The Kii THREE’s are the first loudspeakers to come into my room where I did not feel the need to apply external Digital Signal Processing (DSP) to smooth the frequency response. As some of you know, I am big into DSP.
The frequency and timing response of these loudspeakers are just about “textbook” perfect and translate extremely well into my room with a simple tweak using the Kii Controller.
The Kii THREE’s produce a “big” color free sound that belies their size. Along with constant directivity to low frequencies, make these speakers just about perfect sound reproducers.
Let’s Talk Tech
One can see that the Kii THREE’s are unlike any other speaker design. At the front is a tweeter mated to a diffraction waveguide, along with a midrange driver, in a cabinet that resembles a computer tower, except the cabinet material is inert, weighing in at 33 lbs each. The “knuckle rap” test on the cabinet indicated we are no longer in Kansas, as does the woofers on the sides and rear of the cabinet:
Kii uses this driver arrangement, coupled with on-board DSP, to produce a cardioid, controlled directivity dispersion pattern, down to low frequencies. Kii calls it Active Wave Focusing. Simply, the DSP controls the sound dispersion, so the sound pressure is directed to the sweet spot. The DSP manipulates the phase and timing response of the side and rear drivers, relative to the front, so that they reinforce the output to the front, but cancel the output to the back, producing a cardioid dispersion pattern. Clever.
Why is this important? Speaker boundary interference response (SBIR) is responsible for huge dips in the bass response caused by the destructive interaction of the direct sound from the speaker and the reflected, indirect sound from nearby boundaries, such as front and side walls, relative to the speaker.
See this technical article and scroll down to the bookshelf speaker’s response and look at the frequency responses to understand why this is a big deal:
One can see the huge dips in low frequency response due to speaker boundary interference issues. While there are a few ways to mitigate this destructive interference (e.g. soffit mounting loudspeakers like in studio control rooms or external custom DSP that cancels these reflections), it is all but inevitable for the home audiophile, regardless of speakers or room… until now. Note in above example the bookshelf tweeter has narrowing directivity at 12 kHz. Technically, these two issues are the raison d’etre for what the Kii THREE solves and what makes them unique.
The term constant or controlled directivity is routinely used in pro sound and has been finding its way into consumer audio for some years. Earl Gedde’s wrote the preeminent white paper years ago on why constant directivity is a good thing. I have been using constant directivity waveguides in my home stereo setup for years. The issue is, to get constant directivity low enough requires large baffles and constant directivity waveguides, which is what I have now.
Floyd Toole and Sean Olive have written books and articles on research showing that smooth off axis frequency response is just as important as on-axis response for subjectively good sounding loudspeakers. The JBL M2 and Revel Salon2 are great examples of speakers with smooth on and off axis frequency response or constant directivity. The Kii THREE’s take it to the next level with constant directivity response down to below 100 Hz. Most other speakers are either not constant directivity designs or if they are, constant directivity to approximately 1 kHz. Or if a very large baffle and waveguide, maybe down to around 400 Hz, which is what my JBL Cinema speakers can do, but still have SBIR issues below 400 Hz.
If you look at AudioXpress’s excellent review of the Kii THREEs, Paul Wilke produced this polar map, ala Geddes style, in the measurement section. This is the best controlled directivity response I have seen for any loudspeaker to date:
Not only controlled at low frequencies, but also at very high frequencies, where the tweeter’s directivity typically narrows, but with Kii’s diffraction waveguide, the directivity holds virtually constant all the way to 20 kHz.
Zooming in on the polar map, we are looking at a “controlled or constant” directivity pattern where the -6 dB points are approximately ±75 degrees overall, using 0 degrees, as on-axis. The directivity narrows a bit to under ±50 degrees at 20 kHz and widening to about ±150 degrees at 80 Hz.
This is an impressive technical feat of controlled directivity down low, unheard of from a bookshelf loudspeaker. This implies that the THREE’s can be placed in virtually any room, in whatever location, free standing, or up close to one or more boundaries, and still achieve accurate sound reproduction with minimal SBIR issues.
