MEASUREMENTS: Emotiva Airmotiv B1+ bookshelf speakers. A comment about Andrew Robinson's Fluance Reference XL8S video review and "Do measurements ultimately matter?".

Emotiva Airmotiv B1+: Comes with small manual, and some stick-on rubber footers in the plastic bag. Fabric covers in background.

Since mid-2020, I've been on a quest to find some replacement speakers for my computer desktop. Along the path, I've listened to and measured speakers like my old AudioEngine A2, the active Edifier S2000 MkIII, borrowed the Tannoy REVEAL 501a, the passive Fluance Reference XL8S, and of course tried the KEF LS50.

For this post, let's look at the Emotiva Airmotiv B1+ 2-way bookshelf speakers (US$229/pair), released in early 2020. Externally, these look like their predecessor, the Airmotiv B1 that first came out in 2016. I bought this pair direct from Emotiva in late 2020, arriving a little before Christmas.

They came well packaged with plenty of foam, thick box, and they came inside some black fabric covers for protection.

As you can see in the first image, these speakers are 2-way affairs with 32mm folded ribbon tweeter in a wave guide and 5.25" woven 'fabric' mid/woofer. I noticed that the older B1 was said to have Kevlar woofers, not sure if this is the same material with the B1+. Mid/woofer surrounds spec'ed as SBR (styrene-butadiene rubber).

Check out the mechanism of how AMT "folded-ribbon" drivers work through the application of the Lorentz effect to move air. Potential benefits of the folded ribbon tweeter include more controlled dispersion (less ceiling and floor bounce) and low mass so it can respond to the signals faster, improving transient detail.

The crossover is specified at 2kHz, and the 25mm front panel is made of HDF (high density fiberboard) which should add to the stiffness of the enclosure with 15mm thick sides and rear panel. The enclosure is described as internally braced as well. When you hold it in hand, instead of wood veneer, the outer material is a softer vinyl covering which is actually quite nice and I can't quite convey this in the pictures. It's a matte black finish and the vinyl does not pick up fingerprints easily which is also nice.

Here's the rear, notice the round, flared, 2" port. Standard 5-way speaker posts, recessed and angled so banana plugs don't stick out much beyond the rear.

Side view: if you zoom in, you can make out the textured matte vinyl covering and smooth milled HDF front panel which is angled to reduce edge diffraction. Notice that there are no holes for the grilles; the grilles are held in place with judiciously positioned magnets that are attracted to the metal around the drivers, nice.

Grille: Magnets behind as seen in right image.

The appearance of these speakers clearly isn't that of a "classic" box like the old-school charm of the Spendor SA1! With an all-black appearance (except for the metal-etched polyethylene AMT tweeter), numerous angled edges, there's a sense of austerity to it. I have heard folks call this "Darth Vader's Speakers". I actually like that description. ;-)

Unlike Darth Vader, these are not big speakers measuring only 10.75" x 7.13" x 8.25" which is a great size for my desktop, weighing about 9lbs each.

5 year warranty offered by the company.

I. MEASUREMENTS

I received the speakers more than 2 months ago and I have been using them daily as my computer desktop speakers since then. I don't know how many hours of playback I've had on these speakers; I don't count hours, I just listen. Can't say I noticed much change in sound quality over the weeks and I've probably used them enough to have them "broken in". :-)

Okay then, here's the left speaker on the stand ready to "spin":

As usual with my "qSpin" procedure, I'll be powering the speaker with my Emotiva XPA-1L monoblock, measuring with the little miniDSP UMIK-1. One change is that I'll be using my ASUS Xonar Essence One as the DAC for signal generation (convenient since it's also a preamp which saves me from using the passive Nobsound); I have a bunch of DACs around here and resolution would be way beyond speaker, microphone, and amplifier distortion levels regardless of which I use.

Here's the speaker impedance graph with phase and EPDR:

Port tuning frequency is just below 60Hz. Nominal impedance is "8Ω" according to Emotiva. The minimum measured impedance is 3.8Ω. The EPDR (Equivalent Peak Dissipation Resistance) accounting for both impedance and phase shows that this speaker can torture amplifiers with portions down to 2.2Ω-equivalent with minima around 20Hz and 130-170Hz. I would recommend that your amplifier should be able to handle 4Ω speakers. Note that the EPDR result is comparable to the KEF LS50.

Notice in the impedance graph that there are three "wrinkles" to be aware of that might reflect resonance frequencies at 180Hz, 425Hz, and 650Hz.

Let's have a look at sensitivity using a 2.83V signal sweep at 1m:

Notice how the frequency response pretty much stays within the +/-3dB guides.

