Loudspeakers: Objective evaluations - Part 5: Professional monitor loudspeakers

Consumer audio is what it is, a mixture of marketing and engineering, and inconsistency is to be expected. However, the professional side also has its share of issues. There is more than a hint of complacency, and perhaps condescension, among professionals: Newell and Holland (2007) said, "Whilst professionals tend to work with standardized, known, and objectively designed equipment, domestic equipment tends to be individualistic, and marked by diversity more than commonality." In the scientifically based, engineering-driven mainstream of consumer audio loudspeakers, that is a perspective lacking evidentiary support. The loudspeaker measurements in this chapter identify excellence and mediocrity in both professional and consumer camps.

The large survey reported in Figure 2.4 can only be interpreted with alarm - essentially identical high-quality loudspeakers that in the hands of professionals exhibit an enormous range of performances. The very popular monitor shown in Figure 2.6b was not only accepted in spite of its substantial imperfections, but a new fashion in control room design assisted in its rehabilitation by creating a "dead end" and absorbing most of the (terrible) off-axis sound.

In professional worlds, standards are often relied on to maintain levels of excellence. However, standards often describe the way things are, as decided by a committee of practitioners, not the way they should be. For certain kinds of standards, those dealing with the normalization of physical dimensions, labeling, and so forth, things are straightforward. However, in the development of a standard, if a controversial opinion enters the discussion, progress ceases. Only agreement moves the document forward, and so the content gravitates toward a comfortable middle-of-the-road position, often representing long-standing traditions in the industry.

It is rare for a standard to espouse the state-of-the-art because, inevitably, it is not widespread within the industry and, equally inevitably, there will be detractors. In fact, it is common for such documents to be issued as recommendations not as standards because of lack of universal support. The result is that many of these standards and recommendations may prevent truly bad things from happening, but excellence is not assured, nor is consistency from location to location.

I thought about using such a standard as an example and examining it in detail using perspectives taught earlier in this book. When this was done with one of the more popular international recommendations for choosing and using "reference monitor loudspeakers," the result was greatly disappointing; there was very little left intact. All of these documents are revised periodically, so rather than criticize a particular example, I have chosen to describe what I perceive as serious shortcomings that are commonly found in such documents and hope that future revisions may incorporate improvements in these areas.

• First, it is common for all acoustical measurements to be detuned to 1/3-octave resolution, so all medium- and high-Q resonances and sharp discontinuities will be substantially attenuated, if not rendered invisible. In existing documents this is done for anechoic measurements on loudspeakers (inexcusable), as well as measurements of these loudspeakers as installed on site, measured at the listening location (understandable).

• Anechoic measurements frequently focus on a narrow frontal perspective, rarely exploring beyond 30° off axis. As has been seen in this chapter, this fails to provide any insight into directivity, early reflections, sound power - all of the information that can help us to understand how the loudspeaker may sound in a room. Question: where might ordinary audio professionals find such data, even in the inadequate form specified? Answer: it should be (but often is not) provided by the loudspeaker manufacturers, as discussed in Section 18.2.7, as evidence of their qualification to play in this league. Ideally, these data should be available in a standardized format, with standardized frequency resolution, and so on.

• All of the preceding measurements are commonly allowed to vary within tolerances as generous as ±3 dB, or even more, meaning that there are no assurances of sound quality whatsoever.

• Measurements made at the reference listening location are required to fall within a tolerance that is never less than ±3 dB and that increases in the downward direction as the upper- and lower-frequency extremes are approached. The generosity of the tolerances is much appreciated by studio owners, who should be able to qualify with minimal effort (except see Figure 2.4 for examples of installations that appear not to have even made an attempt).

• Mounting variations - adjacent-boundary effects - are rarely discussed. Interestingly, a loudspeaker that passes the anechoic test for axial flatness may fail this test if it is installed in a soffit or 2p wall mounting (see Figure 12.8) or even close to a wall. It will have different problems if it is placed on the meter bridge.

