Finding the Sweet Spot
I teach audio engineering at the California State University in Dominguez Hills. Every spring the students form small teams and spend the semester recording a band of their choice and then preparing a DVD or Blu-ray disc containing stereo and 5.1 surround mixes. Today was the first day of “crits” on the these projects. It is allows enjoyable to see the students work through production problems, organize the sessions, work with the ensembles and deliver their best efforts. Sometimes they’re brilliant and other times they miss the mark, but the efforts are always worthwhile and the educational value unquestionable.
As I listened to the new crop of tracks this morning, I noticed that the mixes of the first group delivered a distinctly different experience for each of the tunes. I asked each of the four team members about their process and learned that each individual was responsible for one of the songs. One person did the production, engineering, mixing and mastering in isolation. They only heard the other team members work when they got together to master the whole project.
What I noticed was the particular placement of the lead vocal in the 5.1 surround mixes. The usual place for the lead singer is in the center of the front array of speakers. This spatial placement can be accomplished in a number of ways. First, a mixing engineer can simply pan the vocal track to the center speaker. Presto! The sound of the lead vocalist is now coming from a single speaker…the center in the case of a surround mix. This method was employed by one of the members of the first group of students. The next mix had the vocalist evenly distributed between the left and right front speakers with an attenuated amount in the center speaker. This is called “phantom center” and is the standard operating procedure for all stereo mixes. It is, in fact, the very essence of mixing in stereo. Without the ability to pan between a simple pair of left and right channels, records would lack the depth and spatial spread that makes stereo so much of an improvement over mono. The third mix didn’t include any signals in the center speaker at all…nada! The phantom image of the lead vocalist was in tact but without the support of the center speaker. And finally, the last member of the group put the vocalist in all 3 of the front speakers equally creating a seamless vocal presentation across the entire front of the studio.
I sat in the sweet spot with my other colleagues to my left and right. We all experienced the tracks on their DVD from a slightly different perspective. So which one produced the best stereo image? Which one had the greatest amount of depth and spatial accuracy? Let’s think about it.
Figure 1 A typical ITU 5.1 speaker setup. Note that the speakers are equidistant from the center listening position, identical AND pointed towards the center.
In a perfect world, all of the speakers in a playback system are identical and placed along the circumference of a circle. This is rarely the case in reality. Even professional studios and custom home theaters have different speakers and sometimes they’re in the wrong place according to the ITU 5.1 standard. Even Robert Harley, the editor of The Absolute Sound, in his book “The Complete Guide to High-End Audio” has them in the wrong place in his chapter on multichannel audio! No wonder he complained about there being too much direct sound coming from the rear in my “stage” perspective mixes.
Placing the lead vocalist alone in the center speaker is not a good idea in my experience.
So in reality, a mixing engineer cannot count on the center speaker being of the same quality as the left and right stereo pair. Placing the lead vocalist alone in the center speaker is not a good idea in my experience. Putting the singer equally and only in the left and right speakers does a good job of creating a nice “phantom center” image but then the center speaker is left with nothing to contribute. I’ve actually had customers call or email complaining about the fact that some of their surround music doesn’t have anything in the center or the sub. Heck, they purchased a nice center speaker and they want to see it in use. Not the best reason for sending something to the center speakers. But I do include the center in my mixes but at about 6 dB lower than the left and right.
How about the student mix that had all of the front speakers firing with the lead vocal? This is problematic because it can overemphasize that part and/or diminish the spatial accuracy of the whole mix because there are too many copies of the same signal coming from too many places. It is a judgement call but I’ve found that using phantom center with a little piece in the actual center speaker works really well.
I asked the students in the group this morning which version they preferred and they all liked their own mixes…no surprise there.
Having a good deal of experience mixing in a stereo environment (hopefully using the right techniques…), I have no experience in mixing in 5.1 surround. It’s reassuring that stereo techniques, like creating a phantom center, are used in 5.1 mixing…it’ll be a little easier to tackle that 5.1 beast when I get there!