Kii THREE Review

[CtheArgie]

Was it called ‘New Music’ by any chance?

Yes! New Music, near the ponds. We were there last week. Back in CA now. Too cold and wet for my bones! But the beer, oh my god!

 

[richard12511]

People With that sort of money can build a room so big that the precedence effect doesn’t even happen. Let that sink in.

I can afford the Kii Three, but I can't afford a giant well treated room, at least not without taking on debt. Based on some quotes I got a few years ago when looking into building a dedicated listening room, the latter is like 10-15 times the price of the Kii.

 

[richard12511]

It’s not similar, the 8C is mostly 3rd harmonic while the Kii is second.

That's a good point. It's also got quite a bit more distortion around 1kHz.

That said, I tend to think distortion matters very little. What matters more is how the speakers compress(change the FR) as the volume is turned up. Will be curious to see how much the D&D's bigger cabinet helps it here.

 

[Massimo]

That's a good point. It's also got quite a bit more distortion around 1kHz.

That said, I tend to think distortion matters very little. What matters more is how the speakers compress(change the FR) as the volume is turned up. Will be curious to see how much the D&D's bigger cabinet helps it here.

Erin's perspective on distortion from his review of the D&D 8C:

"Speaking of distortion, yes, the measured harmonic distortion is relatively high. But, in my listening tests I had this speaker playing as high as 105dB @ 4 meters with no audible distortion other than the things that were buzzing in my living room. I even used test tones to see if I could hear the distortion in my MLP but, alas, the fireplace was vibrating as were some of the items in the bookshelves. Not to mention room modes. I’m not saying that someone else couldn’t hear the distortion if trying. But in my case, in a real living room with other things about, the distortion measured didn’t correlate to distortion heard if for no other reason than it was masked by all the things that make a real room a room. Aside from THD/HD profiles, the real “selling point” was the lack of mechanical issues. I have had numerous speakers in my room that couldn’t get past 90 or 95dB without bottoming out, flopping around, or otherwise emitting audible issues. These issues are much, much more audible than distortion. And in all of those cases, those issues still remained even when a HPF was implemented to help diminish those things. Yet, at >100dB, the DD8c speaker had no such issues while playing down to 20Hz. I can’t name another speaker I have tested thus far that has done this. Not even the venerable F226Be which rolled off at around 50Hz. I’m not making excuses. I have no reason to. Truthfully, I was surprised by the relatively high level of HD components. But, in my real-world listening they didn’t manifest in to issues. YMMV."

 

[pozz]

I spent some time looking at Erin's in room vs. PIR measurements.

I compiled a bunch more but realized, after looking at 10 or so and wondering if I should scrape/import the data, that comparing them isn't very useful for understanding the contribution of rear radiation, or lack of it, to SBIR. Erin uses MMM measurements, which show pretty well what we can expect in terms of overall curve but not this specific aspect of it.

I don't know if a single point measurement at MLP will really show it. Maybe many fixed points around the room might.

Edit: Accidentally truncated the X axis in the original measurement.

 

[hardisj]

I spent some time looking at Erin's in room vs. PIR measurements.


I compiled a bunch more but realized, after looking at 10 or so and wondering if I should scrape/import the data, that comparing them isn't very useful for understanding the contribution of rear radiation, or lack of it, to SBIR. Erin uses MMM measurements, which show pretty well what we can expect in terms of overall curve but not this specific aspect of it.

I don't know if a single point measurement at MLP will really show it. Maybe many fixed points around the room might.

Edit: Accidentally truncated the X axis in the original measurement.

I wouldn't waste my time. I can't guarantee they were all taken in the exact same manner. I know I was using the UMIK-X for the DD8c (which caused a bump >10kHz). Nevermind the position of the DD8c & Kii are close to the front wall while the other speakers are out in the room. The only real reason I provide the PIR vs MIR is to solidify the notion of usefulness of the PIR in that above the transition frequency you can count on the PIR to give you an incredibly accurate idea of the in-situ response for most setups.

 

[hardisj]

Would it be possible to have the group delay graph, for both in Phase Exact mode and in Minimum Latency mode? it would be interesting to see this. Thanks!

I can for phase exact but not minimum latency (unless you want quasi-anechoic which will be useless below a few hundred hertz)

I just updated my review with the GD. I'll save you a trip, though:

 

[hardisj]

I also updated the cardioid discussion section with this polar which shows better the transition from omni to cardioid which - depending on how you define it - happens somewhere between 70-120Hz.

 

[hardisj]

... and that's it. I've got to move on to other reviews. I've spent a LOT of time on this one and I've got a pile of stuff mounting up in the garage.

 

[edhamgtr]

Kii Three:
Preference Score is 7.7 and would be 8.3 with a perfect subwoofer.
Preference Score is 8.1 with an EQ and would be 8.7 with a perfect subwoofer and the same EQ.
Bass extension (LFX): 20Hz at -6dB

Dutch, Dutch 8C:
Preference Score is 7.4 and would be 8.0 with a perfect subwoofer.
Preference Score is 7.8 with an EQ and would be 8.5 with a perfect subwoofer and the same EQ.
Bass extension (LFX): 20Hz at -6dB

FWIW - I had the Kii3 and 8C in my room, at the same time for about two weeks and agree with this assessment.

When comparing recordings that were done in rooms that I know well, The Kii's sounded like those rooms while the 8C's had just a bit of coloration to them. Just an unnatural midrange quality that made very different rooms sound much more alike than they are.

