5.1-channel Sound: From the studio to your home theater
After years of stereo mixing, mixing for surround can be a creative challenge.
By the end of the last century, recording engineers had pretty much figured out how to mix music for stereo playback. We knew what worked. We knew where things should be panned. We had a good grasp of what types of speakers people were listening to. We knew, generally, how people set up stereo speakers in their homes. Stereo mixing and stereo playback was a done deal. With the advent of surround sound, be it for movies or music, everything we knew has gone out the window. In these still early days of mixing for surround, there seem to be no rules, no set-in-stone standards. There are as many approaches to a mix as there are engineers.
Let's take a look at what challenges 5.1 creates in the studio. How does it differ from mixing in stereo? Why is mixing for the movie theater different from mixing for home playback? What works and what doesn't? To get a better understanding of the intricacies of some mixes, it might be time to start experimenting! I know I'm not the only listener who unplugs speakers so I can hear exactly what an engineer is doing and what is going on in a particular channel we all do it, right?
For the most part, we'll use music as our source, because it causes the most problems for mixers, and also opens up a world of new opportunities. Movie sound effects are typically tied to the picture if something is making a noise on screen right, then that's where the sound effect must be placed. Dialogue is almost always in the center channel. A space ship flies in from behind the camera across to the front left side of the screen, and the sound must do the same. However, anything is possible in a music mix.
How is mixing in 5.1 different from mixing for stereo?
Ah, the simple joys of a stereo music mix. Vocals dead center. Reverb panned hard left and right. Kick, bass and snare also securely in the center. The tom-toms spread across the stereo panorama, hi-hat hard right. The other instruments scattered across the soundstage, so if you close your eyes, you can point to where each player was onstage. The audience in a live recording is also panned to the full stereo width, hard left and hard right. If it is a classical recording, it's even simpler. Each instrument is placed exactly where they would appear on stage violins on the left, cellos on the right, and so on.
A typical surround setup (left) has 5 speakers and a subwoofer, compared to a two-channel stereo setup (right). All those extra speakers and that enveloping sound mean that a 5.1-channel source gets mixed differently than a 2-channel source.
Now, don't think for an instant that this makes mixing completely simple. Because the sounds are coming from just two speakers, room has to be made in the mix for each instrument. Frequencies play a great role. The engineer has to make sure that each instrument occupies a unique sonic space, to differentiate it from all the other instruments coming out of those two speakers.
Also, on the plus side, quite a lot can be hidden in a stereo mix. The perceptual concept of masking plays a big role here. If there is some noise on a track, some distortion, and even some tuning issues, these can be covered up by everything else blending into the two channels of a stereo mix.
Now, consider what happens when that mix, with each of those little indiscretions, is spread out across six speakers. In a stereo mix, a background vocal that was slightly out of tune caused a nice chorusing effect in the stereo mix. Now, that vocal part is sitting by itself in a surround channel, noticeably off pitch. Or, the amplifier noise on a guitar track (that was masked under the rest of the band in the stereo mix) is humming away all alone in the other surround channel.
There is another problem that is encountered frequently. Whenever I start a session, I always ask the producer where this mix will be played. Is it for playback in a theater? Is it strictly being played over a television set? A huge ballroom? Knowing this helps me determine what to mix for and what speakers to use to monitor my mix, either large or small. With a DVD release, the engineers never know if it's going to be played over a custom-tuned, THX-approved $80,000 home theater or over a $250 home-theater-in-a-box. While this has always been an issue with a stereo mix, the number of speakers in a 5.1 setup multiplies the complexity.
There are a lot of different center channel speakers out there! Sound engineers can never predict what kind of system their mixes will be played on.
The center of the issue
The use of the center channel in multichannel music is the cause of much concern. Engineers might find themselves physically attacked by irate singers if the lead vocal was placed dry (without reverberation) in the center channel, with the reverberation only in the front left/right speakers. This would mean the vocals would be naked and exposed, without reverb, for anyone to hear in the center channel. Most singers' voices need that reverberation to smooth out pitch problems and soften up the tone. (Did you ever wonder why your own singing sounds so much better in the shower? It is because the natural reverberation created by the hard tile surfaces smoothes out your own pitch problems.) There are some exceptions, but only when the singer's talent can withstand it; in Sting's Ten Summoner's Tales (DTS Entertainment), the lead vocals, without reverberation, are placed in the center channel. In a typical stereo mix, the reverberation that engineers add is a wide stereo program, so for multichannel mixing, it isn't as simple as adding all of the reverberation to the center channel vocals. Even for very good singers, most engineers will add a small amount of reverberation to the vocals in the center channel just enough to take off the edge. Gary Paczosa does this with Alison Krauss's voice, whether it's the studio recording New Favorite (Rounder Records) on SACD, or a concert recording, Live (Rounder Records), also on SACD.
