Andrew talked about how he linearized the concentric motor system and by doing so stabilized the fixed field so it doesn't get modulated by the signal current, which leads to nonlinearities, ie, distortion products that, because they occur after the crossover, will not be attenuated. (The crossover filters the signal going into the cone, not the sounds coming out of it.) He also discussed how, in a concentric drive unit, there's always a limit to how much magnetic energy you can get to the tweeter, which you need for it to have a high sensitivity. Andrew came up with a structure where the woofer and tweeter magnets contribute to each other's magnetic field, resulting in a greater flux density than either motor could achieve alone. He was going to call it a "compound coupled magnet structure," but MoFi decided on "Twin Drive." "Being a twin, I'll take that one!" exclaimed Andrew, whose twin brother Owen is also an audio engineer.
Setting up
I left it to Jones and Jon Derda to optimize the positions of the SourcePoint 10s in my room. I had initially set the loudspeakers vertically on 18" single-pillar stands. However, after listening to some familiar recordings, Jones and Derda felt that 24" stands would be better, with the speakers sitting horizontally (see photo), where they resembled the Gale 401 speakers I had used for a while in the late 1970s. The MoFis were attached to the stands' top plates with Blu Tack pads, and although the stands' pillars were filled with a mixture of sand and lead shot, we stabilized the bases with small bags of lead shot. The stands were spiked to the wooden floor beneath the carpet.
The front baffles ended up 73.5" from the wall behind the speakers, the right-hand woofer 51.5" from the books that line that speaker's closest sidewall, and the left-hand woofer 31.5" from the LPs that line that speaker's sidewall. Jones and Derda toed out the SourcePoint 10s slightly from the listening position, and while the coaxially mounted tweeters were 32" from the floor, a few inches below the height of my ears, the 3° taper of the speakers' sidewalls meant the drive units were tilted up toward my ears by exactly that much.
The main source of music was my Roon Nucleus+ server feeding audio data over my network to a Roon Ready MBL N31 CD player/DAC, which in turn was connected directly first to the Schiit Tyr monoblocks I reviewed in the January 2023 issue, then to my regular Parasound Halo JC 1+ monoblocks. I didn't use the MoFi SourcePoint 10s' magnetically attached grilles, and the speakers were single-wired with AudioQuest Robin Hood cable.
Listening
To make sure the SourcePoint 10s had settled down after setup, I used the speakers for a week's worth of casual music playback before I started my critical listening. Then, as always, I started with the 1/3-octave warble tones on my Editor's Choice CD (STPH016-2). The SourcePoint 10s cleanly reproduced the tones down to the 40Hz band, and the 32Hz tone was reinforced by a room mode. The 25Hz tone was audible at my usual listening level, but I couldn't hear the 20Hz tone. There was no audible wind noise from the ports with these last two tones. The warble tones sounded clean, with no distortion. The half-step–spaced tonebursts on Editor's Choice spoke cleanly and evenly down to 32Hz, though those around 2kHz were slightly accentuated. Listening to the enclosure's walls with a stethoscope while the tonebursts played, I could hear some liveliness in a narrow band centered around 400Hz. The dual-mono pink noise track on Editor's Choice was reproduced as a stable central image. The noise sounded smoothly balanced and uncolored, though with a slight lift in the high treble. I was aware of this character imparted to the violin when I listened to Mozart's Flute Quartet in D Major from Serenade, which I recorded live at the Santa Fe Chamber Music Festival in 1996 and released as a Stereophile CD (STPH009-2). The balance wasn't bright as such, but it sounded slightly tipped up, emphasized by the lack of low frequencies on this recording. With orchestral recordings that were more fullrange, like the 1966 performance of Elgar's Sospiri with Sir John Barbirolli conducting the Philharmonia Orchestra (16/44.1 ALAC, Angel 67264), which I recently ripped from the CD, the presentation was tonally well-balanced.
Only when the recording itself had a treble-forward balance, like the Classic Records reissue of the early 1960s performance of Mendelssohn's Fingal's Cave with Peter Maag conducting the LSO (16/44.1 ALAC rip from CD, Classic Records CSCD 6191), did I feel that the SourcePoint 10's high-frequency balance was getting in the way of the music. But even then, the MoFi speaker's treble sounded clean—just too high in level.