Further, the cardioid polar pattern that these speakers throw is something I have never heard before. Whether listening on or (way) off axis, one is still hearing an accurate frequency response, especially in the last octave from 10 to 20 kHz. In subjective terms, the “air” is still there, even way off axis.
Setup, Configuration, and Calibration
I pushed my large JBL speakers to the sides and set up the Kii THREE’s on my 24” high, sand filled, Monolith stands in the same location where the JBL’s were. Which is the same location for other speakers that I have reviewed on CA. Vibrapods are used to isolate the speakers from the stands and the stands from the floor. I did not move the subs out of the way, but they were not hooked up for this evaluation.
JRiver MC 24 is the software music player, connected to my Lynx Hilo via USB and then using the AES/EBU digital output of the Hilo to the AES digital input on the Kii on the left (master) and then using an Ethernet cable to link the left speaker to the right speaker. You will notice the Kii Control on the coffee table. For critical listening, I case the guitar, put a comforter over the drum kit and move the coffee table out of the way. During the evaluation, I also used the balanced analog outputs of the Hilo to the balanced analog inputs of the THREE’s and did not notice any difference in sound quality.
As former recording/mixing engineer, I use guidelines from the ITU and EBU to set up my speakers in an equilateral triangle, with the speakers toed in, on axis, pointing directly at my ears. I also calibrate my listening level, so when I am performing critical listening, I monitor at ~83 dB SPL, C weighting, slow integration, using a calibrated sound level meter. Bob Katz’s article that I linked, provides an excellent overview of the process and why. Most recording/mixing/mastering engineers use the same equilateral triangle setup and monitor level calibration for producing the art. I use the same approach for reproducing the art.
The calibration process also includes using REW, as I measure the frequency response at the same reference level and adjust the speaker’s frequency response to a preferred target frequency response at the listening position. There is good scientific research on subjective listening tests correlating to objective measurements from Sean Olive and Floyd Toole on The Subjective and Objective Evaluation of Room Correction Products and The Measurement and Calibration of Sound Reproducing Systems respectively.
From Sean’s slide deck is a preferred subjective ranking of average magnitude responses, objectively measured at the primary listening position:
The top preference (red trace) is a flat, but tilted measured response. If 0 dB is 20 Hz, then it would be a straight line to -10 dB at 20 kHz.
Note that this tilted measured response is perceived by our ear/brain, as subjectively flat or a neutral response according to Sean’s research:
See how an objectively measured response of 20 Hz and straight line to -10 dB at 20 kHz is subjectively perceived as a neutral or flat response to our ears/brain (red trace overlaid in the above chart). Most participants in the study preferred a frequency response from 20 Hz with a straight line to -10 dB at 20 kHz. A measured “flat” in-room frequency response is not the preferred target, as it sounds too thin or lacking bass.
“The Science of Preferred Frequency Response for Headphones and Loudspeakers” goes into more detail and provides links to further studies, which show the same preferences, for both loudspeakers and headphones. Dr. Floyd Toole says, preferred is synonymous with accurate.
I have been using computer based software DSP since 2011 to custom design digital FIR correction filters in both the frequency and time domain for loudspeakers in rooms. In my own listening tests, I prefer the tilted response from 20 Hz to -10 dB @ 20 kHz. To my ears, sounds subjectively balanced or neutral from top to bottom. Whatever your preference is, this is a good place to start, as the subjective listening tests that Sean and team have performed, multiple times, with multiple participants, does correlate to a preferred in-room measured response, assuming good loudspeaker design with smooth directivity.
After taking a few measurements with REW, it was easy to make two simple adjustments on the Kii Controller to achieve my preferred frequency response.
