Stand Loudspeaker Reviews

John Atkinson auditioned the Dynaudio Focus 10 with Dirac Live in November 2023 (Vol.46 No.11):

Jason Victor Serinus was impressed by this small, two-way, active all-in-one speaker from Denmark (footnote 1). He concluded his February 2023 review by writing, "Designed as a 'complete wireless sound system,' the Dynaudio Focus 10 active monitors are the most musically satisfying one-stop playback solution I've had the pleasure of hanging with. ... Within its limitations of image size and bass quantity, the Focus 10 is one of the most musically satisfying active loudspeakers I've had the pleasure to audition. Highly recommended."

I was also impressed by the Focus 10's measured performance, which, as I wrote in the Measurements sidebar, "indicates a well-sorted design." Aside from a slight excess of energy between 1kHz and 1.6kHz and a slight lack between 5kHz and 7kHz, the farfield frequency response, averaged across a 30° horizontal angle centered on the tweeter axis, was respectably even (fig.1).

JVS wasn't able to investigate one aspect of the Dynaudio speaker's performance in his review: built-in Dirac Live, which can be used to equalize and optimize the Focus 10's in-room behavior. I still had the Dynaudios I had measured for the original review (footnote 2), and as I had used Dirac Live successfully with the sample of NAD's M10 integrated amplifier that I had purchased after my review, I decided to examine the effect of Dirac Live, which is the objective of this follow-up review.

I followed the instructions in the manual to set up the Dynaudio Focus 10s, first installing the Dynaudio Connect and Control app on my iPad mini then connecting the Primary speaker to my router. After the speaker was connected to the network, it updated its firmware to version 1.2.880xb3ed1c3. (Another firmware update, which will offer bass-management options and allow Dirac to work with external subwoofers, will be available later this year.)

The source for my auditioning was my Roon Nucleus+ server sending audio data to the left, Primary speaker over Ethernet. The Primary speaker was connected to the right, Client speaker with a 4m length of coaxial S/PDIF cable. I placed the speakers in the positions where my KEF LS50s worked well: The left speaker's woofer was 31" from the LPs that line the nearest wall; the right speaker's woofer was 49" from the books that line its nearest wall. Both speakers sat on 24" Celestion stands, their single pillars filled with a mixture of dry sand and lead shot, and were 73" from the wall behind them. I left off the grilles.

I used the Dynaudios without Dirac room correction for a couple of weeks in order to become familiar with their sonic character. The high frequencies were a little on the hot side, so I used the Dynaudio app to set the tonal balance to "Dark." The blue trace in fig.2 shows the difference in the anechoic tweeter-axis response with the "Dark" setting. The output at 20kHz is tilted down by 1.1dB; the output at 100Hz is tilted up by the same amount.

With this tonal balance, the highs were at the optimal level referenced to the natural-sounding, uncolored midrange. Stereo imaging was precisely focused, and the soundstage extended behind the speakers when appropriate. What impressed me most was the Focus 10's bass. I was streaming random tracks with Roon when my attention was caught by a recording of Japanese percussionist Kuniko Kato performing J.S. Bach's Cello Suite No.1 in G Major on a marimba (24/96 FLAC, Linn Records/Qobuz). The low notes had more weight than I was expecting from these small speakers with 5.5" woofers.

As discussed in JVS's review, the Focus 10 uses digital signal processing to adjust its low-frequency extension and sealed-box woofer alignment depending on the output level. At the lowest sound pressure levels, the Focus 10 has full output down to around 30Hz. As the SPL increases, the LF rolloff starts higher in frequency with a shallower slope. At the SPL I was playing the Kuniko track, averaging around 77dB(C), the Dynaudios were offering extended low frequencies.

This can be seen in fig.3. The red trace shows the spatially averaged response of the Dynaudio Focus 10s at the same SPL in my room (footnote 3), with the "Dark" balance setting. The peak just below 30Hz is due to the excitation of the lowest-frequency mode in my room; the response extends at full level to 55Hz, though with a slight boost in the upper bass. For reference, the blue trace shows the spatially averaged response of the KEF LS50s. Both pairs of speakers offer a smoothly balanced upper midrange and treble; the peaks and dips below 400Hz are due to interference with the direct sound by boundary reflections. However, the LS50s' low frequencies roll off prematurely compared with the Focus 10s'. (For critical listening with the KEFs, I use Roon's parametric equalizer to extend their in-room low-frequency response to 42Hz. As long as I keep the SPL below 85dB(C), the little KEFs offer effective low-frequency performance without noticeable distortion.)

It was time to investigate the effect of Dirac Live (footnote 4). I installed the Dirac Live app on the Mac mini and plugged the USB microphone supplied with my NAD M10 into the USB port on my Mac mini. When I ran the app, it found and identified the NAD microphone and the Dynaudio speakers, set the speaker's tonal balance to Neutral, reduced the playback volume, and performed a loudness check. I then followed the on-screen instructions, placing the microphone in each of the nine positions specified by the app and performing a "chirp" test at each. After the last test, Dirac Live calculated a correction filter, named it "Anduin," and asked for it to be saved to one of the Focus 10's eight filter slots. I could then activate Anduin with the Dynaudio iPad app.

I started listening to some favorite tracks, but it was immediately obvious that there was too much midbass and upper bass. Bass guitar was too rich-sounding; the piano's left-hand register, in Vladimir Ashkenazy's performance of Rachmaninoff's Piano Concerto No.3, with Bernard Haitink conducting the Concertgebouw Orchestra (16/44.1 rip from CD, Decca), had way too much weight.

