Bookshelf Speaker Reviews

For decades, the minispeaker has been a touchstone for audio enthusiasts. Because you can get a respectable pair of minis for a few hundred dollars, the mini is where most audiophiles start their journey into sound.

Although there are decent minispeakers at $100 a pair, and incredible minispeakers at $1,000 (or even $10,000) a pair, the range between $249 and $399 per pair has long been the "sweet spot"—the price at which manufacturers can afford to use good drivers, a more complex crossover network, and a reasonably rigid enclosure. Appealing as these affordable minis can be, it's hard to find trustworthy information about them, because audio mags understandably focus on big, sexy tower speakers.

In this roundup, the minis at long last move to center stage. We collected eight models priced between $249 and $398 (later adding a ninth), and reviewed 'em in the best way possible: a blind test with multiple listeners. Then I measured each one with my Clio FW audio analyzer to get an idea of its technical performance. We'll show the frequency response graphs here; for the full technical rundown on each speaker, visit soundandvisionmag.com.

Of course, the hardcore audiophiles will want to know the particulars, so here's the small print, so to speak:

All speakers sat behind thin black fabric scrims on 28-inch-high stands. We auditioned the speakers in two groups of four, with the speaker order changed for every panelist. All speakers were auditioned with the grilles off. My custom-built active switcher allowed me to match the levels of the speakers within approximately 0.2 dB. Both ATI AT2007 and Outlaw Audio Model 7700 amplifiers (which are made on the same production line and use identical circuit topology) provided the power. A handheld controller let the panelists switch among the four speakers in each group. I offered each panelist the choice of sourcing audio from their smartphone or using their laptop computer with my Firestone Audio ILTW USB digital-to-analog converter.

Our panelists were mostly our usual West Coast testing crew: me (for whom the test was not blind), contributing tech editor Geoffrey Morrison, and voice actress Lauren Dragan. I also brought in a fresh set of ears: Greg Fischer, a journalist who recently started working on a couple of A/V-focused websites.

Note that while some of these speakers definitely outperform others, there was no clearly superior product. We generally agreed on which of these speakers are real contenders, but our personal preferences among the favorites differed. Picking a "winner" in this case would be arbitrary, so instead we'll help you figure out which one or two might be the best choice for you.

The models reviewed for this test are listed below. Click on the link for a specific model for our impressions and lab results.

• Axiom Audio M3v3 ($378/pr)
• Hsu Research HB-1 MK2 ($318/pr, $378/pr in Rosenut wood veneer shown)
• KEF C3 ($350/pr)
• Klipsch Reference RB-41 II ($299/pr)
• Monitor Audio Bronze BX1 ($379/pr)
• Music Hall Marimba ($349/pr)
• Paradigm Atom Monitor 7 ($398/pr)
• Polk Audio RTiA1 ($325/pr)

Measurement Notes

I measured the frequency response of the minispeakers using quasi-anechoic technique to remove the effects of reflections from nearby objects. I measured the speakers with each placed atop a 2-meter-high stand. The microphone was placed at a distance of 1 meter, which with small speakers like this is adequate to capture the contributions of all drivers and cabinet diffraction. For each speaker, I positioned the microphone directly in front of the tweeter, then adjusted the mike position slightly to get the flattest response. The quasi-anechoic measurements were made without grilles and smoothed to 1/12^th octave. Bass response of all speakers was measured using close-miking technique, with the mike positioned about 5mm from each woofer and port. Bass response measurements were scaled and summed, then spliced to the quasi-anechoic measurements at 220 Hz. Results are normalized to 0 dB at 1 kHz. All frequency response measurements were made with a Clio FW audio analyzer then imported into a LinearX LMS analyzer for post-processing.