In audio production, no single item in the sound chain is more important than the reference loudspeakers, and no process in evaluating the product is more important than monitoring. The sound chain begins with the microphone, whose signal is sent to a console, mixer, or computer for routing, processing, and recording and heard through a loudspeaker. Although the loudspeaker is last in this signal flow, no matter how good the other components in the audio system, the quality of every sound evaluated is based on what you hear from a loudspeaker interacting with the room acoustics. Knowing a loudspeaker’s sonic characteristics in relation to the acoustic environment is fundamental to the accurate monitoring and appraisal of audio material.
Loudspeakers are sometimes compared to musical instruments in that they produce sound. But it is more accurate to say that they reproduce sound. And, unlike any musical instrument, a loudspeaker has the capability of generating a far wider frequency response and a greater dynamic range.
Loudspeakers are not like purely electronic components such as consoles, which can be objectively tested and rationally evaluated. No two loudspeakers sound quite the same. Comparing the same make and model of loudspeakers in one room tells you only what they sound like in that acoustic environment; in another room, they may sound altogether different. Furthermore, a loudspeaker that satisfies your taste might be unappealing to someone else.
A loudspeaker’s specifications can be used only as a reference. No matter how good a speaker looks on paper, keep in mind that the measurements are based on tests made in an anechoic chamber—a room with no reflections of any kind.
Another problem in loudspeaker evaluation is that in comparison tests, which are done by switching back and forth between two monitors or switching among more than two monitors, auditory memory is quite short—only a few seconds at best. By the time one listens to a second or third set of monitors, recalling what the comparisons in the first set were or what the first monitor sounded like is unreliable. This is why employing objective testing is an essential part of loudspeaker evaluation.
Even as methods for evaluating loudspeakers improve, it is still difficult to suggest guidelines that will satisfy the general population of audio professionals, though they will agree that monitors used for professional purposes should meet certain performance requirements.
Evaluating frequency response in a loudspeaker involves two considerations: how wide it is and how flat, or linear, it is. Frequency response ideally should be as wide as possible, from at least 40 to 20,000 Hz. But the relationship between the sound produced in the studio and the sound reproduced through the listener’s receiver/loudspeaker becomes a factor when selecting a monitor loudspeaker.
For example, TV audio can carry the entire range of audible frequencies. You may have noticed during a televised music program, however, that certain high- and low-pitched instruments are played, but you hear them only faintly or not at all. Overhead cymbals are one example. Generally, their frequency range is between 300 and 12,000 Hz (including overtones), but they usually begin to gain good definition between 4,500 and 8,000 Hz, which is well within the frequency of television transmission. The highest response of many home TV receivers is about 6,000 Hz, however, which is below a good part of the cymbals’ range.
Suppose you wish to boost the cymbal frequencies that the home TV receiver can barely reproduce. Unless you listen to the sound over a monitor comparable in output level and response to the average TV speaker, you cannot get a sense of what effect the boost is having. But that monitor will not give you a sense of what viewers with upgraded TV sound systems, such as stereo and surround sound, are hearing (see Figure 5-1). (The audio in high-definition television—HDTV—has great potential to level the playing field between the sonic fidelity of the transmission and that of the reception, but it depends on the audio quality of the commercial TV tuner and the willingness of the consumer to purchase an HDTV set.)
Table 5-1. Table of frequency responses.
Device | Typical frequency response |
|---|---|
AM radio | 80 Hz —5 kHz |
Portable AM/FM radio | 80-100 Hz — 10-12 kHz |
Table-model standard stereo TV | 100 Hz — 5-8 kHz |
7-inch portable high-definition TV | 300 Hz — 3.5 kHz |
Good-quality standard stereo TV | 50 Hz — 15 kHz |
Large-screen high-definition TV | 30 Hz — 20 kHz |
Mediocre-quality loudspeaker | 150 Hz — 10 kHz |
Good-quality small loudspeaker | 60 Hz — 20 kHz |
Good-quality large loudspeaker | 35 Hz — 20 kHz |
Good-quality CD/DVD sound system | 20 Hz — 20 kHz |
Call phone | 300 Hz — 3 kHz |
Due to the differences between the potential sound response of a medium and its actual response after processing, transmission, and reception, most professional studios use at least two types of studio monitors: one to provide both wide response and sufficient power to reproduce a broad range of sound levels, and the other with response and power that reflect what the average listener hears. Many studios use three sets of monitors to check sound: (1) mediocre-quality loudspeakers with mostly midrange response and limited power output such as those in portable and car radios, portable and desktop TVs, computer-grade speakers, and small playback devices; (2) average-quality loudspeakers with added high and low response such as moderately priced component systems; and (3) high-quality loudspeakers with a very wide response and high output capability.