• Equalization is almost always required, and the topic is rarely discussed. It needs to be at the core of the document, because it changes the requirements for anechoic performance and performance at the listening position. Anechoic data should be used to eliminate loudspeakers with unacceptable resonances and nonuniform directivity. Broad inconsistencies in frequency response can be corrected with equalization. Performance below about 300 Hz can only be addressed after installation in the room.

• Measurements made at the reference listening location - room curves - are subject to all of the misgivings expressed in Section 18.2.6 in connection with the "X" curve. They include aspects of room acoustics that are normally not adequately specified.

• In control rooms, attenuation of early reflections, particularly those from the side walls, is usually a requirement. The need for this has been discussed in detail earlier, and it is an option, not a requirement. All too commonly, the requirement to attenuate these reflections applies to frequencies above 1 kHz, meaning that 1-in.-thick (25 mm) fiberglass board or slab foam will suffice, and all that is accomplished is an attenuation of first reflections from tweeters. All sound from 1 kHz down is fully reflected, and, as shown in Figure 6.18, the audible threshold of the reflection has been negligibly changed. The spectral balance of the sound has been altered, and possibly a good loudspeaker has been made to sound less good. A review of Chapters 5 through 9 offer persuasive arguments that there is more to this than is represented in typical documents. Figure 21.9 shows that traditional random-incidence acoustical measurements of absorption coefficient do not describe what happens in a "first-reflection" situation.

• Most recommendations set limits on room proportions implying that compliance with the dimensional requirements leads to audible advantages. As shown in Section 13.2.1, without specific knowledge of listener and loudspeaker locations within the room, dimensional proportions are of little value, and having five full-range loudspeakers adds complication that none of the normal predictive calculations account for, making them totally useless.

• The option of using subwoofer/satellite systems is almost never mentioned, even though it is the most commonly used and most likely the best possible configuration for multichannel playback. It is also the most likely configuration to achieve a transfer of a high-quality listening experience from the control room to the home. At the very least it needs to be included as an option so recording personnel can hear their product as it is likely to be heard in home listening rooms and theaters.

• When subwoofer/satellite systems are used, the acoustical crossover from the subwoofer(s) to each of the satellite loudspeakers must be individually measured and the low-pass and high-pass filter characteristics adjusted to achieve smooth summing in the crossover region. Doing that requires high-resolution transfer-function measurements (amplitude and phase) and a flexible electronic crossover customization routine because each installation will be different. It is not sufficient to rely on fixed-slope electronic filters.

• Stereo programs need to be evaluated through the upmix algorithms commonly used in homes.

• Reverberation time is always specified, and its importance is exaggerated by requiring a precision and frequency-dependent consistency that are excessive for this application. Often there are no requirements for how it should be measured (see Chapter 4).

This list could be extended, but if even these points could be considered for incorporation into an agreed-upon recommendation, the industry will have made a substantial step forward.

REFERENCES
Newell, P.R., and Holland, K.R. (2007). “Loudspeakers for music recording and reproduction,” Focal Press, Oxford, U.K.

Printed with permission from Focal Press, a division of Elsevier. Copyright 2008. "Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms" edited by Floyd Toole. For more information about this title and other similar books, please visit www.elsevierdirect.com.

Related links:
Loudspeakers: Objective evaluations - Part 1: Sound source radiation patterns | Part 2: Measuring the essential properties of loudspeakers | Part 3: Comparing the subjective and objective domains | Part 4: Measurements of real-world consumer loudspeakers
Acoustics and Psychoacoustics Applied - Part 1: Listening room design
Acoustics and Psychoacoustics: Introduction to Sound, Part 1: Pressure waves and sound transmission | Part 2: Sound intensity, power and pressure level | Part 3: Adding sounds together | Part 4: The inverse square law | Part 5: Sound Interactions | Part 6: Sound Interactions (cont.) | Part 7: Time and frequency domains
Using the Decibel - Part 1: Introduction and underlying concepts | Using the Decibel - Part 2: Expressing Power as an Audio Level