I have written before that the acid test for a loudspeaker's lack of midrange coloration is solo piano. The absence of masking with the instrument's intrinsic tonal character coupled with the frequency gaps in the equal-tempered music scale allow problems no place to hide. Robert Silverman's Steinway D in the Stereophile recording of the Canadian pianist performing Liszt's B-minor Sonata (16/44.1 ALAC, from Sonata, STPH008-2) sounded uncolored, as did Lars Vogt's piano in his recording of the complete Brahms Violin Sonatas with violinist Christian Tetzlaff (DSD128 files, Ondine ODE1284-2D/HDtracks). That midrange cabinet resonant mode I heard with a stethoscope didn't appear to color the sound of the piano, though there was again a touch of character to the sound of the violin at the top of its range in the Brahms.
The piano's left-hand register in both the Liszt and Brahms recordings sounded appropriately powerful, without any boom or hangover. The low frequencies on my unreleased recording of Jonas Nordwall performing Widor's Organ Symphony No.5 (24/88.2 AIFF file) were reproduced with a combination of massive weight and good control. With the sub-40Hz pedal notes at the work's climax and an spl at the listening position of around 93dBC (slow ballistics), measured with the Studio Six app on my iPhone, the woofer's excursion appeared to be less than ½" peak–peak.
The SourcePoint 10s' stereo imaging was precise and stable. The positions of the three string players on the stage of Santa Fe's St. Francis Auditorium in my Mozart's Flute Quartet recording were clearly presented in the soundstage behind the central image of flutist Carol Wincenc. The supportive acoustic of the Albuquerque church in the Liszt Sonata could be heard surrounding the image of the piano. And on "Somewhere in Hollywood," from one of my picks for this issue's Records 2 Live 4 feature, 10cc's Sheet Music (16/44.1 FLAC, Tidal), the many multitracked voices and instruments were stably presented in a layered soundstage.
Jumping forward 50 years, from the 10cc album to a modern pop record, a man my age has no business being a "Swiftie." But my wife and I were driving into Manhattan and she played a song on the car stereo that I didn't know at all but wanted to. It was "All Too Well (Taylor's Version)" from Taylor Swift's album Red (Big Machine Records). The following day, I cued up the 24/96 Qobuz version in Roon and pressed Play. OMG. Taylor's voice had a touch of sibilance emphasis, but the sound was impressively clean. I turned up the volume so that the spl at my chair averaged around 100dB (C-weighted). This is about as loud as I can stand to listen, but the sound remained clean, with the bottom-octave bass line in the chorus swelling magnificently into the room.
The SourcePoint is a high-dynamic-range, almost full-range design. Going back to my regular KEF LS50s, which I love for their lack of coloration and their superbly accurate stereo imaging and soundstaging, I was painfully aware that the KEFs are small, limited-loudness loudspeakers with restricted low frequencies and a slightly suppressed top octave.
Conclusion
Its high-frequency balance means that the MoFi SourcePoint 10 will have a more neutral treble in rooms larger than mine. But even in my room, it didn't sound bright as such, though it was very revealing of the sonic differences between the amplifiers with which I used it. But when you consider the clean, superbly well-defined low frequencies, the natural-sounding midrange, the high sensitivity, the easy-to-drive impedance, the ability to play loudly without strain, and the affordable price, the SourcePoint 10 gets a thumbs-up from this reviewer. Good job, Mr. Jones.
I left it to Jones and Jon Derda to optimize the positions of the SourcePoint 10s in my room. I had initially set the loudspeakers vertically on 18" single-pillar stands. However, after listening to some familiar recordings, Jones and Derda felt that 24" stands would be better, with the speakers sitting horizontally (see photo), where they resembled the Gale 401 speakers I had used for a while in the late 1970s. The MoFis were attached to the stands' top plates with Blu Tack pads, and although the stands' pillars were filled with a mixture of sand and lead shot, we stabilized the bases with small bags of lead shot. The stands were spiked to the wooden floor beneath the carpet.