The second consideration in evaluating frequency response in a loudspeaker is how linear it is. Linearity means that frequencies fed to a loudspeaker at a particular loudness are reproduced at the same loudness. If they are not, it is very difficult to predetermine what listeners will hear. If the level of a 100 Hz sound is 80 dB going in and 55 dB coming out, some—if not most—of the information may be lost. If the level of an 8,000 Hz sound is 80 dB going in and 100 dB coming out, the information loss may be overbearing.
Loudspeaker specifications include a value that indicates how much a monitor deviates from a flat frequency response, by either increasing or decreasing the level. This variance should be no greater than ±3dB.
To generate adequately loud sound levels without causing distortion, the loudspeaker amplifier must provide sufficient power. At least 30 W for tweeters (high-frequency loudspeakers) and 100 W for woofers (low-frequency loudspeakers) is generally necessary. Regardless of how good the rest of a loudspeaker’s components, if the amp does not have enough power, efficiency suffers considerably.
There is a commonly held and seemingly plausible notion that increasing amplifier power results in a proportional increase in loudness; for example, that a 100 W amplifier can play twice as loud as a 50 W amplifier. In fact, if a 100 W amp and a 50 W amp are playing at top volume, the 100 W amp will sound only slightly louder. What the added wattage gives is clearer and less distorted reproduction in loud sonic peaks.
Discussion of amplifier power naturally leads to consideration of distortion—appearance of a signal in the reproduced sound that was not in the original sound (see Figure 5-2). Any component in the sound chain can generate distortion. Because distortion is heard at the reproduction (loudspeaker) phase of the sound chain—regardless of where in the system it was generated—and because loudspeakers are the most distortion-prone component in most audio systems, it is appropriate to discuss briefly the various forms of distortion: intermodulation, harmonic, transient, and loudness.
Figure 5-2. Generalized graphic example of distortion. Ideally, a reproduced sound would change in direct proportion to the input. Realistically, the relationship between input and output is rarely linear, resulting in distortion, however slight or significant it may be.
The loudspeaker is perhaps most vulnerable to intermodulation distortion (IM), which results when two or more frequencies occur at the same time and interact to create combination tones and dissonances unrelated to the original sounds. Audio systems can be most vulnerable to intermodulation distortion when frequencies are far apart, as when a piccolo and a baritone saxophone are playing at the same time. Intermodulation distortion usually occurs in the high frequencies because they are weaker and more delicate than the low frequencies.
Wideness and flatness of frequency response are affected when IM is present. In addition to its obvious effect on perception, even subtle distortion can cause listening fatigue.
Unfortunately, not all specification sheets include percentage of IM, and those that do often list it only for selected frequencies. Nevertheless, knowing the percentage of IM for all frequencies is important to loudspeaker selection. A rating of 0.5 percent IM or less is considered good for a loudspeaker.
Harmonic distortion occurs when the audio system introduces harmonics into a recording that were not present originally. Harmonic and IM distortion usually happen when the input and the output of a sound system are nonlinear, that is, when they do not change in direct proportion to each other. A loudspeaker’s inability to handle amplitude is a common cause of harmonic distortion. This added harmonic content is expressed as a percentage of the total signal or as a component’s total harmonic distortion (THD).
Transient distortion relates to the inability of an audio component to respond quickly to a rapidly changing signal, such as that produced by percussive sounds. Sometimes transient distortion produces a ringing sound.
Overly loud signals give a false impression of program quality and balance; nevertheless, a loudspeaker should be capable of reproducing loud sound levels without distorting, blowing fuses, or damaging its components. An output capability of 120 dB-SPL can handle almost all studio work. Even if studio work does not often call for very loud levels, the monitor should be capable of reproducing them because it is sometimes necessary to listen at a loud level to hear subtlety and quiet detail. For soft passages, at least 40 dB-SPL is necessary for most professional recording/mixing situations.
It is worth noting that although monitors in professional facilities have a dynamic range of up to 80 dB (40 dB-SPL to 120 dB-SPL), the dynamic range in consumer loudspeakers is considerably less. Even the better ones may be only 55 dB (50 dB-SPL to 105 dB-SPL).
Sensitivity is the on-axis sound-pressure level a loudspeaker produces at a given distance when driven at a certain power (about 3.3 feet with 1 W of power). A monitor’s sensitivity rating gives you an idea of the system’s overall efficiency. Typical ratings range from 84 dB to more than 100 dB.