In some surround mixes, engineers don't want to use the center channel speaker at all. That's because there is a wide range of center channel speakers being used in surround systems. The problem is especially true in the all-important car playback market. In a mobile environment, the center speaker might be a tiny, low-powered speaker mounted in the head unit. To get around the center-channel speaker, engineers will sometimes create a phantom center image, which is how the voice appears to come from the center of the stereo panorama in a stereo mix. By using the same signal at exactly the same level in both front speakers, it will appear to be coming from the center. This technique can be used quite effectively in surround mixing. Listen to Buena Vista Social Club (Nonesuch Records) on DVD-Audio to see how pleasant it can sound.
The center channel can also be problematic in home playback. Some people (we won't name names here, but you know who you are) use the speaker built into a television for the center channel. While it might work for the dialogue in a movie, where the intelligibility of the spoken word is all that some folks listen for, it doesn't work in music playback, when the all-important lead singer is there. The richness and fullness of a male vocalist would be destroyed. On the other hand, some engineers throw caution to the wind, and heavily use the center channel. Listen to the center channel during the song "Hotel California" on the Eagles Hell Freezes Over DVD-Video concert (which should be in everyone's collection). The kick drum and bass guitar are mixed to the center channel, along with Don Henley's clean, dry (reverb-free) voice can your center speaker handle that? Engineer Elliot Scheiner was betting it could. Like I said, when it comes to surround music mixing, there is no consensus.
The Great Bass Debate
One of the most disputed aspects of mixing in 5.1 is how much bass to use, and where to put it. Should an engineer directly place bass content in the discrete LFE (.1) channel? Is it better to place it in the satellites and rely on bass management circuitry (tuned for that particular system) to direct low frequency information away from the satellite speakers and filter it to the subwoofer? The problem is that some people have large front speakers with no bass management employed. Others use full-range front speakers and a subwoofer with bass management to convey the real deep bass that the main front speakers cannot produce. Still others use small satellites and rely on the subwoofer to produce all of the lows. A mix needs to have the right balance that will sound good on all systems.
What most engineers tend to do, for movies and music, is to show restraint when it comes to using the discrete LFE channel. Reserving it for the kick drum, bass, and deep movie sound effects ensures that it won't cause problems in the home. Engineers just have to be careful about what they put in the other channels that might sum together to an overwhelming level in the subwoofer. A little low end in each of the main channels could add up to too much bass from a subwoofer.
The cavernous space of a movie theater poses a very different sonic challenge to good surrouns sound than the average home's living room.
What are the differences between mixing for the theater and mixing the DVD release?
There are quite a few things that need to be considered when remixing a theatrical release for DVD release. One is the amount of bass that is used. While a theater can pump out as much low end as the audience can stand, people at home might need to be concerned about overall volume. To keep the subwoofer from bothering the neighbors or the sleeping kids upstairs, mixing engineers might bring down its overall level.
Another problem is dynamic range. In a quiet theater, a movie soundtrack can be alternately loud and soft. But that can be a problem when playing back a DVD soundtrack at home. When a soundtrack has extremely loud sections and extremely soft sections, people at home may wind up constantly adjusting the volume. One problem is that most people (contrary to what we all would like to think) don't listen at as high a volume as in a theater. The other problem is that homes are noisier than a theater the softest dialogue needs to be audible above the dishwasher and the lawnmower down the street.
The front left and right channels in many movie theater systems do not convey frequencies below around 40 Hz, and the engineers mixing the movie don't hear it on the mixing stage, because their main monitor speakers have the same roll-off. However, many home systems use front left and right speakers with plenty of response below 40 Hz, or else the system's bass management directs front left and right channel bass content to the subwoofer. That means a home system might reproduce information that the original mixing engineers had no idea was down there. This can be information that was in the front left and right monitor speakers but was lower than they could produce. Your bass management system, rightly, has taken those frequencies to the subwoofer, which has no problem producing them. But, you end up with too much bass, and moreover, muddy bass. In either case, it's not the mix the mixing engineers intended. Thus, when mixing for a DVD release, engineers must start over, and rebalance everything on playback systems that are more like home playback systems.