First, I set the boundary eq for each speaker independently. The issue is due to my room set-up constraints, where the stereo is offset to the left side of the room along the long wall. The left speaker is more in the corner and the right speaker is almost centerline in the room. More boundary compensation is required on the left versus the right:
Here we see the boundary eq settings that I arrived at when individually calibrating the speakers to produce the smoothest response and matches my preferred target frequency response.
The left channel (Master) required more compensation, -6 than the right channel (Slave1), -2.
Once that was set, I moved on to the Contour and Tone controls, which are Baxandall type tone controls that allows one to set the corner frequencies for both the lows and highs and adjust the gain.
The boundary eq is so good, I did not need to make any additional low frequency adjustments.
Folks can adjust to their own preferences without the measurement gear. The Toole/Olive research shows there is a direct correlation between an in-room measurement and the preferred frequency response that subjectively sounds neutral to groups of people, myself included.
With that in mind, let’s look at the in-room measurements.
With the simple boundary eq and tone adjustments dialed in as described above, I measured the frequency response at the listening position:
That’s an excellent “in-room” frequency response using no external eq or DSP. The “ups and down” are the standing waves or room modes of my room. I have unfavorable room ratios where the only room ratio worse is a cube shaped room.
The offset of the stereo to one side of the room is why the peaks and dips aren’t the same for both channels (and why I use digital room correction). Subjectively, the bass still sounds neutral to my ears. Meaning I don’t hear much of those peaks and dips. But more importantly, the destructive Speaker Boundary Interference Response (SBIR) has been nullified, as it does not show up in the measurement.
How closely can I match the Kii THREE’s to Olive’s research on preferred (neutral) frequency response?
An excellent match. The Kii’s measure -3 dB at 21 Hz in my room with high frequency response out past 20 kHz. This is with 2 simple adjustments on the Kii Controller. Bravo! The Blue and Mauve “reference” measurements are from my much larger tri-amplified speaker system, with subs, using a custom digital filter specifically designed for my loudspeakers and room to tightly match to the Olive target. To be clear, I did not apply any external DSP to the Kii’s in this review.
Even moving the THREE’s back towards the front wall, so they are 40 centimeters away, measured from the rear woofers, have very little effect on the low frequency response:
Here we have the two sets of left and right measurements that are overlaid. One is with the THREE’s 80 cm (basically free standing) from the front wall and the other is 40 cm from the front wall. The results are pretty much identical. The Kii THREE’s boundary eq is very effective and seems impervious to any destructive boundary interference, whether placed free field or close to a boundary.
I moved the measurement mic 3 feet to the left, from the center listening position, then 3 feet to the right, from center. The speakers are back to their 80cm position from the front wall:
Impressive, as the distance across between mic positions is 6 feet, which covers both ends of my 3 seat couch. Still achieving a nice smooth envelope response with the top end predictably down in level, but the frequency response (i.e. the timbre or tone quality) is the same. Meaning the off-axis frequency response all the way up to 20 kHz is perfectly smooth and no indication of any narrowing (i.e. notches) in the frequency response at the top end, as per our example at the beginning of the article.
In the low end, we are just seeing the beginning of SBIR issues, as minor as they are, from the speakers that are furthest away from the microphone. Meaning, in my 9 foot equilateral setup, when the mic is 3 feet left of the center position, the right speaker is approximately 12 feet away from the microphone. This is getting into far field response…
What about the timing response? Here are the Kii THREE’s step response, showing that all drivers are time aligned or time coherent at the listening position:
It is the vertical step at time 0ms that we are most interested in. As can be seen, a vertical, non-discontinuous step means that all direct sound from the drivers are arriving at the measurement microphone (or ones ears), all at the same time. The ripples over time are the low frequency “peaks and dips” of the standing waves or room modes, based simply on the physical dimensions of my room.
Here is a quick refresher on what a time coherent loudspeaker’s step response looks like as compared to non-time coherent loudspeakers. Folks should understand that time alignment is not just for one mic location either.