In real terms, however, a sensitivity rating of, say, 90 dB indicates that the loudspeaker could provide 100 dB from a 10 W input and 110 dB from a 100 W input, depending on the type of driver. It is the combination of sensitivity rating and power rating that indicates whether a monitor loudspeaker will be loud enough to suit your production needs. Generally, a sensitivity rating of 93 dB or louder is required for professional applications.
Polar response indicates how a loudspeaker focuses sound at the monitoring position(s). Because it is important to hear only the sound coming from the studio or the recording, without interacting reflections from the control room walls (vertical surfaces) and ceiling or floor (horizontal surfaces), dispersion must be controlled at the monitoring locations so it is a relatively reflection-free zone (see Figure 5-3).
This is easier said than done, particularly with bass waves, which are difficult to direct because of their long wavelengths. With bass waves, the room’s size, shape, and furnishings have more to do with a speaker’s sound than the speaker’s inherent design. Therefore, bass traps and other low-frequency absorbers are included in control room design to handle those bass waves not focused at the listening position (see Figures 5-4 and 5-5; see also Figures 4-13 and 4-16).
Figure 5-4. Improving control room acoustics. Using strategically placed absorbers, diffusers, and bass traps can help control unwanted room reflections.
Figure 5-5. Room and loudspeaker influences on the response of selected musical instruments. At frequencies lower than 300 Hz, the room is the dominant factor. Above 300 Hz the inherent loudspeaker characteristics dominate response.
Frequencies from the tweeter(s), on the other hand, are shorter, more directional, and easier to focus. The problem with high frequencies is that as the wavelength shortens, the pattern can narrow and may be off-axis to the listening position. Therefore, the coverage angle—defined as the off-axis angle or point at which loudspeaker level is down 6 dB compared with the on-axis output level—may not be wide enough to include the entire listening area.
To help in selecting a loudspeaker with adequate polar response, specifications usually list a monitor’s horizontal and vertical coverage angles. These angles should be high enough and wide enough to cover the listening position and still allow the operator or listener some lateral movement without seriously affecting sonic balance (see Figure 5-6).
Figure 5-6. Coverage angles of monitor loudspeakers. (a) Desirable coverage angle. Alternative coverage angles include (b) loudspeakers with narrow radiations, which produce a sharp aural image and less envelopment, and (c) loudspeakers that radiate more broadly, which increase envelopment but reduce imaging.
Figures 5-4, 5-5, and 5-6 display examples of reflection-free zones and coverage angles using conventional two-loudspeaker setups for stereo monitoring. Surround sound presents altogether different acoustic challenges because monitors are located to the front and the rear and/or the sides of the listening position. Theoretically, therefore, for surround monitoring there should be an even distribution of absorbers and diffusers so that the rear-side loudspeakers function in an acoustic environment similar to the frontal speaker array. The problem is that this approach renders mixing rooms designed for stereo unsuitable for surround sound because stereo and surround require different acoustic environments (see “Monitoring Surround Sound” later in this chapter).
Even if coverage angles are optimal, unless all reproduced sounds reach the listening position(s) at relatively the same time they were produced, aural perception will be impaired. When you consider the differences in size and power requirements among drivers and the wavelengths they emit, you can see that this is easier said than done.
In two-, three-, and four-way system loudspeakers, the physical separation of each speaker in the system causes the sounds to reach the listener’s ears at different times. Arrival times that differ by more than 1 ms are not acceptable in professional applications.
Sometimes, although dispersal and arrival time are adequate, sound reaching a listener may not be as loud as it should be, or the elements within it may be poorly placed. For example, a rock band may be generating loud levels in the studio, but the sounds in the control room are, in relative terms, not so loud; or an actor is supposed to be situated slightly left in the aural frame but is heard from the right loudspeaker.
These problems may be the result of the loudspeakers’ polarity being out of phase: One loudspeaker is pushing sound outward (compression), and the other is pulling sound inward (rarefaction) (see Figure 5-7). Polarity problems can occur between woofer and tweeter in the same loudspeaker enclosure or between two separate loudspeakers. In the latter case, this usually happens because the connections from loudspeaker to amplifier are improperly wired; that is, the two leads from one loudspeaker may be connected to the amplifier positive-to-negative and negative-to-positive, whereas the other loudspeaker may be connected positive-to-positive and negative-to-negative. If you think sound is out of phase, check the sound-level meters. If they show a similar level and the audio still sounds skewed, the speakers are out of phase; if they show different levels, the phase problem is probably elsewhere.
Another way to test for polarity problems is to use a 9 V battery. Attach the battery’s positive lead to the loudspeaker’s positive lead and the battery’s negative lead to the loudspeaker’s negative lead. Touching the battery’s positive lead should push the speaker outward. If the speaker moves inward, it is out of polarity. All speakers in a cabinet should move in the same direction.