REW's average tool calculates a final result of 87.3dB/2.83V/m using the gated response (pseudoanechoic) within 500Hz to 7kHz I use which is within the expected value and in fact a little higher than the official "86dB".

Now, let's get down to the heart of the speaker tonality graphs, let's look at the "qSpin" CTA-2034A-inspired tracings:

That's quite nice. Notice that there's no bass "bump" here to accentuate the limited low-end which we can see in some small speakers. Remember that I've standardized on a 30dB Y-axis range and 25dB/dec aspect ratio for all these graphs so this can accentuate the bumps and valleys a bit compared to other published work but consistent throughout my speaker measurements.

The black "Listening Window" curve consists of nearfield data blended with the gated response from 350-500Hz. Baffle diffraction and step compensation have been applied. It shows that overall, the Airmotiv B1+ maintains a relatively flat frequency response within the listening window with mild +2.5dB accentuation from 1-1.7kHz and a slight rise in the upper treble above 13kHz.

Likewise, the directivity indices show relatively stable dispersion until 9kHz after which the sound narrows and "beams" into 20kHz. Except for small bumps like at 6kHz, no major anomalies otherwise with directivity. You can see this directivity pattern in the polar maps:

On the horizontal axis, it's good to listen at +/- 30° to not roll off the higher end content above 10kHz. The vertical axis dispersion is tighter than most other small speakers I've tried (reduced table, ceiling, possibly floor reflections). It's best to stay between the +10° and -15° range for listening.

Let's have a look at the time domain:

Step response shows us that the tweeter and woofer are arranged in positive acoustic polarity. Acoustic axis is probably just slightly below the mid AMT tweeter level based on the slight kink between the tweeter and mid/woofer responses.

The Cumulative Spectral Decay waterfall overall looks OK. As usual, I measure with an amplitude ~80dB SPL and offer a 25dB range below. Remember that we're looking at a waterfall plot over just 3ms. Remember these parameters when comparing the results and thinking about audibility.

The initial waterfall decline looks good with steep fall-off although there does appear to be a little more low-level energy present than some other speakers like the KEF LS50 or Spendor SA1 but better than the Radio Shack Minimus 7 and Fluance Reference XL8S.

Notice that the steep fall-off in the waterfall is best with the higher frequencies above 2kHz which correlates with the tweeter; a finding we probably expect from the AMT driver.

Using a 200ms multitone burst, the linearity is excellent with a tiny range of +0.4 to -0.2dB between 70 to 100dB SPL with 0dB at 85dB SPL. Basically this is as good as any other small speaker measured and better than the inexpensive Fluance Reference XL8S. 

Moving on to distortion amount, here is the THD vs. Frequency Sweep with -40dB (1%) distortion line shown, measured with average ~85dB SPL volume level:

This is good with essentially all frequencies above 200Hz below 1%. Notice that at the crossover frequency ~2kHz, there is a rise to about 1%. As for the stuff below 200Hz, this is not an anechoic measurement so it is affected by the room and frequency response falls off below 100Hz as well thus should be disregarded.

Using dual tones at 300 and 1300Hz, we can look at intermodulation distortion (IMD):

Again, not bad at all. Every measurement from 60dB to 90dB SPL is lower than 1% (-40dB) THD. I like in particular how at 90dB SPL, the IMD is still <0.2%. Compare this to the Fluance Reference XL8S or Radio Shack/Realistic Minimus 7 to see speakers that cannot handle the extra amplitude and will distort. Likewise, an inexpensive active speaker like the Tannoy REVEAL 501a when pushed to this higher amplitude will demonstrate increasing distortion as well suggesting strain on the internal amps.

Finally, the Emotiva Airmotiv B1+ came as a pair, let's look at the matching of the 2 speakers:

Nice! Less than 1dB variation between the 2 speakers across the audible frequency. This speaks well to the manufacturing quality/consistency.

Likewise the Cumulative Spectral Decay looks equivalent for both speakers. And finally:

Again, a very consistent match between the pair of speakers for impedance. I mentioned above some kinks I see in the impedance graph like at 425Hz, I see this is present with both speakers. Notice the left speaker also has a little bump below 30Hz, not present in the right speaker.

So did the speaker grille affect the sound?

Yup. Looks like with the speaker grille on, measured at mid tweeter axis, a 5dB dip of moderate width centered around 7kHz has been introduced! The result was consistent between the 2 speakers. I didn't check if the effect might be lower off-axis. It reminds me of the foam ring tweak put around the Spendor SA1 and it's quite noticeable if you leave the grille on on one side only and play some music with high frequency content to hear the imbalance.