The THREE’s has another mode of operation called “Minimum Latency” which reduces the delay in the Exact mode from 90 milliseconds down to 1 millisecond. This is useful for when watching movies or television for lip sync purposes.
There is a subtle, but audible difference to my ears, not only on the Kii’s but I ran a separate experiment switching from time aligned to non-time aligned, while keeping all other parameters the same. I prefer the time aligned response. Technically, it is the more accurate response that does not distort the phase response of the music waveform. Kii calls it, the Exact response. From the Kii THREE manual:
“In the latency menu you can choose whether you are able to run the Kii Three in their most exact setting, which offers a full phase correction but introduces roughly 90ms of latency. In an audiophile listening environment this doesn’t impose any practical problems, so the “Exact” setting is always the preferable and best sounding option. Full phase correction means, that the whole frequency spectrum radiated by the speaker does not have any unnatural phase shift, an artefact that is commonly induced by classical analogue crossovers. Clever DSP correction allows us to instead achieve a phase response that is identical to the original signal on the recording. This powerful calculation needs time, therefore latency cannot be avoided in this case.”
Completely agree, and at this price point, I expect loudspeakers to be time aligned to qualify for accurate sound reproduction.
The Kii THREE’s measures full range, with a very smooth frequency response that can easily be tailored to one’s preferred in-room response using the Kii Controller. The measurements show the THREE’s to be impervious to speaker boundary interference. The dispersion or polar response of these speakers measure excellent off axis for a very wide, but controlled horizontal coverage pattern. Finally, all drivers are time aligned to achieve a phase response that does not alter the original signal on the recording.
Technically, these loudspeaker measure as an accurate sound reproducer in both the frequency and time domains, plus have a constant directivity polar pattern that is pretty well unmatched by any speaker on the market today (except perhaps the BeoLab 90 and Dutch and Dutch 8c).
It is amazing to achieve this level of measured technical competence in a “bookshelf” loudspeaker.
Do the Kii THREE’s sound as good as they measure? ?
Subjective Listening Results
I have posted this before, as a small sampling of music I listen to for testing gear. Part of it is that I have listened to these tracks hundreds of times over the years with many equipment and know the recordings extremely well. I also appreciate the high dynamic range of these recordings which gives a much more lifelike music reproduction to my ears:
I got in around 80 hours of listening over a couple of weeks, as I work mostly from home. These speakers sound considerably bigger than their size suggests. I normally listen to (very) large floor standers. The THREE’s sound full range with deep bass extension. Very surprising from a speaker of this size.
The bass response, is not only deep, but uncolored, and did not in any way sound disconnected from the mids or highs, like I find in some other speakers. I attribute this to several aspects of the Kii’s unique design where the speaker boundary interference is being nullified, the smooth polar response down low, and the bass drivers time aligned with the mids and highs. This results in very tight bass response, even down low, and can play effortlessly all day long at my 83 dB SPL critical listening reference level without ever sounding uncontrolled. Turning up the level to be as twice as loud at 95 dB SPL, the bass reproduction never lost its composure.
The THREE’s can’t match my large double 15” woofers per side bass cabs for impact with 12 Hz response from two 12” sealed Rythmik subs. But at critical listening levels, the Kii’s sound competent and uncolored. Not really a fair comparison, as the JBL bass cabs are huge. However, I don’t know of any other relative sized bookshelf loudspeaker that can play down this low (in my room).
Listening to SRV’s Tin Pan Alley (DR18) with the volume turned way up feels like one is at the hall during the live off the floor recording. It is easy to get lost in the envelopment. With the volume still way up, the buildup crescendo at the beginning of Money for Nothing is enough to break out laughing at the goosebumps. Spontaneous dancing occurs when Thelma’s, Don’t Leave Me this Way starts cooking during the chorus with the backup singers. At 2 mins 44 sec in, the tambourine player sounds right in the room.
Marilyn Mazur’s, Bell Painting (DR25 on CD!) attack, clarity and decay of the percussion instruments sounds realistic to my ears, as does LessLoss individual drum presentation (DR21). The bells sound so clear and pure.