Where you place monitor loudspeakers also affects sound quality, dispersal, and arrival time. Loudspeakers are often designed for a particular room location and are generally positioned in one of four places: well toward the middle of a room, against or flush with a wall, at the intersection of two walls, or in a corner at the ceiling or on the floor. Each position affects the sound’s loudness and dispersion differently.
A loudspeaker hanging in the middle of a room radiates sound into what is called a free-sphere (or free-field or full space) where, theoretically, the sound level at any point within a given distance is the same because the loudspeaker can radiate sound with no limiting obstacles or surface (see Figure 5-8). If a loudspeaker is placed against a wall, the wall concentrates the radiations into a half sphere, (or half space) thereby theoretically increasing the sound level by 3 dB. With loudspeakers mounted at the intersection of two walls, the dispersion is concentrated still more into a one-quarter sphere, thus increasing the sound level another 3 dB. Loudspeakers placed in corners at the ceiling or on the floor radiate in a one-eighth sphere, generating the most concentrated sound levels in a four-walled room. A significant part of each increase in the overall sound level is due to the loudness increase in the bass (see Figure 5-9). This is particularly the case with frequencies at 300 Hz and below. It is difficult to keep them free of excessive peaks and boominess, thereby inhibiting a coherent, tight bass sound.
Figure 5-8. Four typical loudspeaker locations in a room and the effects of placement on overall loudness levels.
For informal listening, one of these monitor positions is not necessarily better than another; placement may depend on a room’s layout, furniture position, personal taste, and so on. In professional situations, it is preferable to flush-mount loudspeakers in a wall or soffit (see Figure 5-13). The most important thing in monitor placement is to avoid any appreciable space between the loudspeaker and the wall and any protrusion of the loudspeaker’s cabinet edges. Otherwise, the wall or cabinet edges, or both, will act as secondary radiators, degrading frequency response. By flush-mounting loudspeakers, low-frequency response becomes more efficient, back-wall reflections and cancellation are eliminated, and cabinet edge diffraction is reduced.
Stereo creates the illusion of two-dimensional sound by imaging depth—front-to-back—and width—side-to-side—in aural space. It requires two discrete loudspeakers each of which, in reality, is producing monaural sound but creating a phantom image between them (see Figure 5-10). In placing these loudspeakers, it is critical that they be positioned to reproduce an accurate and balanced stereo image. The monitoring system should be set up symmetrically within the room. The distance between the speakers should be the same as the distance from each speaker to your ears, forming an equilateral triangle with your head. Also, the center of the equilateral triangle should be equidistant from the room’s side walls (see Figure 5-11).
Figure 5-10. Why we perceive a phantom image. From Dave Moulton, “Median Plane, Sweet Spot, and Phantom Everything,” TV Technology, November 2007, p. 34.
Figure 5-11. To help create an optimal travel path for sound from the loudspeakers to the listening position, carefully measure the loudspeaker separation and the distance between the loudspeakers and the listening position. (a) The distance between the acoustic center (between the loudspeakers) and the listening position is equal. In this arrangement head movement is restricted somewhat and the stereo image will be emphasized or spread out, heightening the sense of where elements in the recording are located. Shortening D2 and D3 will produce a large shift in the stereo image. (b) The distance between the acoustic center and the listening position is about twice as long. With this configuration the movement of the head is less restricted and the stereo image is reduced in width, although it is more homogeneous. The more D2 and D3 are lengthened, the more monaural the stereo image becomes (except for hard-left and hard-right panning).
The locations of the front-to-back and side-to-side sound sources are where they should be. If the original material has the vocal in the center (in relation to the two loudspeakers)—the first violins on the left, the bass drum and the bass at the rear-center, the snare drum slightly left or right, and so on—these should be in the same spatial positions when the material is played through the monitor system. If the listener is off the median plane, it skews the stereo imaging (see Figure 5-12). The designated stereo listening position is known as the sweet spot.
Figure 5-12. Phantom imaging on and off the median plane. From Dave Moulton, “Median Plane, Sweet Spot, and Phantom Everything,” TV Technology, November 2007, p. 34.
Most professional audio-mixing rooms use at least two sets (or groups) of frontal loudspeakers. One set is for far-field monitoring, consisting of large loudspeaker systems that can deliver very wide frequency response at moderate to quite loud levels with relative accuracy. Due to their size and loudness capabilities, these loudspeakers are built into the mixing-room wall above, and at a distance of several feet from, the listening position (see Figure 5-13). Far-field loudspeakers are designed to provide the highest-quality sound reproduction.