Anyhow, listen for yourself and decide. I would obviously suggest taking off the grille for best sound quality.

II. SUBJECTIVE

Here they are on my computer workstation powered by the small SMSL SA300 Class D amplifier and fed by the Topping DX3 Pro DAC which is what I base most of my listening impressions/experience on:

The Archimago's Musings Command Centre. :-) Notice that I'm using my old AudioEngine DS1 desktop risers (DS2 might be more stable) to angle them up towards the ears. Sounds better this way.

Subjectively I really like how these sound. One of the first things I noticed was that I could pump up an album like the John Wick: Chapter 3 - Parabellum soundtrack to high levels and even though the music itself is highly compressed as a DR6 album, the distortion didn't get out of hand. This is a different experience from the old Realistic Minimus 7's where clearly they were unable to handle volume with grace (yes, they go loud - but not cleanly!). Check out the tracks "Winter At The Continental" and the ending credit "Really Pissed Off" for the driving pace and rhythm. Like him or not, Keanu Reeves stars in some cool movies...

Maybe it has to do with the AMT tweeter. There's an effortlessness to string classical music - for example Janine Jansen's Vivaldi: The Four Seasons (2005) sounded beautiful. I particularly loved the harpsicord part played by her dad Jan Jansen.

Something that struck me with these was the soundstage coherence. Benjamin Clementine on At Least for Now was clearly front-and-center and an "oldie" like Ella Fitzgerald on Pure Ella sounded gorgeous with the presence of an "embodied" voice standing a few feet in front in the nearfield listening configuration above. 

The recently released (2019) extended remixes from A Flock of Seagulls - Inflight: Extended Essentials is a fun album if you like pop. It's chimeric music that brings back memories of the '80s with a modern take using the synth production qualities of today. Of course "I Ran" is an essential part of any '80s fan's collection.

On balance, these sound neutral to me (take those grilles off!) and I think would make for great home-studio monitors for acoustic recording work. I was able to work on some home piano recordings for example and was impressed by how well these little speakers handled the attack and decay of the instrument. For the size, don't expect full-range bass of course. The track "Moonlight on Spring River" (The Dali CD, Vol. 3) highlighted both the fantastic speed and effortlessness of the tweeters but also the limitation of low bass performance with only hints of the depths on this recording which are impossible physically for the small 5.25" woofers to reproduce with any authority.

III. SUMMARY

BTW, in this age of Zoom/Skype/MS Teams teleconference calls, the inexpensive USB Maono AU-902 microphone (~US$60) works really well, trouble-free, with good voice quality.

Let's just say that my search is over for a small pair of speakers to use on the computer desktop. The Emotiva Airmotiv B1+'s are simply fantastic for this job! The size is convenient for me. My wife certainly appreciates that they're not deep and will not take up the shared table space when she works on the other side. I like the matte vinyl finish that doesn't pick up fingerprints and is easy to wipe/dust off. The black, "Darth Vader" look might not be for everyone but they look good IMO in real life gracing each side of my monitor on the computer workstation. 

As an objective comparison, we can look at Average Joe Audiophile's Emotiva B1 Review from 2017 (remember, the B1 is the previous version of the updated B1+). Not bad, certainly the measured data looks quite similar with relatively flat characteristics from 200Hz onward. There's similar bass roll-off below 100Hz and areas of slight accentuation between 1-2kHz and from 9kHz up as I showed here with the B1+.

Of no surprise is that the bass could be reinforced a bit maybe with closer wall placement if possible. Remember that ported speakers can extend the bass output a bit lower but the roll-off is steeper than a closed box like the Spendor SA1 (-24dB/octave ported, -12dB/octave sealed below resonant frequency). Based on the data, I see that these Emotiva B1+ speakers have about -3dB at 80Hz and -6dB by 60Hz. One could use a small subwoofer like the Polk PSW111 crossed over at 80Hz to extend that bass down to frequencies in the 30Hz region.

On my desktop, I will routinely use parametric EQ in Roon for music with these speakers with grilles on for protection (kids hanging around):

Notice that I'm using conservative +3dB gains here and have activated the "Headroom management" option to make sure the signal doesn't significantly clip. I add a bit more punch down below using the "low shelf" filter and a "peak" around 7kHz to compensate for the grille dip.

[As an aside, I find that when I'm doing ReplayGain volume normalization already with a target of -14LUFS, a slight -1.5dB headroom management amount typically will take care of all but the loudest of music with the above PEQ settings.]

I think we truly are living in a "golden age" of low-priced, high performance gear whether it be excellent speakers or electronics. Who knows if the future will remain so "golden".