The front baffles ended up 73.5" from the wall behind the speakers, the right-hand woofer 51.5" from the books that line that speaker's closest sidewall, and the left-hand woofer 31.5" from the LPs that line that speaker's sidewall. Jones and Derda toed out the SourcePoint 10s slightly from the listening position, and while the coaxially mounted tweeters were 32" from the floor, a few inches below the height of my ears, the 3° taper of the speakers' sidewalls meant the drive units were tilted up toward my ears by exactly that much.
To make sure the SourcePoint 10s had settled down after setup, I used the speakers for a week's worth of casual music playback before I started my critical listening. Then, as always, I started with the 1/3-octave warble tones on my Editor's Choice CD (STPH016-2). The SourcePoint 10s cleanly reproduced the tones down to the 40Hz band, and the 32Hz tone was reinforced by a room mode. The 25Hz tone was audible at my usual listening level, but I couldn't hear the 20Hz tone. There was no audible wind noise from the ports with these last two tones. The warble tones sounded clean, with no distortion. The half-step–spaced tonebursts on Editor's Choice spoke cleanly and evenly down to 32Hz, though those around 2kHz were slightly accentuated. Listening to the enclosure's walls with a stethoscope while the tonebursts played, I could hear some liveliness in a narrow band centered around 400Hz. The dual-mono pink noise track on Editor's Choice was reproduced as a stable central image. The noise sounded smoothly balanced and uncolored, though with a slight lift in the high treble. I was aware of this character imparted to the violin when I listened to Mozart's Flute Quartet in D Major from Serenade, which I recorded live at the Santa Fe Chamber Music Festival in 1996 and released as a Stereophile CD (STPH009-2). The balance wasn't bright as such, but it sounded slightly tipped up, emphasized by the lack of low frequencies on this recording. With orchestral recordings that were more fullrange, like the 1966 performance of Elgar's Sospiri with Sir John Barbirolli conducting the Philharmonia Orchestra (16/44.1 ALAC, Angel 67264), which I recently ripped from the CD, the presentation was tonally well-balanced.
The SourcePoint 10s' stereo imaging was precise and stable. The positions of the three string players on the stage of Santa Fe's St. Francis Auditorium in my Mozart's Flute Quartet recording were clearly presented in the soundstage behind the central image of flutist Carol Wincenc. The supportive acoustic of the Albuquerque church in the Liszt Sonata could be heard surrounding the image of the piano. And on "Somewhere in Hollywood," from one of my picks for this issue's Records 2 Live 4 feature, 10cc's Sheet Music (16/44.1 FLAC, Tidal), the many multitracked voices and instruments were stably presented in a layered soundstage.
Jumping forward 50 years, from the 10cc album to a modern pop record, a man my age has no business being a "Swiftie." But my wife and I were driving into Manhattan and she played a song on the car stereo that I didn't know at all but wanted to. It was "All Too Well (Taylor's Version)" from Taylor Swift's album Red (Big Machine Records). The following day, I cued up the 24/96 Qobuz version in Roon and pressed Play. OMG. Taylor's voice had a touch of sibilance emphasis, but the sound was impressively clean. I turned up the volume so that the spl at my chair averaged around 100dB (C-weighted). This is about as loud as I can stand to listen, but the sound remained clean, with the bottom-octave bass line in the chorus swelling magnificently into the room.
The SourcePoint is a high-dynamic-range, almost full-range design. Going back to my regular KEF LS50s, which I love for their lack of coloration and their superbly accurate stereo imaging and soundstaging, I was painfully aware that the KEFs are small, limited-loudness loudspeakers with restricted low frequencies and a slightly suppressed top octave.
Its high-frequency balance means that the MoFi SourcePoint 10 will have a more neutral treble in rooms larger than mine. But even in my room, it didn't sound bright as such, though it was very revealing of the sonic differences between the amplifiers with which I used it. But when you consider the clean, superbly well-defined low frequencies, the natural-sounding midrange, the high sensitivity, the easy-to-drive impedance, the ability to play loudly without strain, and the affordable price, the SourcePoint 10 gets a thumbs-up from this reviewer. Good job, Mr. Jones.
