The Sonus Faber Guarneri Evolution's B-weighted sensitivity on its tweeter axis was 85dB/2.83V/m, which is slightly lower than the specified 86dB and significantly lower than the norm. The electrical impedance (fig.1) remains above 5 ohms from the bass through the low treble, but drops to a minimum value of 4 ohms between 4 and 5kHz. The electrical phase angle is generally benign, but there is a combination of 5 ohms and –45° at 3200Hz that will require the use of a good, 4 ohm-rated amplifier with this speaker. A small wrinkle between 200 and 300Hz in the impedance traces correlates with a cabinet resonance at 254Hz that was detectable on the enclosure's side panels (fig.2). As Art Dudley didn't remark on any coloration that could be laid at the door of this behavior, I suspect this measurement looks worse than it sounds. (It is of very high Q or Quality Factor, which will work against its audibility; in addition, I couldn't detect this resonance on the enclosure's other surfaces.)
Fig.3 shows the individual outputs of the port (red trace), woofer (green), and tweeter (blue). The woofer's minimum-motion point occurs at 43Hz, as predicted by the impedance curve, which has a saddle in the magnitude trace at that frequency, and the port's output peaks between 30 and 100Hz. The port's output rolls off smoothly above 100Hz, and is free of the pipe resonances that so often afflict small reflex-loaded speakers. Both the woofer and tweeter, measured on the tweeter axis, are flat within their passbands, with the crossover set at 2.9kHz, very close to the specified 2800Hz. Both drive-units are well behaved outside their operating regions.
Fig.4 shows how these outputs sum in the farfield, averaged across a 30° horizontal window centered on the tweeter axis. The bump in the upper bass is entirely due to the nearfield measurement technique used to capture the drivers' low-frequency behavior; the speaker will be flat down to around 70Hz, with the output down 6dB at the port tuning frequency of 43Hz, which is close to the lowest note of the four-string electric and orchestral basses. The Guarneri Evolution is also extremely flat in response throughout the midrange and treble.
Despite its flat and extended top-octave output on axis, AD found the Sonus Faber to sound a little mellow. This is explained by the plot of the speaker's lateral dispersion normalized to its on-axis response (fig.5). The tweeter becomes quite directional above 9kHz or so, which will reduce the top-octave energy level in all but very small, live rooms. Other than that, the Guarneri Evolution's horizontal dispersion pattern is textbook in its evenness, with no off-axis peaks or suckouts to color the reverberant soundfield. In the vertical plane (fig.6), a suckout develops above the tweeter axis, but the balance doesn't change significantly for 10° below that axis. Sonus Faber's 31"-tall stands are optimal for this speaker.
In the time domain, the Guarneri Evolution's step response on the tweeter axis (fig.7) indicates that the drive-units are connected in the same positive acoustic polarity. The smooth blend of the decay of the tweeter's step into the rise of the woofer's step correlates with the excellent integration of their outputs in the frequency domain on this axis. This is a testament to the crossover design. Finally, the cumulative spectral-decay plot (fig.8) demonstrates a superbly clean decay, with no ridges of decayed energy to suggest the presence of resonances. (The black ridge just below 16kHz is the usual leakage from my computer's video circuitry.)
Yes, this gorgeous little speaker is expensive. But its superb measured behavior both correlates with its superb sound quality and suggests that Sonus Faber has some serious speaker-engineering expertise on its staff. It also measures better in many respects than the speaker it replaces, the Guarneri Memento, which Michael Fremer reviewed in August 2007.—John Atkinson