Even in the best monitor-control room acoustic environments, the distance between wall-mounted loudspeakers and the listening position is often wide enough to generate sonic discontinuities from unwanted control room reflections. To reduce these unwanted reflections, another set of monitors is placed on or near the console’s meter bridge (see Figure 5-13).
Near-field monitoring reduces the audibility of control room acoustics by placing loudspeakers close to the listening position. Moreover, near-field monitoring improves source localization because most of the sound reaching the listening position is direct; the early reflections that hinder good source localization are reduced to the point where they are of little consequence. At least that is the theory; in practice, problems with near-field monitoring remain (see Figure 5-14).
Figure 5-14. Near-field monitoring. (a) If a meter bridge is too low, early reflections will bounce off the console, degrading the overall monitor sound reaching the operator’s ears. (b) One way to minimize this problem is to place the near-field monitors a few inches in back of the console.
Among the requirements for near-field monitors are: loudspeakers small enough to put on or near the console’s meter bridge without the sound blowing you away; a uniform frequency response from about 70 to 16,000 Hz, especially smooth response through the midrange; a sensitivity range from 87 to 92 dB; sufficient amplifier power; and good vertical dispersion for more stable stereo imaging.
It is worth remembering that many near-field monitors have a built-in high pass filter to protect the bass unit from overload. The filter is generally set around 50 Hz. Not knowing that the loudspeaker is removing this part of the spectrum can lead to low-frequency balancing problems.
Surround sound differs from stereo by expanding the dimension of depth, thereby placing the listener more in the center of the aural image than in front of it. Accomplishing this requires additional audio channels routed to additional loudspeakers.
The most common surround-sound format (for the time being) uses six discrete audio channels: five full-range and one limited to low frequencies (typically below 125 Hz), called the subwoofer. Hence the format is known as 5.1. The loudspeakers that correspond to these channels are placed front-left and front-right, like a stereo pair; a center-channel speaker is placed between the stereo pair; and another stereo pair—the surround speakers—are positioned to the left and right sides, or left-rear and right-rear, of the listener. (Unlike stereo, there is no phantom image in surround’s frontal loudspeaker arrangement. The center-channel speaker creates a discrete image.) The sub-woofer can be placed almost anywhere in the room because low frequencies are relatively omnidirectional, but it is usually situated in the front, between the center and the left or right speaker (see Figure 5-15). Sometimes it is positioned in a corner to reinforce low frequencies.
Figure 5-15. International Telecommunications Union guideline for arranging loudspeakers in a surround-sound setup. In a 5.1 system, the front-left and front-right speakers form a 60-degree angle with the listener at the apex; the center-channel speaker is directly in front of the listener. The surround speakers are usually placed at an angle between 100 and 120 degrees from the front-center line.
A problem in positioning the center-channel speaker is the presence of a video monitor if it is situated on the same plane as the loudspeakers. Mounting the center speaker above or below the monitor is not the best location. If it is unavoidable, however, keep the tweeters close to the same plane as the left and right speakers, which may require turning the center speaker upside down. If the image is projected, it becomes possible to mount the center speaker behind a microperforation screen, as they do in movie theaters.
A primary consideration with surround-sound setups is making sure the room is large enough, not only to accommodate the additional equipment and loudspeaker placement specifications, but also to handle the relatively high loudness that is sometimes necessary when monitoring surround sound.
Evaluating the Monitor Loudspeaker[1]
The final test of any monitor loudspeaker is how it sounds. Although the basis for much of the evaluation is subjective, there are guidelines for determining loudspeaker performance.
Begin with rested ears. Fatigue alters aural perception.
Sit at the designated stereo listening position—the optimal distance away from and between the loudspeakers. This sweet spot should allow some front-to-back and side-to-side movement without altering perception of loudness, frequency response, and spatial perspective. It is important to know the boundaries of the listening position(s) so that all sound can be monitored on-axis within this area (review Figures 5-6 and 5-11).
If repositioning near-field monitors is necessary, changes of even a few inches closer or farther or from side to side can make a considerable difference in how the sound is perceived.
There is no one best listening position with surround sound. Moving around provides different perspectives without significantly changing clarity, dimension, and spatial coherence (see Figure 5-15).
When evaluating monitors in general and near-field monitors in particular, the tendency is to be drawn to the bass response. The focus should be on overall clarity.
Use material with which you are intimately familiar for the evaluation, preferably on a high-quality digital disc. Otherwise, if some aspect of the sound is unsatisfactory, you won’t know whether the problem is with the original material or the loudspeaker.
