PristineSound
Major Contributor
I came here to say that I am not a fan of the visual aesthetic of these speakers. But beauty is in the eye of the beholder.
Just saw a FB post of his that says he will continue to work for Mofi as well as with his new venture with two associates.Mr. Jones sure moves around a lot.

I'll take the TD124 vinyl spinner with arm, you lot can have the rest (no room in my future for large loudspeakers nowThe ‘Troubadour’ !
Photo courtesy of Mono&Stereo.
Keith
If they're perfectly fit for purpose (cf. the classic Altec theater and sound reinforcement loudspeakers), they're beautiful.Those Dutch and Dutch speakers on page one of the thread are UGLY! Just sayin.......
Courtesy of spinaroma.org.If that is in fact part of his process it seems to be working pretty well. Plenty of satisfied customers and his speakers for MoFi have generally measured pretty well.
In fact Erin of Erin’s Audio Corner who Klippel measures speakers and who prefers well measuring speakers, essentially found AJ’s MoFi V10 speakers to be tremendous performers and his second favourite speaker after the KEF Blades.
Who knows? Maybe his speakers measure better after the measurement portion and AJ dials them in a bit more by ear.
Or it could be the opposite - maybe he starts out with the speakers measuring a bit better/linear but after he’s done dialing a bit more by his ear they measure a bit worse. In any case if he’s using what he perceives to some significant degree in his designs his perception can’t be too badly off; it’s clearly still leading to some very good loudspeakers, 10 to measure pretty well. (Which is why even ASR members have been excited about possibly owning some of his new MoFi designs)!
I thought optimization by ear is all about “voicing” that is verifying that the speaker produces naturally sounding speech? No of us have reference to the recorded material or the sound in the control room, but for sure we know instinctively, and without reference to the original sound, if voices sound unnatural. Isn’t this voicing all that is being done?If that is in fact part of his process it seems to be working pretty well. Plenty of satisfied customers and his speakers for MoFi have generally measured pretty well.
In fact Erin of Erin’s Audio Corner who Klippel measures speakers and who prefers well measuring speakers, essentially found AJ’s MoFi V10 speakers to be tremendous performers and his second favourite speaker after the KEF Blades.
Who knows? Maybe his speakers measure better after the measurement portion and AJ dials them in a bit more by ear.
Or it could be the opposite - maybe he starts out with the speakers measuring a bit better/linear but after he’s done dialing a bit more by his ear they measure a bit worse. In any case if he’s using what he perceives to some significant degree in his designs his perception can’t be too badly off; it’s clearly still leading to some very good loudspeakers, which tend to measure pretty well. (Which is why even ASR members have been excited about possibly owning some of his new MoFi designs)!
Courtesy of spinaroma.org.
If you ask me, it does match the tuning by the ear of an elderly gentleman, where most have high frequency hearing loss. This is not a knock on elderly gentlemen, nor am I here to age discriminate; Dr. Toole openly mentioned once he got to a certain age, he completely excluded himself from listening tests.
I have heard this MoFi in a showroom, like Erin's feedback, I find it on the bright side.
View attachment 520815
www.audiosciencereview.com
And perhaps Andrew Jones took in the feedback from the first SourcePoint 10 and tamed it down. Remember the SourcePoint 10 was the first speaker in the series.I was referencing this newer model in terms of Erin’s remarks:
![]()
MoFi Sourcepoint V10 Master Edition
Erin's Review is up. He likes the speaker a lot even asking "Is This the Best Speaker of 2025?" It's not super linear (notice this is the 15° measurement): But not bad at all.www.audiosciencereview.com
(I also heard the top end rise in the original SourcePoint 10, as did JA in his Stereophile review as I remember)
According to his subjective and objective review, he liked the 8 better due to flatter response overall and much cleaner in the top octave. He also claimed that the 888 was the best speaker overall he has tested and in their price range, FWIWIn fact Erin of Erin’s Audio Corner who Klippel measures speakers and who prefers well measuring speakers, essentially found AJ’s MoFi V10 speakers to be tremendous performers and his second favourite speaker after the KEF Blades
Large credit goes to his ability to show a speaker in best light at shows with very nicely selected content, English accentIf that is in fact part of his process it seems to be working pretty well. Plenty of satisfied customers and his speakers for MoFi have generally measured pretty well.
There is enough research that shows majority of people preferring a linearly flat anechoic FR. As such, the prudent thing to do is you design for the majority. If the speaker is well designed, having "reasonable off-axis" as you put it, then those minority who have a certain preference can go to town and EQ it.Admittedly, the standard tuning highlighted the treble detail, potentially leaving a better first impression. And at any rate, it would be impossible to know 100% the correction is needed across all people and all content. But there is enough reasoning here to make a change. As a compromise, Asci chose to use a softer filter than I used.
I cannot imagine a product being designed over tens to hundreds of hours of sketches, modeling, analysis, component testing, etc. and then built, and then not testing the hell out of it.To me, the listening part isn't just a final check, it's an integral, important part of the process.
I've written about this several times, but I don't expect people to memorize everything I've written of course. Here's a quote from an article about designing the Manta that sums it up nicely:
While we are inherently science driven, it's of course not just about measuring the speakers. Listening sessions are a crucial and major part of the development. The speakers go through countless iterations of crossover / digital filter configuration, and basically all iterations has one version based on the measurements, and a subsequent variant "By ear".
This means I first tune them based on measurements on- and off-axis (attempting to balance the two), then I listen for a few days and do additional adjustments based on what I hear. Then perhaps something is difficult to get perfect, so I measure again to help me understand what the problem is. Then do some adjustments based on the measurements, and then go back to listening and additional adjustments based on the listening sessions. This way the response is continuously balanced based both on on objective data and real listening impressions over time, and across a broad range of music genres.
And a similar section about the Saranna development:
We use both measurements and listening sessions to dial in the rich, natural sound that is the signature of all Sigberg Audio products. A linear anechoic response is a good starting point. Then we measure and listen our way to a combined dispersion characteristics that translate into that "just right" sound in the sweetspot.
This isn't really about voicing it to a "house sound", but trying to make it sound right on everything. Which sounds impossible perhaps, but it's actually possible to come interestingly close. Despite circles of confusion and the huge variety of production approaches across music genres and decades.
If that is his position, then he is wrong. As you know, I have tested hundreds of speakers where I check the audibility of deviation from flat on-axis response. I am beyond convinced that we need flat on-axis response. Medium to large variations are 100% wrong. You could argue about how flat is flat but outside of that, someone's subjective remarks without published protocol and research has zero value to me.@Arindal has strong conviction that measurements are not good enough to tell the story of a speaker, and he professes to have vast experience with empirical testing. I understand that he cannot share that data here because it is proprietary, which is unfortunate, but--and perhaps you've already done so and I've missed it--I am interested in your opinion of his heterodox (on ASR) views.