Measurements Here are the measurements for the HFF speaker (click on each chart to view it in a larger window).
Frequency response On-axis: ±5.0 dB from 37 Hz to 20 kHz Average ±30° horiz: ±4.7 dB from 37 Hz to 10 kHz, ±6.1 dB to 20 kHz Average ±15° vert/horiz: ±4.5 dB from 37 Hz to 10 kHz, ±4.7 dB to 20 kHz
Impedance min. 4.5 ohms/1.1 kHz/-3.8°, nominal 7 ohms
Sensitivity (2.83 volts/1 meter, anechoic) 84.7 dB
The first chart shows the frequency response of the HFF. The second chart shows the impedance. The computer that runs my LMS analyzer broke down as I was putting these measurements together, so I am temporarily unable to present charts with average responses. In the meantime, I've presented a chart showing the response at 0° on-axis and 10°, 20°, 30°, 45°, and 60° off-axis. Ideally, the 0° curve should be more or less flat, and the others should look the same but should tilt down increasingly as the frequency increases.
The response of the HFF is pretty flat overall, although there's a definite downward tilt in the tonal balance. (Translation: more bass, less treble.) That's likely a choice rather than a flaw. The only real anomaly in the response is that roughly two-octave-wide dip in the treble, centered at about 11 kHz. It's reasonable to speculate that this is why, as you'll learn later in the article, I found the treble to sound "polite." Adding the grille has only mild effects on the sound, but it does reduce treble energy an additional -1.5 dB or so from 6 to 8 kHz and 12 to 15.5 kHz.
Close-miked measurements of the woofers and midrange driver told me more about the crossover. It turns out there's a lot of overlap between the woofers and the midrange. Above the 530 Hz crossover point, the woofer response rolls off at about -10 dB/octave, suggesting a first-order (6 dB/octave) electrical crossover. Below 530 Hz, there's not much roll-off on the midrange driver, only about -4.75 dB/octave, which a much sharper rolloff below 90 Hz. This is a place where an apparent cost-cutting move did impact sound quality; with a steeper crossover here (which would raise the parts cost), the midrange driver could play louder with less distortion, and the woofers would stay out of the midrange where they don't belong. The midrange transitions to the tweeter with a roughly -27 dB/octave low-pass roll-off, which is more in keeping with common practice.
Sensitivity of the HFF is a little on the low side at 84.7 dB (measured at one meter with a 2.83-volt signal, averaged from 300 Hz to 3 kHz), which means the HFF should hit 100 dB with about 32 watts. Impedance is about average at seven ohms; so, as long as your amp can put out about 50 watts per channel, it shouldn't have much problem driving the HFF.
Here's how I did the measurements. I measured frequency response using an Audiomatica Clio FW 10 audio analyzer with the MIC-01 measurement microphone, and the speaker driven with an Outlaw Model 2200 amplifier. I used quasi-anechoic technique to remove the acoustical effects of surrounding objects. The speaker was placed atop a turntable that elevated it three inches off the ground. The mic was centered on the tweeter axis and placed at a distance of two meters from the front baffle and a pile of denim insulation was placed on the ground between the speaker and the mic to help absorb ground reflections and improve accuracy of the measurement at low frequencies. Bass response was measured by close-miking the woofers and ports, then scaling the port responses appropriately and adding that sum to the woofer responses. I spliced this result to the quasi-anechoic results at 220 Hz. Results were smoothed to 1/12th octave. I made measurements with the grille off except as noted. Post-processing was done using TrueRTA software.
The Downside The HFF is a really good speaker for $699/pair, but don't go thinking it's a $3,000/pair speaker for $699. It does have three shortcomings that aren't evident in better, costlier speakers.
First is that the treble is a little on the polite side. I rarely found myself craving more highs, but many audiophiles do like to hear a little more treble than the HFF will give. The speaker has that classic soft-dome tweeter sound (it's clear, but in no way dazzling), which some audiophiles love and some don't. Probably because of this softer response, while the HFF sounded spacious, I didn't hear the fine resolution of spatial details and positioning that I've heard with costlier models.
Second is a slight cupped-hands coloration, which made Cecile McLoren Salvant's recording of "What's the Matter Now?" sound as if she had a hand ... well, not cupped around her mouth, but maybe one hand cupped a couple of inches away from her mouth. (Perhaps it's a result of the half-inch-deep recess the midrange driver sits in, or the fact that the woofers are active well up into the midrange.) I'm particularly sensitive to this coloration because I'm a hardcore speaker geek; I'm not sure most listeners, even most audiophiles, would even notice it, but you know, it's my gig to point out these things.
Third is that the upper bass isn't as tight as what I heard with the more expensive models I compared with the HFF. On pianist Tsuyoshi Yamamoto's recording of "Honeysuckle Rose," there was plenty of bass, and as with the Steely Dan recording, the pitch definition was solid. But the drops--subtle off-beat accents that double bassists throw in to keep things interesting--didn't have the punch and impact they do with some other speakers, particularly those with smaller woofers. The HFF's eight-inch woofers handle everything up to 530 Hz, and at those higher frequencies they don't have the "speed" and definition of a good 6.5-inch woofer.
Comparison and Competition Using my Van Alstine AVA ABX switcher, I pit the HFF in different groupings against three other tower speakers, all far more expensive: the $3,500/pair Revel Performa3 F206, the new $2,000/pair Revel Concerta2 F36, and the $11,495/pair Monitor Audio PL200 II. (Sorry, I didn't have anything in the HFFs' price range to which I could compare them.) While all of the more-expensive models sounded better overall, the HFF didn't commit any real errors. It simply sounded like a less costly speaker. Its treble was less smooth and clear, its midrange revealed a bit of cupped-hands coloration, and its bass didn't sound as precise.
Tower speakers are mostly a $1,000/pair-and-up category, so the models to which the HFF could be fairly compared are few. They include the $379-each Elac Debut F6, the $415-each Monitor Audio Bronze 6, the $349-each Klipsch R26F, and the $399-each Polk RTiA5. None of these speakers packs a driver complement as robust as the HFF's; at the most they have dual 6.5-inch woofers compared with the HFF's dual-eights. The HFF also has nicer fittings, although all of the competitors listed above are smaller and sleeker.
How do they compare sonically? The only ones I've heard are the Elac and Monitor Audio models. Neither is quite as muscular in the bass as the HFF; if you want to play stuff loud or listen to a lot of hip-hop or heavy rock, the Fluance is almost certainly your best choice of the bunch, although the others will likely have better upper-bass and lower-midrange definition. When it comes to the sound of the mids and treble, my guess (since I didn't have a chance to compare the speakers directly) is that the midrange might be kind of a toss-up, with the Elac and Monitor Audio models sounding a little more open, and I'd probably have a slight preference for the more vivid treble of the Elac and Monitor Audio models.
Conclusion The HFF has to be considered one of today's best values in a tower speaker. There are certainly reasons to spend more, and there are speakers at this price you might prefer if you're willing to sacrifice some bass response for midrange and treble clarity. But if I had to pick a "best tower in the $700/pair range for most people," the HFF would be the first one to come to mind.