For example, with speech the male voice is a good test to reveal if a loudspeaker is boomy; the female voice helps determine if there is too much high-end reflection from nearby surfaces. With music, consider how the following sonic details would change tonally from speaker to speaker: an overall sound with considerable presence and clarity; the depth and the dynamic range of symphonic music; the separation in the spatial imaging of instruments; a singer whose sound sits in the lower midrange; the sounds of fingers on guitar strings and frets or fingernails plucking the strings; the kiss of the hammers on piano strings; the wisp of brushes on a snare drum; the deep bass response of an organ; the punch of the kick drum; and the upper reaches of a flute’s harmonics. Such tests, and there are many others, facilitate the judging of such things as a loudspeaker’s bass, midrange, and treble response; dynamic response; imaging; clarity; and presence.
Keep in mind that as bass response becomes more prominent, midrange clarity suffers due to masking.
Listen at a comfortable loudness level; 85 dB-SPL is often recommended.
Listen for sounds you may not have heard before—such as hiss, hum, or buzz. Good monitors may reveal what inferior monitors cover up; or it could indicate that the monitors you are listening to have problems. In either case, you have learned something.
In listening for spatial balance, make sure the various sounds are positioned in the same places relative to the original material. As an additional test for stereo, put the monitor system into mono and check to make sure the sound appears to be coming from between the two loudspeakers. Move toward the left and right boundaries of the sweet spot. If the sound moves with you before you get near the boundaries, reposition the monitors, then recheck their dispersion in both stereo and mono until you can move within the boundaries of the listening position without the sound following you or skewing the placement of the sound sources. If monitor repositioning is necessary to correct problems with side-to-side sound placement, make sure it does not adversely affect front-to-back sound dispersion to the listening position.
In evaluating monitor loudspeakers for surround sound, consider the following: Is the imaging of the sound sources cohesive or disjointed? Does the sound emanate from the appropriate loudspeaker location, or does it seem detached from it? In moving around the listening environment, how much of the positional information, or sonic illusion, remains intact? How robust does that illusion remain in another listening environment? If there are motion changes—sound moving left-to-right, right side to right front, and so on—are they smooth? Does any element call unwanted attention to itself?
In evaluating treble response, listen to the cymbal, triangle, flute, piccolo, and other high-frequency instruments. Are they too bright, crisp, shrill, harsh, or dull? Do you hear their upper harmonics?
In testing bass response, include such instruments as the tuba and the bass as well as the low end of the organ, piano, bassoon, and cello. Sound should not be thin, boomy, muddy, or grainy. For example, notes of the bass guitar should be uniform in loudness; the piano should not sound metallic.
Assess the transient response, which is also important in a loudspeaker. Drums, bells, and triangle provide excellent tests, assuming they are properly recorded. Good transient response reproduces a crisp attack with no distortion, breakup, or smearing.
If you are evaluating a number of different loudspeakers, compare only a few at a time and take notes. Trying to remember, for example, how different speakers color the presence range, their differences in low-end warmth, or their degree of darkness or brightness, is unreliable. Our ability to retain precise auditory information is limited, particularly over time and when comparison testing.
Other important elements to evaluate, such as intermodulation, harmonic, transient, and loudness distortion, were discussed earlier in this chapter.
When doing a session in a facility you have not worked in before, it is essential to become thoroughly familiar with the interaction between its monitor sound and its room sound. A relatively quick and reliable way to get an objective idea of that interaction is to do a real-time analysis. It is also crucial to put the real-time analysis into perspective with reference recordings and your ears.
Reference recordings on CD, CD-ROM, or DVD can be commercial recordings with which you are entirely familiar or discs specially designed to help assess a listening environment or both. Knowing how a recording sounds in a control room whose monitors and acoustics you are intimately familiar with is a good test in determining the sonic characteristics of, and among, the monitors and the acoustics in a new environment. For example, if you know that a recording has a clear high-end response in your own control room but the high end sounds thinner in another control room, it could indicate, among other things, that the other studio’s monitors have inadequate treble response; the presence of harmonic distortion; a phase problem; the room’s mediocre sound diffusion; or any combination of these factors.
Digital discs specially produced for referencing are also available. They are variously designed to test monitor and/or room response to a particular instrument, such as a drum set; individual or groups of instruments, such as the voice and clarinet, or strings, brass, and woodwinds; various sizes and types of ensembles, such as orchestras, jazz bands, and rock groups; room acoustics; spatial positioning; and spectral balances.
