The Surround Sound Playback/Performance Topology

Mission

The restated mission of an idealized Surround Sound system and environment might reasonably be: To provide a superb array of musical and audio experiences in a comfortable, ergonomically conducive and flexible domestic space, without particular regard to style, genre, or specific production preferences, but encompassing as broad a range of such experiences as possible while maintaining an acceptably high level of performance for all of them.

Criteria for a high-quality domestic listening environment

Multiple uses

It is not reasonable, given the investment involved and the similarity of architectural and hardware requirements, to devise separate and unique spaces for listening to original music, to reproduced music or for viewing film and/or video. I assume, therefore, a single, multi-use space. I have even envisaged the integration of live acoustic music performance as well, but that implementation remains beyond the scope of this paper. Some fascinating possibilities exist.

Acoustical levels and isolation

The principles of acoustical isolation and the engineering requirements needed to achieve various values are well-known and understood. Suffice it to say that the noise floor, with all systems running, should be less than 40 dBA SPL Leq, and maximum levels should probably not exceed 115 dBC SPL (fast detection). Isolation of the listening space from its surroundings should probably be at least 40 dB, depending of course on the noise issues pertaining to those adjacent spaces.

Minimum and maximum dimensions

Based on personal experience with numerous purpose-built and adapted home theatre and studio-type listening rooms, I suggest that the minimum size for a successful high-quality room might be 20’ long, 15’ wide and 8’ high. At the other extreme, I have come to believe that no single dimension should be significantly greater than 50’ and that the overall floor area should be not exceed approximately 1500 square feet, in order to constrain the onset of long-term reverberance in the playback space which may interfere with impulsive transient characteristics inherent in any given recording. Ceiling height, depending on topology, should probably be no greater than 20 feet, for a total volume of no more than 30,000 cubic feet. It should be noted that such a maximum volume will require significant absorption at all frequencies to obtain satisfactory performance.

Optimum Listening area/population size

The question of the optimum listening position and area, as well as the number of people that can share that space, is a key concern that has serious implications for loudspeaker music. It is widely accepted in stereophonic practice that the optimum listening position can exist only along the median plane (with a lateral tolerance of approximately six inches). Therefore, for “optimum” stereophonic listening, it has come to be accepted that listeners should be positioned in a single column placed along the median plane. In end-user practice, this constraint is usually set aside and listeners listen side-by-side in a “near-optimum area” within a few feet of the median plane, accepting the phantom-image distortions that occur without major complaint.

With a multi-channel array, however, we no longer have a median plane, but rather only a unique median point. Therefore, the number of people who can appropriately enjoy a music performance is severely constrained by this, and may be exacerbated by some of the philosophies and approaches of recordists who prepare highly correlated multi-channel music recordings. In the worst case (where phase coherence is required up to 1 kHz.), a single individual may be the maximum number that can enjoy a satisfactory performance at a single time.

However, if we assume reasonably conventional management of correlated signals among channels and traditional musical standards of rhythmic integrity, it seems to me that the upper limit of the optimum listening area is, in general, going to be limited by the amount of time error that is tolerable for rhythmic materials used in hockets “bouncing” between multiple speakers.

I estimate the threshold of audibility of that time error, based on my own production and testing experience, at approximately 6 ms. (at which point an error in consistent rhythm can be detected). A 20 ms. error results in a distinct “loping” rhythmic offset which I believe will be found unacceptable by most musicians. A 10 ms. error seems to me to be audible but generally not annoying. This suggests that the greatest error in distance between speakers should be no greater than approximately eleven feet, which in turn suggests an optimum listening area of approximately 11’ in diameter (94 square feet). From this, we can posit that a group of up to twenty-five listeners of close acquaintance (thirteen would be better and nine would be best) might be a reasonable maximum. Consider the following graphics:

	Figure 1. The optimum listening area for a surround sound system where the maximum time error between speakers in the array is 10 ms. Rhythmic (time) and amplitude degradation increase as listeners move outside of the optimum area. Nine listening positions (in close proximity) can fit entirely within the optimum listening area.

	Figure 2. Thirteen listeners can fit “nearly” within the optimum listening area, while another twelve can fit “just outside.” This represents a reasonable maximum population that can listen to a high-quality surround sound performance at one time.

Therefore, it is clear that the population that can be musically well-served by a five-channel surround sound system similar to the one we are discussing is necessarily quite small. It is not possible, due to the errors that accumulate due to the propagation rate of sound in air, to present such music with high quality to large groups of people. Multichannel loudspeaker music is, therefore, by this primary physical constraint suitable ONLY for domestic use in small groups. It cannot be a viable public medium.

Proposed room surface treatments and supporting criteria

Management of Early Reflections and Decay

Early reflections will begin to arrive at listeners in the optimum listening area approximately 10-15 ms. after their direct predecessors, and will build up over the next 40 ms. (the nature of that buildup is strongly affected, of course, by the room volume and absorptivity at frequency). It has been my experience that these early reflections (and particularly their high-frequency content) are extremely beneficial in generating strong phantom images, depth, reverberance and envelopment.

Some constraints apply:

• loudspeakers need to be at least 6’ from the nearest wall or else installed in the wall;
• loudspeakers need to have broad and uniform lateral high frequency dispersion (and it is desirable to have constrained high frequency vertical dispersion);
• in general, the “front” wall (the wall behind the front loudspeakers) and the ceiling (at least the front part) need to be treated with broadband absorption;
• the side and rear walls need to be hard and reflective at all frequencies. It is desirable for the rear wall to include a large vertical cylindrical diffuser at the median plane in order to specularly reflect high frequencies laterally. Such a diffuser can also serve as a bass trap as needed.
• the floor needs to be carpeted to provide additional high-frequency absorption.

The goal here is to provide a rich, broadband array of lateral early reflections from the loudspeakers for at least 50 ms., but not much after that. Meanwhile, vertical early reflections are suppressed, which serves to support the asymmetrical behavior of our auditory localization mechanism. These behaviors permit the playback room to carry whatever phase-locked lateral early reflection and reverberant artifacts exist in the recorded music without significant interference.

Management of Standing Waves and Room Modes

It is important to account for the room modes of the playback room. In general, in rooms with these dimensional limits, the widely spread lowest modes will be high enough in frequency to impose a significant tonal signature based on room dimensions.

The use of broadband absorption on at least two adjacent surfaces in the playback room will tend to absorb and dissipate most of the resonant room modes that are excited by the loudspeakers. Only the lateral axial mode will remain essentially unaffected, and it will tend to be comparatively unimportant to the loudspeaker/room/listener interaction. By designing the room so that side walls are not parallel, the lateral axial mode can be diffused as well.

Lateral symmetry

It is desirable to make the playback room symmetrical along the front-to-back median plane (and, of course, to place the loudspeakers so that they use the same median plane for their Left/Right orientation). Such symmetry enhances the effectiveness of lateral early reflections in supporting phantom images, phantom reverberance and envelopment.