What are engineers doing in the mixes?
At this point, in the early stages of the technology, mixing for 5.1 has no rules. Engineers are experimenting, seeing what works, and learning the hard way what doesn't. For starters, a lot of equalizing falls by the wayside; there is no need to clear out a 'sonic place' in the mix for an instrument by focusing it into one frequency range. If it is getting lost in the mix, either move it or other instruments to a different speaker. Problem solved! The same applies to dynamic range compression. To keep all the nuances of a performance audible in a stereo mix, engineers would have to apply dynamic compression to even out the levels. Again, they can now move an instrument out of the way of competition and keep all of the original dynamics. Part of this is also due to the high-resolution recording formats being used these allow for greater transients and dynamics.
In live recordings, the audience tracks play a greater role in surround mixes. In traditional stereo mixes, this track had to be used judiciously, and brought in only when necessary, at the beginning and end of a song or after a solo, for example. In a surround environment, the audience, if panned to the rear surround channels, can remain up throughout the entire song. Many live recordings use the surrounds strictly for audience ambience. This is evident on Bruce Springsteen's Live In New York City (Sony/Columbia) where the audience remains at a fairly high level in the surrounds. There's only a little bit of ambience and a few keyboards mixed into the rear channels.
There are clearly different theories of mixing live music in surround sound. One is to put you in front of the band (engineers like to aim for a seat in the 20th row, center of the house) and the other approach puts you on a bar stool in the center of the group. This latter surround feel is used in the Buena Vista Social Club DVD-Audio. The former approach is used in the Alison Krauss + Union Station Live recording.
Engineers sometimes do funny things. For instance, the entire band is mixed in the center channel on Disturbed's Believe (Reprise Records) on DVD-Audio. This causes the overall mix to sound very narrow and focused on the mono center speaker. The DTS CD release of Santana's Abraxas (DTS Entertainment), remixed for surround, goes to the other extreme. The number of instruments swirling around the surround speakers might have worked in the psychedelic 60s, but today, it just gives you a headache.
Other times, there are strokes of genius. Check out the movie soundtrack of Magnolia (New Line Home Video) which features songs by Aimee Mann. In Chapter 9 (at 2:18:56), the song "Wise Up" plays a pivotal role, rising from background music to center stage (so to speak) as each of the film's main characters sing along with the track. It starts out with the entire track in the center channel. As it shifts from one scene to another, the instruments and the reverberation on Mann's voice pan out to the front left and right speakers, keeping the character's vocals in the center channel. Throughout the song, vocal reverberation first, and then the instruments, slowly enter the surround channels as well, wrapping you in the evocative moment. Rather quickly, as the song reaches its somber climax, the mix collapses back to the front channels. This is a subtle, yet beautiful use of the surround soundstage.
Other mixes are just good showcases for a truly engaging surround experience. Big Phat Band's Swingin' for the Fences (Silverline Entertainment) on DVD-Audio is a great disc to show off a well-designed system. The big band is spread across the entire sound field, placing you in the middle of the action, with a massive drum sound filling out the low end of the spectrum. This disc uses a phantom rear center signal the soloing instruments are placed in the center channel as well as equally in the rear surround left and right speakers, to give a very solid, balanced performance, no matter what type of speaker you use for the center channel. Aaron Neville's Devotion (Silverline) is another great piece to showcase your system. The song "Mary Don't You Weep" features a male chorus, with a group of vocalists in each of the four channels. You feel like you are standing among the singers, literally surrounded by the men.
As for movies that completely blow away the listener, my current favorites are The Lord Of The Rings: The Fellowship of The Ring (Warner) and Spiderman (Columbia/TriStar). The prologue of The Fellowship has an amazing low-frequency response, along with detailed center channel information. Spiderman is an exaggerated comic-book over-the-top mix. The fight scene in Chapter 27 has very exaggerated, discrete sound effects in all of the channels, and the beginning of the scene lacks music scored underneath, so the sound effects are distinct and clean.
Bringing it home
No one can say for sure if there will ever be a right and wrong way to mix for surround. What you can do at home is start listening and pay attention to what works for you personally. Get up off the couch and listen closely to the center channel, or one of the surrounds. Disconnect speakers. See what is going to your subwoofer. Get a feel for what works for you. Strive to make your system sound as good as possible. You might suddenly realize that your surround speakers aren't carrying their weight, or you might just need to adjust the bass management crossover frequency to your subwoofer. There is certainly a wide variety of material out there. All you have to do is to start listening.