Headphones (also referred to as “cans”) are an overlooked but important part of monitoring, especially in field production. The following considerations are basic when using headphones for professional purposes: Although frequency response in headphones is not flat, listen for as wide and uncolored response as possible; you must be thoroughly familiar with their sonic characteristics; they should be circumaural (around-the-ear), as airtight as possible against the head for acoustical isolation, and comfortable; the fit should stay snug even when you are moving; and, although it may seem obvious, stereo headphones should be used for stereo monitoring, and headphones capable of multichannel reproduction should be used for monitoring surround sound (see Figure 5-16).
Figure 5-16. Headphone monitoring system. This particular system is called Headzone. It includes stereo headphones and a digital processor that accepts 5.1 signal sources. The processor uses modeling software to reproduce a surround-sound image. It also provides the ability to turn your head while the virtual sound sources stay in the correct spatial positions as they do in control room monitoring through separate loudspeakers. The processor takes various types of connectors, such as a three-pin socket for static ultrasonic receivers, six phono (RCA) sockets for unbalanced 5.1 analog inputs, and a standard sixpin FireWire socket for linking to a computer. The processor can be remotely controlled.
Headphones are usually indispensable in field recording because monitor loudspeakers are often unavailable or impractical. In studio control rooms, however, there are pros and cons to using headphones.
The advantages are (1) sound quality will be consistent in different studios; (2) it is easier to hear subtle changes in the recording and the mix; (3) there is no aural smearing due to room reflections; and (4) because so many people listen to portable players through headphones or earbuds, you have a reasonably reliable reference to their sonic requirements.
The disadvantages include (1) the sound quality will not be the same as with the monitor loudspeakers; (2) the aural image forms a straight line between your ears and is unnaturally wide; (3) in panning signals for stereo, the distance between monitor loudspeakers is greater than the distance between your ears—left and right sounds are directly beside you; (4) no interaction exists between the program material and the acoustics, so the tendency may be to mix in more artificial reverberation than necessary; and (5) in location recording if open-air, or supra-aural, headphones are used, bass response will not be as efficient as with circumaural headphones, and outside noise might interfere with listening.
As we discussed in Chapter 2, high sound levels from everyday sources such as loudspeakers, loud venues, hair dryers, and lawn mowers are detrimental to hearing. Potentially, this is even more the case with headphones because they are mounted directly over the ear or, with earbuds, placed inside the ear. The problem is exacerbated outdoors by the tendency to boost the level so that the program material can be heard above the noise floor. Listening over an extended period of time can also result in cranking up the level to compensate for listening fatigue.
It has been found that listening to headphones at a loud level after only about an hour and a quarter can cause hearing loss. One recommendation is called the 60 percent/60-minute rule: Limit listening through headphones to no more than one hour per day at levels below 60 percent of maximum volume. You can increase daily listening time by turning down the level even farther.
One development in headphone technology to help reduce the potential of hearing loss is the active noise-canceling headphone. Designs differ, but generally the noise-canceling headphone detects ambient noise before it reaches the ears and nullifies it by synthesizing the sound waves. Most noise-canceling headphones act on the more powerful low- to midrange frequencies; the less powerful higher frequencies are not affected. Models may be equipped with a switch to turn the noise-canceling function on or off and may be battery-operated.
Passive noise-canceling can be employed with conventional circumaural headphones. The better the quality of headphones and the tighter the fit, the more effective the reduction of ambient noise.
The in-ear monitor (IEM) was developed for use by musicians in live concerts to replace stage monitors. They allow the performers to hear a mix of the on-stage microphones or instruments, or both, and provide a high level of noise reduction for ear protection. Due to the convenience in adjusting the sound levels and mix of audio sources to personal taste, in-ear monitoring has become increasingly popular with performers in broadcasting and recording. The problem with IEMs is that a performer who prefers loud sound levels or has hearing loss may increase the loudness of the IEM to a harmful level, thereby negating its noise reduction advantage. A sound level analyzer is available that measures the sound pressure levels reaching the ear canal. Users can determine not only their actual monitoring level, but how long they can safely listen at that volume.
The sound chain begins with the microphone, whose signal is sent to a console, mixer, or computer for routing, processing recording, and heard through a loudspeaker.
In evaluating a monitor loudspeaker, frequency response, linearity, amplifier power, distortion, dynamic range, sensitivity, polar response, arrival time, and polarity should also be considered.
Frequency response ideally should be as wide as possible, from at least 40 to 20,000 Hz, especially with digital sound.
Each medium that records or transmits sound, such as a CD or TV, and each loudspeaker that reproduces sound, such as a studio monitor or home receiver, has certain spectral and amplitude capabilities. For optimal results, audio should be produced with an idea of how the system through which it will be reproduced works.
Linearity means that frequencies fed to a loudspeaker at a particular loudness are reproduced at the same loudness.