While I tend to have a different viewpoint on "High End" audio than Steve Guttenberg and the "old-guard" subjective reviewer cohort in many ways, I agree with him that these are excellent speakers as expressed in his video.

--------------------

On the other hand, I am a bit disappointed by Andrew Robinson for his very positive review of the Fluance Reference XL8S as "CRAZY GOOD" which I saw only recently. While I certainly appreciate the very low cost of the speakers, I think it's worth discussing differences of opinion around sound quality; this and his opinions on specs/measurements.

Although I speak up once awhile, typically, I just leave reviews like these be; there are just too many out there to comment on anyway. But when a superlative like "CRAZY GOOD" in CAPS is used, that implies something special and caught my attention. Other than being inexpensive, these Fluance speakers certainly didn't seem that special as much as I like that Fluance is a Canadian company (as a Canadian myself).

Compared to the Emotiva Airmotiv B1+, the Fluance Reference XL8S (yes, about US$60 less expensive), was clearly a step down in terms of objective and subjective performance IMO. Contrary to what some might claim, many of the objective findings are audible especially for speakers; we're not talking about ruler-flat frequency responses and <<0.1% distortions like with decent DACs and good amplifiers!

After watching his Fluance video, YouTube suggested that I watch Robinson's other recent video - "DO SPECS REALLY MATTER in Audio? - Understanding Speaker Measurements!" Oh boy. Andrew Robinson appears to be getting way too opinionated here with as far as I can tell little rationale built upon experience with correlating measurements and what he might/might not hear.

First, he does have good points that published "specs" might be unreliable and manufacturers can use different measurement techniques (eg. whether they have an anechoic chamber or not). A specification can be correct but incomplete. For example, "8Ω" impedance is a common specification, but incomplete in the context of a full impedance/phase/EPDR graph. More complete measurements might show us that in fact the "8Ω" speaker has impedance dipping down to 2Ω in important frequencies and this will have real-life consequences with certain amplifiers. Likewise, manufacturers exaggerate figures and will publish very poor resolution graphs like frequency response using 1/3-octave smoothing and 100dB Y-axis to make it look flatter. This speaks to the lack of standardization and that too is another major topic. As such, I think it's a bit inappropriate to use the words "specs" and "measurements" in the title of his video as if they're all the same without qualifications.

However, to say at around 10:35 in the "DO SPECS REALLY MATTER" video:

"Do measurements ultimately matter? Well, for me not really. Because if I don't enjoy myself, all of those fancy numbers and measurements add up to a whole lot of nothing. But that's just me, that's just how I feel."

Is a rather sad admission for a reviewer of engineered devices in a science-based hobby. Regardless of whether a statement is profoundly true or absolute rubbish, a lazy "that's how I feel" can apply just the same!

While not all objective measurements might be important or particularly audible especially with DACs these days, assuming the results are accurate, at least there is an attempt at expressing a truth about the physical properties of the device which others can independently verify. Subjective opinions can just as well "add up to a whole lot of nothing" because if all it is is "just how I feel", then there might not even be any truth in the opinion whatsoever. It's much easier to make misleading subjective claims also.

One could I suppose argue that watching videos like these really isn't about "truth" at all. I don't think it's inappropriate if someone were to label such videos and articles as "for entertainment only". In fact, I think Robinson was quite honest in a previous video describing the "infotainment" nature of such productions rather than necessarily an attempt at "truth" (notice that he stumbles over this word)!

It's fine to tell stories about how one found one's life partner, express an anecdote in life, or express preferences for artwork/music because these are genuinely, subjectively, personal. Nobody will argue with this usually. But isn't subjective-only opinion-making a bit empty when we're trying to convey value to others for a tech product that's supposedly reproducing high quality sound with physical characteristics that can be expressed quantitatively? Is it good enough that reviews of hi-fi products be such self-indulgent affairs as to reflect just "how I feel"? IMO, "golden ear" audiophiles have gone down this path over the many decades resulting in disappointments (for many believing that ridiculous money bought ridiculously good sound quality), ridicule by the public (the negative connotation for being an "audiophile" / "audiophool"), and unscrupulous companies peddling pseudoscience snake oil products. Whether it's megabuck electronics, krazy kables, quantum doodads, or even budget speakers, the limitations of a purely-subjective mindset are the same.

For Robinson to express that "measurements do not matter" suggests that he has not experienced how subjective and objective characteristics are inextricably linked. Like two sides to the same coin that look and "feel" different, but each just as important in a product's evaluation. To wit, I think the Fluance review is a good example of this as I have quite different opinions on those speakers.

