonies and jazz acts. It utilizes a combination of two PZM (Pressure Zone Microphone[tm]) elements on a shoe-box sized enclosure. The SASS (which is a acronym for Stereo Ambient Sampling System) can make a beautiful stereo image with lots of binaural timing cues. The real trick is, instead of requiring headphones for playback, it works well over speakers.

As shown in the figure above, a SASS-PMKII is simply placed in the center of the audience and oriented towards the rear of the room so that it “hears” the reflections of the back wall. I like to fly it about 12 to 15 feet above the crowd in whatever way works the best. Then record those two channels as the last tracks on your multi-track deck. Now when mixing to surround, simply use the SASS tracks for the rear surround channels, adding additional reverb on those tracks if the room tone was too dry to begin with, or some signal delay in the 30 to 40 millisecond range if the SASS was too close to the stage to sound BIG. Be sure to monitor for downmixing compatibility. Sometimes longer delays are better than very short ones for the rear channels, since short delays tend to make really phasey sounding cancellations during a downmix to stereo.

Of course, you can substitute other stereo mics in place of a SASS for the audience response. I’ve also tried a Shure VP-88 Mid-Side stereo mic, or a pair of AKG-535 mics in a standard ORTF stereo array. Of course, these mics are always oriented to point towards the back wall. The key is to get a rear stereo image that sounds good on its own to begin with, and then do some sonic manipulations with reverbs and delays. Like they say, you can’t make a silk purse out of a sow’s ear, and trying to make convincing surround room tone out of the wrong type of mic tracks is
just as frustrating.

Copyright 2001 Mike Sokol – All Rights Reserved

Leave a Reply

Your email address will not be published. Required fields are marked *