A measurement of the in-room response revealed that Dirac's Anduin filter was boosting the region between 60Hz and 200Hz by up to 5dB. Here's what I think happened. I had performed the Dirac response measurements at too high a level, which caused the Focus 10s to roll off their low prematurely. The Anduin filter was compensating for that rolloff. When, subsequently, I played music at a lower level, Dirac's correction in the bass was too much. It is necessary, therefore, to perform the Dirac measurements at the same SPL as will be used for the auditioning.

I repeated the Dirac Live calibration routine with the Dynaudio's volume control set to the same level, "57," that I had been using for my non-Dirac listening. I saved the new filter, "Rhine," to the Focus 10's second slot and measured the spatially averaged response at the same volume control setting (fig.4, red trace). Other than a small peak centered on 250Hz, the in-room response in the midrange and low-treble region was now superbly even compared with the red trace in fig.3, falling within ±0.7dB limits from 300Hz to 3.7kHz. The mid-treble slopes down a little more with Dirac's Rhine filter than it had with the "Dark" balance setting, which I heard as a slight lack of air on the Rachmaninoff recording. The upper bass was still around 3dB too high in level, and the speakers excited the lowest-frequency room mode to a greater degree than with the "Dark" setting.

To create the Anduin and Rhine filters, I used Dirac Live's suggested target response. For my next attempt, I modified the target response by reducing the amount of correction below 200Hz. The spatially averaged response with the resultant "Hegel" filter is shown as the blue trace in fig.4. The upper bass is now close to the average level in the midrange, though now there is insufficient midbass energy.

The lowest-frequency room mode was still excited with Hegel but not as much as with Rhine. My final experiment was therefore to adjust the target response to roll off the low bass. I also tried to compensate for the small peak at 250Hz. The in-room response below 310Hz with this filter, named "Broglie," is shown as the green trace in fig.4. Its output between 40Hz and 200Hz is identical to that of Hegel, and the peak between 200Hz and 300Hz has been reduced in level. The low frequencies now roll off precipitously below 35Hz, though there is a return to the Hegel output below 25Hz. (Dirac Live's target response appears to be restricted to frequencies above 20Hz.)

With the Dynaudio Connect and Control app, I could switch between the four correction filters I had created. I started listening with Broglie. While this filter's low-frequency rolloff will protect the Focus 10's small woofers from excessive excursions, I was not satisfied with what I heard. While stereo-imaging specificity was excellent and low frequencies were well-defined, soundstage depth seemed restricted and the hall ambience seemed suppressed on Encore, featuring live recordings I had made at the 1997 Santa Fe Chamber Music Festival (16/44.1 ALAC, Stereophile STPH011-2). Too much correction is too much of a good thing.

I abandoned Broglie and continued listening, alternating Rhine or Hegel. The double bass on Mendelssohn's Sextet in D, Op.110, sounded a little reticent with Hegel. Rhine brought it into the correct balance with the piano and string quartet. But on a recording of Vaughan Williams's Tuba Concerto (24/96 FLAC, Royal Concertgebouw/Qobuz), the blatty bottom register of soloist Perry Hoogendijk's instrument had a touch too much bloom with Rhine. Hegel worked best with this recording, as it did with Cantus performing Curtis Mayfield's "It's Alright" (16/44.1 ALAC, from Outside the Box, Cantus Records), on which I had boosted the upper bass a little in the mix to give the a cappella performance a sufficiently firm tonal foundation.

Overall, I found that the Rhine filter I had created with Dirac Live worked well with all kinds of music. On "Nell's Bells," from Attention Screen's Live at Merkin Hall (16/44.1 ALAC, Stereophile STPH018-2), the hall acoustic surrounding Bob Reina's Steinway grand, Don Fiorino's ukulele, Chris Jones's fretless bass guitar, and Mark Flynn's drums was palpable (footnote 5). The tonalities of the instruments were, for want of a better word, perfect, though, as JVS commented, the stereo image was small. (That's what you get with small speakers.)

Even without Dirac Live, the Dynaudio Focus 10 combines ease of setup and use, stereo imaging precision, and a natural-sounding midrange with surprising low-frequency extension for such a small speaker, provided the SPL is kept below 85dB or so. It would be a great fit'n'forget choice as a complete music system for use in relatively small listening rooms. Dirac Live takes its sound quality to another level.—John Atkinson

---

Footnote 1: The Focus 10 costs $5500/pair. Dynaudio A/S. Sverigesvej 15, 8660 Skanderborg, Denmark. US distributor: Dynaudio North America, 500 Lindberg Ln., Northbrook, IL 60062. Tel: (847) 730-3280. Web: dynaudio.com.

Footnote 2: The serial numbers were 10033212 (Primary, left) and 10033213 (Client, right).

Footnote 3: Using the FuzzMeasure 3.0 program, a Metric Halo MIO2882 FireWire-connected audio interface, and a 96kHz sample rate, I average 20 1/6-octave–smoothed spectra, individually taken for the left and right speakers, in a rectangular grid 36" wide by 18" high and centered on the positions of my ears.

Footnote 4: Something that should be noted if a Focus 10 owner is intending to use this speaker to play movie soundtracks is that Direct Live correction introduces around 61.5ms of additional latency—almost two video frames—to the uncorrected speaker's 11.7ms latency.

Footnote 5: See my essay on how I recorded this concert here.