Listening: Strategies for Choosing Speakers

Home | Audio Magazine | Stereo Review magazine | Good Sound | Troubleshooting

CHOOSING loudspeakers is the most difficult part of buying or upgrading an audio system. The speakers affect the sound of a system more strongly than any other component, and their performance depends greatly on the listening room and their placement in it. And even for a given room there isn’t one “best” loudspeaker; your tastes in both music and sound will—and should—make your final buying decision a very personal one. Just because personal preference is important, though, doesn’t mean that you should simply buy the first speaker that sounds good in the store. Here are some general principles and specific stratagems, the product of years of listening in the lab, in stores, and at trade shows, to help you zero in on the speakers that best fit your taste, style, and budget.

TAKE YOUR TIME. This is the cardinal rule. If you just go to a store intending to come out with new speakers in half an hour, you make it easy for a sales man to steer you wherever he likes. Go to stores where the salespeople have time and patience, because if you follow my advice they’re going to need both to deal with you. In exchange, you shouldn’t expect the deepest of discounts at a store that caters to customers like you.

GET SEVEN-DAY RETURN PRIVILEGES. Your room at home will almost certainly be so different from the store’s listening mom that even the most thorough and expert listening test in the store can only roughly predict how a particular pair of speakers will sound in your system. Buy speakers from a store that will let you return them (in good condition) for a full re fund after a home trial of at least a week.

If a speaker disappoints you, help your self and the store by remembering what went wrong at home. If the bass was weak, pick a speaker with an extra-fat low end for your next try. If the treble was harsh in your room, a speaker that sounds slightly too mellow in the store is a good bet.

LISTEN TO MANY RECORDINGS. The sound of any recording depends heavily on the monitoring system the engineer employed to judge microphone choice and placement. Using only a single recording to help choose your new speakers will prejudice your selection in the direction of the particular monitoring system used in making it. So listen to many recordings, and look for speakers that sound accept able with all of them. (I’ll give you my personal choices for listening material later on.)

TAKE YOUR OWN MUSIC. Back in the Sixties, a speaker company put out a sampler record that successfully sold thousands of one particular model in its line. The speaker in question had a prominent treble peak, but the record had been made with a dip at that frequency, so it sounded right on the target speaker and strangely lifeless on most others. Even in the absence of such obvious trickery, a salesperson can choose, wittingly or not, recordings that favor speakers with certain sonic characteristics. So take your own music, and rely primarily on it rather than the store’s demo material. I use CD’s for this purpose because their sound is so consistent from player to player. But if you’re a die-hard LP fan, use black vinyl for listening tests, as its typically warmer frequency balance (and the altered stereo image it tends to produce) will influence your choice. -

INSIST ON COMPARABLE PLACEMENT. However little a store’s demonstration room may resemble your listening room, you should try to put the speakers you are auditioning in positions similar to the ones they will occupy in your home. If they’re on a shelf and you’ll be using them on floor stands, ask the salesperson to move them. Then listen both from the “sweet spot”—centered between the speakers—and from various other places around the room. If the tonal quality changes for the worse as you move off center, beware: You won’t hear much off-axis sound in most demo rooms because their walls are relatively absorptive, but at home it will likely be a problem.

USE PROPER COMPARISON TECHNIQUES. Most of the time you will be comparing two sets of speakers. It is very important that these be played with equal loudness, since the louder speaker will do better in such a com parison than it would with matched levels, giving it an unfair advantage. Normal differences in response between speakers make it impossible to match levels perfectly for all recordings, so for each selection have the salesperson adjust the level of one system until the two sound equally loud.

Left to his own devices, the sales man will wind up holding the speaker switch. Take control of it yourself, and try to avoid comparing apples and oranges: Don’t switch in the middle of a musical phrase, but look for a repeated passage and play all of it through each speaker in turn. Better still, find a CD player with a phrase-repeat feature and use it to listen to exactly the same musical fragment on both systems. Return to this tactic to verify your impressions whenever one system be gins to sound distinctly better than the other.

USE VOICE TO JUDGE THE MIDRANGE. The balance of frequencies between about 150 and 3,000 Hz is the most important single characteristic of any speaker, so I always begin by checking this range, using the tenor solo in Track 1 of the Christopher Hogwood CD of highlights from Handel’s Messiah, L’Oiseau-Lyre 400 086-2. (If you listen mostly to rock, pick a typical male vocal from a group you like for this test.) Tenor Paul Elliott’s voice should be clear and brilliant but not piercing or thin; if it sounds harsh in the loud passages (at 1:40 or 2:30, for example), you’re probably hearing a coloration that will bother you consistently at home.

The tenor voice should appear to come from a point midway between the speakers, assuming you have positioned yourself carefully at the same distance and angle from both. (This is also true of virtually all pop and rock vocals, which are recorded and mixed in mono,) If the soloist’s location seems uncertain, or if the vowels in his voice seem to come from one place and the consonants (especially 5, F, and T) from somewhere else, the responses of the two speakers in the pair probably don’t match well, and stereo imaging will suffer.

Other things to listen for with the Hogwood CD: In Track 4, the baritones on the right should be clearly distinguishable in both location and tonal quality from the tenors just to the right of center, the boy sopranos on the left should sound slightly too bright in the upper treble (a coloration from the recording microphone), and there should be plenty of room reverberation without any muddiness.

