“Let our rigorous testing and reviews be your guidelines to A/V equipment – not marketing slogans”
Facebook Youtube Twitter instagram pinterest

Evaluate Loudspeakers for Sound and Accuracy - Continued


If you’re going to understand anything about audio—how speakers work, the effect of room acoustics, how to set your subwoofer’s level, the concept of bass management/low-pass filters, anything about audio—you have to understand frequency response. This is the cornerstone of everything in audio. It all starts and stops with frequency response. Amplifiers, receivers, speakers, CD and DVD players, they all have to have as close to perfect frequency response as possible, or else they just won’t sound right. Nothing else matters if the frequency response is no good. Can’t make it any clearer than that.

Why Frequency Response is important: An audio device with good frequency response is able to play all the low, middle, and treble tones correctly—and in the proper proportion to each other—and that’s what tells our ears whether or not this is a high-fidelity unit with rich, vibrant, lifelike sound.

To understand frequency response, remember this: the loudness of sound is expressed in a unit of measure called the decibel, or dB. To define frequency response, we specify a range of frequencies, and then we state by how many decibels (dB) the equipment varies from perfect. For example, a speaker may be said to have a frequency response of 40 Hz-20 kHz (that’s the range), ± 3dB (that’s the variation).

Now it’s time for another hard part: Frequency responses are almost always shown as a graph. This graph is known as the “Frequency Response Graph.” (Clever, no?) You have to know how to read a graph, no excuses. If you paid attention in Mr. Kelleher’s 6th grade class, great. If not, you’ll be sorry now.

Now look at figure 2. The black line is a speaker with excellent frequency response. The frequency response curve (so-called because a speaker’s frequency response curves, or drops off, in the low bass and high treble) is pretty flat (“flat” is good, because it means the device is accurate with very little up or down variation), with no serious peaks or dips. For speakers, ± 2 or 3 dB is considered very good. Electronic equipment (receivers, CD players, etc.) should be within ± .5 dB.

The red line in Figure 2 shows a speaker’s frequency response with a big 7 dB peak (so-called because the graph looks like a mountain’s peak) in the upper midrange around 6 kHz, which will make it sound harsh and irritating.

Speaker FR Curves

Figure 2—Speaker frequency response curves—Good and Peaked

Figure 3 shows what speaker frequency response curves look like that correspond to various subjective descriptions.

Learn these terms and learn how they look as frequency response graphs, and you’ll be on your way to a meaningful understanding of loudspeaker sound.

Subjective descriptions vs FR

Figure 3—Subjective descriptions shown as frequency response curves

Try to listen in an environment similar to your home listening space because room acoustics have a tremendous influence—good and bad—on how a speaker sounds. Some rooms are too “live”, with lots of hard, reflective surfaces, which can make a speaker sound too bright and shrill. Conversely, rooms can be “dead,” or overly absorptive, like a room with thick carpeting, heavy drapes and overstuffed furniture. Speakers in a room like this can sound dull and lifeless. So bear in mind that a crowded, noisy open showroom floor or a small sound room filled to the gills with speakers and other equipment will sound nothing like your 12 x 15-foot den or your 17 x 27-foot living room with vaulted ceilings.

Live vs Dead Room

Room Acoustics—“Live” (left) and “Absorptive” (right)

One advantage of buying from an Internet Direct company with a home trial arrangement is being able to listen to speakers in the actual location where they’ll be used. Many ID companies have a trial period where you can live with the speakers for a week or more and return them if you don’t like how they sound in your home. This is actually a major improvement over the way speakers were typically purchased in years gone by, when customers would listen to several models in a retailer’s sound room. The problem with that, of course, is that a dealer’s sound room bore almost no acoustic resemblance to a consumer’s actual listening room, so people often got an unpleasant surprise when they got their new speakers home and discovered that they sounded quite different than they did at the store.


Subwoofers are certainly important, especially in home theater systems. These days, few people have a two-channel “music only” system with full range speakers and no subwoofer, so listening to a sub is part of the overall equation. Use the guidelines for evaluating bass quality, extension and clarity that we previously outlined, but add a few more for a subwoofer. Perhaps the most important thing about a subwoofer’s sound will be its ability to play loudly enough for your needs without noise or audible distress. Once you have determined that the sub in question has the quality of sound you’re looking for (by listening to it with music), you need to determine if it has the quantity of sound as well.

Play a movie soundtrack with a lot of deep bass special effects—explosions, crashes, dinosaur roars, etc. Play the movie slightly louder than you think you’d ever play it. Listen to the bass—does the sub fill the room with clean output or is it weak and struggling? If it sounds good, then replay the same sequence again—only this time, turn off all the other speakers and listen just to the sub by itself. Does it sound pretty clean, without any really noticeable distortion and is it free from severe distracting mechanical noises? It’s ok if there’s a little audible distortion and a small bit of mechanical noise, because remember, this is at a level higher than you’ll ever really play the system and when the other speakers are playing, they’ll mask the vast majority of the sub’s bad sounds. But….if the sub sounds pretty good all by itself at really loud levels without the other speakers to “cover for it,” then you can be pretty confident that the sub will do a great job in your room, playing your most demanding material at the loudest levels you’re likely to play.

 Velodyne Subwoofer

Velodyne subwoofer

Center Channel Speakers

Center channel speakers are critically important in home theater systems because typically 50-75% of a movie’s sound track is handled by the center channel speaker.