Amplifier power must be sufficient to drive the loudspeaker system, or distortion, among other things, will result.
Distortion is the appearance of a signal in the reproduced sound that was not in the original sound.
Various forms of distortion include intermodulation, harmonic, transient, and loudness.
Intermodulation (IM) distortion results when two or more frequencies occur at the same time and interact to create combination tones and dissonances unrelated to the original sounds.
Harmonic distortion occurs when the audio system introduces harmonics into a recording that were not present originally.
Transient distortion relates to the inability of an audio component to respond quickly to a rapidly changing signal, such as that produced by percussive sounds.
Loudness distortion, or overload distortion, results when a signal is recorded or played back at an amplitude greater than the sound system can handle.
To meet most sonic demands, the main studio monitors should have an output-level capability of 120 dB-SPL and a dynamic range of up to 80 dB.
Sensitivity is the on-axis sound-pressure level a loudspeaker produces at a given distance when driven at a certain power. A monitor’s sensitivity rating provides a good overall indication of its efficiency.
Polar response indicates how a loudspeaker focuses sound at the monitoring position(s).
The coverage angle is the off-axis angle or point at which loudspeaker level is down 6 dB compared with the on-axis output level.
A sound’s arrival time at the monitoring position(s) should be no more than 1 ms; otherwise, aural perception is impaired.
Polarity problems can occur between woofer and tweeter in the same loudspeaker enclosure or between two separate loudspeakers.
Where a loudspeaker is positioned affects sound dispersion and loudness. A loud- speaker in the middle of a room generates the least-concentrated sound; a loudspeaker at the intersection of a ceiling or floor generates the most.
For professional purposes, it is preferable to flush-mount in a wall or soffit loud-speakers to make low frequency response more efficient and reduce or eliminate back-wall reflections, cancellation, and cabinet edge diffraction.
Stereo sound is two-dimensional; it has depth and breadth. In placing loudspeakers for monitoring stereo, it is critical that they be positioned symmetrically within a room to reproduce an accurate and balanced front-to-back and side-to-side sonic image.
Loudspeakers used for far-field monitoring are usually large and can deliver very wide frequency response at moderate to quite loud levels with relative accuracy. They are built into the mixing-room wall above, and at a distance of several feet from, the listening position.
Near-field monitoring enables the sound engineer to reduce the audibility of control room acoustics, particularly the early reflections, by placing loudspeakers close to the monitoring position.
Surround sound differs from stereo by expanding the depth dimension, thereby placing the listener more in the center of the aural image than in front of it. Therefore, using the 5.1 surround-sound format, monitors are positioned front-left, center, and front-right, and the surround loudspeakers are placed left and right behind, or to the rear sides of, the console operator. A subwoofer can be positioned in front of, between the center and the left or right speaker, in a front corner, or to the side of the listening position.
In adjusting and evaluating monitor sound, both objective and subjective measures are necessary.
When evaluating the sound of a monitor loudspeaker, it is helpful to, among other things, use material you are intimately familiar with and to test various loudspeaker responses with different types of speech and music.
Headphones are an important part of monitoring, particularly on location. The following considerations are basic when using headphones for professional purposes: The frequency response should be wide, flat, and uncolored; you must be thoroughly familiar with their sonic characteristics; they should be circumaural (around-the-ear), as airtight as possible against the head for acoustical isolation, and comfortable; the fit should stay snug even when you are moving; and, although it may seem obvious, stereo headphones should be used for stereo monitoring, and headphones capable of multichannel reproduction should be used for monitoring surround sound.
Headphones can be detrimental to hearing because they are mounted directly over the ears or, with earbuds, placed inside the ears. The problem is exacerbated outdoors by the tendency to boost the level so that the program material can be heard above the noise floor.
To help reduce the potential of hearing loss is the active noise-canceling headphone, which detects ambient noise before it reaches the ears and nullifies it by synthesizing the sound waves.
Passive noise-canceling can be employed using high-quality conventional circumaural headphones that fit snugly over the ears.
The in-ear monitor (IEM) allows a performer to hear a mix of on-stage microphones or instruments, or both, and provides a high level of noise reduction for ear protection. Since their development, IEMs are used for other monitoring purposes. A sound level analyzer is available for the specific purpose of measuring in-ear monitoring levels to guard against excessive loudness.
[1] Before evaluating a monitor loudspeaker system, either stereo or surround, it is critical to obtain an objective measure of the correlation between monitor sound and room sound using a calibration system. Calibration helps ensure that a system’s measurements meet a specified or desired standard. Calibration is usually done by an engineer or technician.