For the upper midrange and “presence” range (500 to 3,000 HZ), a female vocal is the perfect test. I use two selections. First is Amanda from Sheffield’s “Crème de la Crème” disc (CD-CRM). Amanda McBroom’s voice is slightly bright in this recording, like the percussion tracks, but it should not sound harsh or thin, and it should remain centered between the speakers. My second selection is Bird on a Wire, from Jennifer Warnes’s “Famous Blue Raincoat” ( Cypress 611111-2). This disc can also help you to identify one kind of high-frequency coloration to avoid in speakers: The vocal on Track 6, Ain’t No Cure for Love, was recorded by a different team of engineers from the others and has an unpleasant-sounding peak at around 6,000 Hz. If you hear this effect consistently, rather than on just the one track, the speaker has a problem. (The repeated-octave bass figures in -this track are useful, too; all three notes should have equal weight on a speaker with good deep bass.)

CHECK FOR MIDBASS BOOM. Resonances in the range between 80 and 150 Hz sound boomy and can become very annoying after a while. Midbass boom can give a deceptive initial impression of good deep-bass response, but it actually tends to make it harder to hear the system’s low bass. The most stringent test I have found for this region of the spectrum is “The World of the Harp” with Susann Mc Donald (Delos DICD 3005). Rare is the speaker that will reproduce the large harp in Track 10 of this disc with complete clarity. Another good test is Train in the Distance, Track 7 of Paul Simon’s “Hearts and Bones” (Warner Brothers 23942-2). Clear, warm- sounding background vocals with no trace of one-note quality are a sign of freedom from cabinet or woofer resonances. But beware: The low end of the spectrum is most susceptible to room effects. Consequently, the most important place to perform this test and the following one is in your home.

TEST FOR DYNAMICS AND BASS POWER. My old standby for low-end testing is the excerpt from Stravinsky’s Rite of Spring on Track 5 of the Telarc Sampler, Volume II (CD 80102). It’s okay for the brass to sound slightly nasty in this selection; what you’re really listening for is the weight and impact of the bass drum. A system with good power handling and bass below 40 Hz will reproduce this excerpt with palpable impact and no sign of strain. But be careful: Many power amplifiers, especially those found in small or medium-size receivers, will distort audibly if you get careless with the volume control, so if you hear crackling sounds, they may not be the speaker’s fault. When you’re comparing speakers, use the CD player’s phrase-repeat feature to single out the section from 1:43 to 2:17 and listen to the entire passage on each system in turn, switching while the player is returning to the starting point of the passage between repeats.

CHECK THE STEREO IMAGING. The apparent breadth, depth, and clarity of the stereo image produced by a pair of speakers are qualities on which audiophiles often lavish considerable time and money. If the speakers are care fully placed, you can keep track of their imaging as you conduct the previous tests. The Hogwood Messiah and Track 2 of the Jennifer Warnes recording are especially revealing of the fine points of placement and depth in the stereo image.

If you plan to entertain friends with your system, you will want it to pre sent a decent image to listeners who are sitting away from the center line between the speakers. Give the speakers a walk-around test, noting whether the musicians seem to remain arrayed between and behind the speakers as you move about the room or, instead, tend to bunch up in the nearer speaker as you move off center.

If, on the other hand, your goal is to create the best image for yourself in the best listening spot, center yourself precisely between the speakers (meaning that your distance from each one is the same within an inch or so) and check out Track 4 of the Talking Heads CD “True Stories” (Sire 25512- 2). There is a percussion track that circles right to left behind the speakers and then, on a good system, from left to right out in front, seeming to pass right through the center of your head. Successful reproduction of this effect (and of the little flute figure at 2:32 that seems to come from 90 degrees to your right) is a sign of the good symmetry and careful placement of speakers and listener that are necessary for really precise imaging.

ADAPT THESE PROCEDURES TO YOUR NEEDS. Although I’ve suggested specific program material for listening tests, there is nothing magical about my selections; I just know from experience that they work well as speaker-evaluation tools. In time you will be able to choose from among your favorite recordings those that work best to supplement (or replace) the music I have suggested. Whatever music you choose, use a wide variety of sonic textures, and look for a speaker that highlights the differences. The more strongly a speaker’s character comes through on all recordings, the worse its colorations. Conversely, the systems that change the most as you try different selections will be easiest to live with. Careful, systematic listening is your best defense against unpleasant surprises—and your best hope for finding speakers that will please you for a long time to come.


If you’re interested in the fine details of speaker performance, there is a test CD that can provide a lot of information without sound-measuring instruments:

“Compact Test: Demonstrations et Essais” on the Pierre Verany label (PV-784031). Track 20 contains a series of tones, each of which varies randomly over a range of one-third of an octave around center frequencies from 1 20,000 HZ. These “warble tones” can quickly and easily reveal just how low a system goes and how much bass distortion it produces. The lowest three bands should be felt, not heard; any audible components are distortion in the amplifier or woofer. This test will also mercilessly reveal any rattles and buzzes in the speakers (or anything else that’s not bolted down).

Finally, Track 18 contains pink noise, a constant waterfall sound with an even distribution of all frequencies and equal energy in every octave. With a little practice you can use pink noise to quickly find the most serious colorations in any system. Listen to 30 seconds of this signal at a moderately loud level on the best systems you can find, either in stores or at friends’ houses, and you will slowly accumulate an approximate mental image of how it should sound.

A word of caution about warble tones and pink noise, though: Either of these signals will tend to empty a room very rapidly. If you want to stay friendly with a store, don’t try these tests with other customers around.

= = == = =

Source: Stereo Review (Jan. 1991) by E. Brad Meyer, president of Point One Audio in Lincoln, Massachusetts, is an audio consultant, writer, and recording engineer and producer.

Prev. | Next

Top of Page   All Related Articles    Home

Updated: Friday, 2016-09-23 21:50 PST