Stewart Copeland’s drums in Murder by Numbers (DR18) crack and punch like a drum kit should, especially during the outro where the drums are turned up in level in the mix. Every little intricate hi-hat showmanship shines through with crystal clear clarity. Even when listening well off axis. For example, walking behind my couch as Stewart’s hi-hat is playing, I could not hear any high frequency response imbalances. One can listen for directivity “dead spots” where the hi-hat is playing at a certain level, move your head a few inches and the hi-hat level goes down, then move a few more inches and its back in level. The so called “head locked in vice listening position.” Not so with the THREE’s, they sound super smooth with Stewart’s hi-hat tricks sounding equal volume, as I walked by behind the couch or sitting anywhere on the couch for that matter. Good show!
Listening to Patricia Barber’s, Regular Pleasures, I listen to her voice closely while the bass slides and the drum hits for any hint of her voice modulating on the bass. To my ears, her voice sounds perfectly clear, as does the bass with no hint of modulation.
I put on Yotto’s Radiance, which has significant with clipped overs and over compressed bass amplitude, still sounds deep and clean on the THREE’s, even with the cones visibly vibrating.
I have to admit, as much as I like to point out flaws with loudspeakers, as the weakest link in the sound reproduction chain, I don’t hear any real flaws with the THREE’s. Sure, if you want to go louder, Kii has a BXT woofer assembly to extend the THREE’s bass response and increase the sound pressure level. But the low bass is there in the THREE’s and sounds remarkably uncolored to my ears.
I hope folks get an opportunity to hear the Kii THREE’s, as an excellent example of what accurate sound reproduction “sounds” like. If you are in Vancouver, head over to Liquid Sound for a listen. This speaker measures as good as it sounds and vice versa. Smooth frequency response from 21 Hz to 20 kHz with time aligned drivers, means the music arriving at one's ears matches as closely as possible to the content on the recording.
The ability for the THREE’s to be placed near or far from a boundary, without speaker boundary interference, is next level loudspeaker engineering. As mentioned before, the only other way to achieve this is soffit mounting speakers or external digital room correction software products. For some folks, hearing these speakers without speaker boundary interference may be a first, and an ear opening listening experience.
Special attention to the top octave (10 to 20 kHz) with Kii’s unique diffraction waveguide, raises the bar for controlled, high frequency dispersion. I have yet to see any other loudspeaker measurement that shows as wide as dispersion in the top octave as the Kii’s. Even speakers with notable controlled directivity, always show a narrowing of the directivity in the top octave.
The THREE’s sound full range, smooth, uncolored, time coherent, and can be listened to over a wide angle without any loss in sound reproduction accuracy. An amazing technical feat for a bookshelf sized loudspeaker.
Thanks Bryan Taylor from The Gramophone in Edmonton, Canada for loaning me the Kii THREE’s for a couple of weeks. Thanks Bryan!
Thanks Paul Wilke for supplying the AudioXpress Polar Map. Much appreciated Paul!
Coming up next on the stands for review…
I wrote this book to provide the audio enthusiast with an easy-to-follow step-by-step guide for designing a custom digital filter that corrects the frequency and timing response of your loudspeakers in your listening environment, so that the music arriving at your ears matches as closely as possible to the content on the recording. Accurate Sound Reproduction using DSP. Click on Look Inside to review the table of contents and read the first few chapters for free.
Mitch “Mitchco” Barnett.
I love music and audio. I grew up with music around me, as my mom was a piano player (swing) and my dad was an audiophile (jazz). My hobby is building speakers, amps, preamps, etc., and I still DIY today.
I mixed live sound for a variety of bands, which led to an opportunity to work full-time in a 24-track recording studio. Over 10 years, I recorded, mixed, and sometimes produced over 30 albums, plus numerous audio for video post productions, in several recording studios in Western Canada.