The 5.1 speaker arrangement specified for home theater viewing of movies can never be the successor to stereophonics where high caliber reproduction of music is concerned. The standard 5.1 speaker arrangement of left, center, and right speakers up front and two speakers toward the left and right rear (plus the new one at the rear center in 6.1) may be adequate for moving pictures and is often better than just plain stereo, but it can never be more than barely acceptable for the reproduction of classical music, jazz, folk, or acoustic rock, hip-hop or other staged forms of traditional musical performance.

In a movie, the scene changes every few seconds and direct sounds can come from the front, rear, sides, back or overhead. Thus, a movie sound system in a theater or in a home must provide, as far as possible, for direct sound to originate from any direction. Also the acoustic space portrayed in a movie may vary from a room in a house, to the outdoors, to a restaurant, to a car, to an office, to an Olympic stadium, etc. so quickly that the viewer cannot and does not really consider whether the ambience of the scene is truly realistic. Since the movie viewer has never been in a space ship, anything goes surround sound wise.

In contrast, musicians usually stay put in a concert hall, a church, an opera house or a Broadway theater for the duration of their performance. Home listeners know what such spaces sound like from attending concerts, plays or church services. When they listen to music over a home theater system they can immediately sense that they are not in attendance at a real live musical event because they now have minutes in which to sense the space and they have a frame of reference to relate what they hear to a similar real life experience.

Traditional music lovers, some of whom may also be audiophiles, know that the stereo triangle, which has been with us since 1931, does not sound very real no matter how many tweaks we apply. Where live music is concerned, one reason, (among many) is that realistic concert hall ambience is missing. When one hears a symphony orchestra in Carnegie Hall one receives early reflections and reverberation tails from just about every hall surface. But in standard stereo reproduction, any recorded hall ambience always comes from just two points in the front, plus any bogus short delay listening room reflections, and this naturally sounds unrealistic. Futhermore, recording engineers cannot allow too much hall sound to leak into the standard stereo channels because this makes the reproduced sound seem like the recording was made in a sewer.

The two rearward surround channels provided in Dolby Digital and DTS for movies is meant to allow for direct sound such as gun shots or a telephone ring to come from the extreme side or the rear. In the case of music these speakers are often used to deliver concert hall ambience. However the idea that just two speakers to the rearward sides can mimic a great hall is ludicrous. Two ambience speakers, plus stereo up front can be better than just stereo but that is not saying much. However, the usual surround multichannel recording system actually provides for six full range channels. This is true for Dolby, DTS, DVD-V, the new MLP 96/24 DVD-A, and multichannel SACD.

The question is, can these six channel delivery media be used to better satisfy the expectations of audiophile music lovers. One of the home theater channels is usually assigned to carry low frequency movie effects such as earthquakes. For music it is sometimes used to carry low bass to a subwoofer. However, in practice this full range channel is generally under-utilized. There is really no reason why it cannot be used to carry concert hall ambient information and feed a third surround speaker.

A front center channel is used in movies to anchor dialog to the screen, even for viewers not centered in the home theater room. This makes sense in movies where dialog is on a separate track and can be isolated and fed to a center speaker. However, in the case of acoustic music recordings, there is really no LCR microphone arrangement known to mankind that can utilize this three-speaker arrangement without problems. The only advantage that using a center speaker has in the case of live music (not three channel mono or virtual reality) is to slightly enlarge the sweet listening area at the expense of imaging broadening, spurious phantoms, comb filtering, etc. Thus, where music is concerned, we might as well stick with the stereo triangle but improve on its lack of concert-hall realism by using the center channel for a fourth hall ambience channel.

Two front speakers and four surrounds go a long way toward delivering a reasonable facsimile of a concert-hall, opera house, or rock pavilion experience especially when compared to 5.1 or plain stereo. Now the problem is how to make six channel recordings. In particular the question is how to capture four realistic channels of ambience and deliver them to our domestic concert hall. First, since we have only four ambience speakers it is a good idea to put them where they will do the most good.

There are as many opinions as to what makes a good concert hall, as there are halls. But some of the best concert halls deliver early reflections from 55-degrees to the right of center and 55-degrees to the left of center. This makes sense since at this angle there is a large difference in the signal between the ears because the head is in the way and the path around the rear of the head is not as accessible. The human brain is fond of interaural differences and listening is more exciting when there is interaural interest. For this reason it is not a good idea to provide an ambience channel directly behind or directly overhead, since the signals reaching both ears in these cases would be identical. So let us put one pair of our surround speakers at these 55-degree locations. Remember, ambience is not critical. Moving the speakers, a lot sideways or up will make little difference and is akin to changing your seat in the hall.

The other two speakers are best left where the movie people put them at 110 to 120 degrees at the side rear. Again, the exact position is not important. What is important is that there is no scientifically valid method known as to how to use microphones during the recording session to record the hall ambient signals to feed these speakers. (A technique called Ambisonics cannot easily be used here if the front speakers stay stereo or a really precise hexagonal speaker arrangement is not used.) While there are too many reasons to list as to why the recording of hall ambience during the recording session cannot be done to audiophile standards, I will discuss two.

To get a wide symphony stage for stereo reproduction and avoid echo effects it is usually necessary to place the main microphone close to the orchestra. Also the main microphone should not pick up too much rear hall ambience since this should be coming from the surround channels for realism. You don't want to put the hall ambience pickup array too close to the stage, since it will then be contaminated with too much direct stage sound. So now we have the stereo speakers delivering a 2nd row center perspective while the ambience speakers are delivering some direct sound and the ambience picked up from say the 25th row. (If you put the ambience mic's too far back you begin to get echo effects because of the direct sound pickup.). Of course you can delay the stereo signals to match the ambience, but by how much, and there is still some direct sound which will cause comb filtering distortion and false localization cues if the delay between front and back is too short. Secondly, and more importantly, this is no way to preserve the directional information of the ambience. The reflections are all lumped together by the microphones and there is no way later on to feed left coming early reflections originating from the right half of the stage to a left side surround speaker and so on.

Fortunately, there is a better way. Concert halls do not change their stripes during a performance. That is, the equations that represent what a hall does to a sound from the stage are always reasonably the same for years at a stretch except if the size of the audience varies. If we measure the characteristics of the hall and get these equations for the best seat in the hall and the left and right sides of the stage, then we can use these equations to calculate the hall ambience level and direction that will arrive at a listener in the hall or their home from the surround speakers we have installed. This mathematical operation is called convolution. Professional convolvers from Sony and Yamaha are available to recording engineers. They come with a variety of already measured halls or with software if you want to measure your own hall and store it in the convolver.

Someday audiophiles will be able to download the impulse responses (equations) of all the world's great halls and play back two channel recordings in the hall they prefer for the particular piece of music. Consumer friendly ambience convolvers will sooner rather than later become as ubiquitous as surround sound processors. Home convolution would allow the use of even more than four surround speakers and provide elevation. However, at the present time, it is necessary to let the recording companies provide the convolved ambience.

Thus, we can summarize what it takes to make a superior and more realistic six-speaker surround recording of music. The recording engineer makes the best stereo recording one can, taking care not to allow the front stereo or any spot microphones to pick up too much rear hall ambience. He/she then convolves the left signal to produce the hall ambience that music from the left of the stage would produce and feeds that, together with the ambience that music from the right side of the stage would produce, to the appropriate speaker. This calculation is then repeated for the other three speakers. Chesky Records has already produced several DVD-A's using this advanced technology and the results resonate for themselves.