To evaluate the center channel speaker, it’s best to listen two ways:

1. Play music through it and evaluate its sound like any other speaker, using same standards we outlined previously. Chances are, you’ll only have one speaker to use, so set the receiver to “mono” so the single center speaker receives all audio channels.

2. Use the center as the left or right speaker of a stereo pair, with one “main” speaker as the other speaker in the pair. Use the receiver’s balance control to go L to R to see how tonally similar they are. They will not be identical twins, but they should be at least close relatives. Then evaluate them as a stereo pair. Is the image stable? That will tell you if their radiation patterns are similar enough. Do they sound “right” as a pair? That will tell you if their tonal signatures are similar enough. A similar tonal signature is absolutely essential for the front stage image to be believable as sounds pan across the front three left/center/right (LCR) speakers.

Atlantic Technology 1400c

Atlantic Technology 1400C center channel speaker

Source Material

Use good old-fashioned stereo music CDs. Don’t use compressed iPod MP3 files, because the audio quality is simply not good enough to use as a test source. Develop a collection of music that you know really well, and that covers different genres and musical formats. You’ll want music that tests for female vocals, male vocals, dense, detailed percussion, deep bass, acoustic piano, horns, guitars, strings, etc. Cover every musical scenario and performance venue. Studio recordings should sound tight and immediate; live recordings should have a realistic, believable sense of “air” and three-dimensionality, etc.

This is important—don’t pick test material based on your favorite music, because if you really love the music you’re playing, it can make the speakers sound better than they really do.

This is also very important—Use the same CDs over and over. The scientific method dictates that you keep the variables to an absolute minimum and try to isolate and identify the data points you’re after. Become very familiar with the sound of a few CDs and use them on a wide range of speakers. You can’t compare the bass performance of different speakers using Steely Dan’s Aja on one speaker and Saint-Saëns Organ Symphony on the next. I’ve come up with my collection over about a 20-year span. They work really well for me, and not necessarily because they’re “perfect” recordings, but because I know what is on them and I can consistently evaluate the sound. If the vocalist on a particular CD sounds laid-back and very smooth on 9 out of 10 speakers, you can be pretty sure that’s what the recording is like. If speaker number 11 makes the vocalist sound too forward and edgy, you can be pretty confident it’s the speaker’s fault.  That’s the advantage of knowing your CDs—they should become like reliable test equipment. The CDs listed in Audioholics’ equipment reports are a very good starting point for a good test collection.


If you follow these guidelines and learn this basic information, you’ll be well equipped to evaluate speakers and be confident about your impressions of their sound:

  •  Use music, not special effects, as your “test” material for listening
  •  Listen in two-channel stereo. Multi-channel music can fool the ear with distracting sonic “trickery.”
  • Try to listen in an environment similar to your home listening space
  • Use CDs, not compressed MP3 files
  •  Choose a wide range of musical genres and formats, to give the speakers a complete workout
  •  Use the same CDs over and over, to ensure consistent, easily-comparable listening results


Confused about what AV Gear to buy or how to set it up? Join our Exclusive Audioholics E-Book Membership Program!

Recent Forum Posts:

FozzieT posts on April 30, 2014 11:26
Thanks for a great article. It really helped me identify things to focus on when evaluating my speakers and system set-up. In particular, the advice on how to evaluate the bass performance by following the bass line in a jazz trio or quartet has been extremely helpful.

I'd recommend the Vince Guaraldi Trio's “A Charlie Brown Christmas” CD as at least one recording to use to evaluate your bass set-up. Not only is it just a lot of fun to listen to, but it has great bass lines running through most of the songs.

Anyway, that's my $.02. Thanks again for a helpful article!
exlabdriver posts on April 29, 2014 19:05

The change covers the electric bass low range perfectly…

markw posts on April 29, 2014 16:29
BoredSysAdmin, post: 1030341
Regardless if this correct or not (and I DO agree - pure direct stereo is great choice for evaluating individual speakers) just because it's published doesn't make it “validated”.

Getting published now a days is so much easier than it ever was. Most people are lazy and anyone who's producing any original content should be considered published author and receive recognition and fame? By this logic I should be world renown expert in my field. After all I had over 420k unique page views on my humble blog

Yeah, it kinda does. But I'm not actually the first person to say it.

Perhaps it's new here but, in the overall scheme of things, it's not a new observation. And, I have said it here, just not recently.
zieglj01 posts on April 29, 2014 15:11
I always test in 2 channel stereo mode for music - if they cannot pass that, then forget it.
The main thing is to have good resolution, dynamics and not annoy/distract me. If they can
handle this, then I find they will also perform well for Home Theater and TV. A lot/most HT,
TV and games have a lot of artificial sound to begin with.
gene posts on April 29, 2014 15:05
exlabdriver, post: 1030321
The lowest note - B0 - on my 5 String Electric Bass is actually 30.868 Hz. These instruments are now widely used in modern music; however, while playing an open ‘B String’ is not all that common, the potential is there.

My Sat Bookshelves + each with its own 8" Sub can easily reach down to that range with authority that makes music reproduction very satisfying…


TAM, I found the statement and changed it to:

This is lower than the vast majority of so-called “full range” speakers can play on their own without the aid of a subwoofer. The lowest note on a 4 string electric bass, which is the lowest musical note in most popular music, is low E at 42Hz and Bo at 30Hz on a 5 string bass.

Thanks for letting me know.
Post Reply