To start, the Fluance XL8S reviewed here was able to convey detail but it's at best merely OK, the kind of "detail, detail, detail" (4:50) Robinson declared with fanfare IMO is untrue and I believe the reasons can be found in the review which I won't expand on here. I noted that the soundstage was not as precise with these small speakers which may be a reflection of the interspeaker variability. Furthermore, and more importantly, they were uneven sounding with dips and accentuations of bass and upper treble. I'm not just saying this subjectively because these limitations are audible "to me" before running tests, but it's evident in the measurements also. For reference, compare this to the Emotiva results above:

While my "qSpin" result is less resolving than a proper anechoic chamber, meticulous outdoor rigs or $100K measurement gear, for illustrative purposes as a hobbyist however, I'm pleased with what it tells me. I see measurements as being able to enhance in detail an understanding of what I'm hearing and helps me consider their relative value.

Although these speakers could be exciting when listening to modern pop and EDM, they were "dark" sounding, a bit unrefined with high dynamic range acoustic recordings (notice increased distortion measured at higher amplitude), and lacked upper-mid sweetness when playing clean vocals. This was why I thought some listeners who were not keen to a more neutral sound could still like them. However, because of those limitations, for music lovers who cared about "good", natural sound (not just audiophiles!), the Fluance was not good enough and why I thought they were more suitable for secondary systems, and surround use (which is what Fluance markets them as). IMO, discerning listeners looking for a front pair of speakers should look elsewhere even within the low budget category (again, the Emotiva is only about US$60 more for example). They did not stay on my computer desktop for very long due to these deficiencies.

As humans, tools have allowed us to reach further and deeper in our understanding in all kinds of ways. Measurement tools expand our appreciation for what a device does and allows us to understand ourselves as well. "Am I a music lover who wants neutral sound with flatter frequency response or do I prefer 'euphonic' coloration?" is a reasonable question that measurements can help us answer personally.

Of course each of us will have our own preferences and I'm not doubting that the Fluance speakers sounded good for Andrew Robinson's home set-up and through his ears. If he's happy with these speakers, that's great but I would take issue just the same with declaring them to be "CRAZY GOOD" as untrue! All I can surmise from how he reviewed the speakers is that he likes this kind of coloration in his sound, in that room, with the music he listened to.

It's easy to find countless strong opinions about almost anything. Some folks might even be easily offended thinking that their opinions somehow must be "respected" or that "my truth" is somehow equivalent to "the truth". Robinson's opinion that "Because if I don't enjoy myself, all of those fancy numbers and measurements add up to a whole lot of nothing" to me is a rather superficial one wrapped up in a "nice guy" kind of persona which ends up intellectually empty. I don't think that's cool. Before dismissing something as "a whole lot of nothing", I would respectfully suggest all of us to search for a higher level of understanding, a more complete picture with exploration and finding "truth" that transcends personal opinions - not just in audio, but life in general. Even if he doesn't enjoy himself with a piece of equipment, might he and the viewers not be interested in why that might be so? Might the "numbers and measurements", if he had any, not inform him of why a piece of gear is lacking so as to learn and avoid such characteristics in the future? If "numbers and measurements" are potentially able to educate him and his viewers, then that in itself is far from being "nothing".

I would have thought that as a "recovering audiophile" looking to kick the "audio addiction" of traditional audiophilia nervosa, he would have achieved a broader perspective. What he's saying in that video is actually not all that different than every other subjective-only audiophile reviewer over the decades trying to justify why they don't have measurements to show. Thankfully, we are just talking about the audio hobby here. I hope that after all these years of living in a modern technologically sophisticated society, the idea that objective testing can get us closer to "truths", achieve understanding, confirm value, and perhaps even find wisdom to spread to others along the journey (as rational audiophiles of course), should be obvious already.

As a reviewer, it's essential to listen for oneself but better yet to make efforts to verify with objective data before rendering strong opinions ("CRAZY GOOD"? "STUPID GOOD"?). Often, such strong subjectively-derived opinions are incomplete and even misleading to viewers/readers more profoundly than an incomplete 'spec'.

Yes. Of course measurements ultimately matter. Opinions matter as well. Both need to be expressed in context. And it's never either/or.

[I see Gene at Audioholics has a rebuttal video as well.]

--------------------

Almost March, dear audiophiles! I wish you well as we look forward to Spring in the Northern Hemisphere.

Happy listening.

Addendum:
As per the comments, here are versions of the Emotiva and Fluance "qSpins" with 50dB vertical range and 50dB/dec aspect ratio for both the main frequency response and directivity indices:

Do these look more comparable/"standard" with respect to NRC results?