Mix translation also went to the Kii's.

We are splitting hairs here for sure, but the 8C's deliver just that extra bit of ear candy that for most listeners would have them win a side by side shootout - which in turn is exactly why I didn't keep them as for both my work, and recreational listening - the Kii's were just more accurate to the source.

Kii's have remained and I'm only visiting speaker forums as I'm considering an Atmos mixing room build and simply can't afford 13 of them.

also fwiw - I had the 8351A's in at the same time. My ears agree with the measurements found in this forum but I just didn't like the tweeter. Some amount of edginess to them that I couldn't live with. They also fell short to the Kii's as far as recordings showing an accurate room tone (for recordings in rooms I've spent a lot of time in).
When you've spend days in a recording space, you get an imprint of that sound. Kii's make my ears feel like they are in that exact room in a way that only my ATC150's could do. I just didn't have space in my new room for my 150's which led to a long speaker testing process that the Kii's won.

my 2cents - which my wife assures me isn't quite worth even 2 cents.

 

[bigjacko]

We don't have the same data on the 8C, do we? So we cannot make any conclusion from this.

Oh I did not notice we don't have that for 8c, makes things hard to compare.......

 

[thewas]

I spent some time looking at Erin's in room vs. PIR measurements.
View attachment 169517

I compiled a bunch more but realized, after looking at 10 or so and wondering if I should scrape/import the data, that comparing them isn't very useful for understanding the contribution of rear radiation, or lack of it, to SBIR. Erin uses MMM measurements, which show pretty well what we can expect in terms of overall curve but not this specific aspect of it.

I don't know if a single point measurement at MLP will really show it. Maybe many fixed points around the room might.

Edit: Accidentally truncated the X axis in the original measurement.

The Kii Three seem to have the smoothest FR/tonality at the LP which is also very close to the 0.8 dB/oct curve which often sounds fine for my setups too:

My current much lower budget listening setup with only (negative) PEQs below 500 Hz:

 

[fluid]

There is another solution called soft-fit; speakers installed into a false cavity; where the baffle becomes part of the new wall.

The pedant in me can't help but clarify that it is soffit mounting and not soft-fit, which sounds more like an ear plug

 

[McFly]

I think we need to start looking at vertical directivity more. I mean these speakers are for all intents and purposes identically smooth in response, albeit one slightly wider horizontal radiation pattern that the other, but the main difference I can see is the mid to tweeter XO point, the kii at 2khz and the DD8C at 1500(?)hz. This causes a dip in sound power and this would be the main difference one would hear "around the midrange"

Might also explain why the well measured R3's don't wow people as much as the measurements would suggest. The more I measure conventional 2-way 2 driver speakers the more I find myself enjoying the LR4 xo's above 2khz and I seem to loose intelligibility and focus with 2 ways that are LR4 crossed below 2khz. The horizontal measures are perfect, so I can only speculate on the vertical dispersion. I had R3's but sold them on, I found them too forward around that upper mid region no matter what I did.

 

[abdo123]

Erin uses MMM measurements, which show pretty well what we can expect in terms of overall curve but not this specific aspect of it.

Actually i would argue that since the listener (microphone) keeps moving, whatever response anomalies we see are primarily SBIR and room modes, since the location of the speaker does not move in MMM. With a single point measurement the location of the microphone will polute the data.

 

[abdo123]

Yes! New Music, near the ponds. We were there last week. Back in CA now. Too cold and wet for my bones! But the beer, oh my god!

Glad you enjoyed it!

 

[FrantzM]

I don't think my post was clear, it was a quarter of a million over my existing property value I meant, on it's own I'm not sure if would get anything around where I live, same as around your way, and lots of other places.

There are a lot of people who could afford these easily, cash rich and space poor, how many of them want good sound quality is the question.

I can see the discussion veering ever slowly on a class warfare issue.
A bit of perspective. A $15000 car for many (most?) people on this forum would be a reasonable outlay of cash. That’ll get you a decent, likely used car in the US. I would count me as one of the people that would prefer spending that amount on a pair of items that provide me much greater pleasure. I am not alone.
Life is a matter of compromises and priorities. At this point in time, to get to that level of quality, it seems that there are very few, or no, less costly alternatives. I wouldn’t mind, emphasis on “I”. In fact, waiting for more review s, as there are indeed, other speakers with similar prices and girth, vying for the top spot.

Peace

 

[BrokenEnglishGuy]

This is a good speaker
Sadly, their price not
Let me know when we see this kind of speaker for 1/4 of this price.
For this money I prefer to buy a random super nice speaker + mic and find the correct placement for the speaker in the room
Always works without the cardiod thing.

 

[TimVG]

I think we need to start looking at vertical directivity more. I mean these speakers are for all intents and purposes identically smooth in response, albeit one slightly wider horizontal radiation pattern that the other, but the main difference I can see is the mid to tweeter XO point, the kii at 2khz and the DD8C at 1500(?)hz. This causes a dip in sound power and this would be the main difference one would hear "around the midrange"

Might also explain why the well measured R3's don't wow people as much as the measurements would suggest. The more I measure conventional 2-way 2 driver speakers the more I find myself enjoying the LR4 xo's above 2khz and I seem to loose intelligibility and focus with 2 ways that are LR4 crossed below 2khz. The horizontal measures are perfect, so I can only speculate on the vertical dispersion. I had R3's but sold them on, I found them too forward around that upper mid region no matter what I did.

I have to disagree on this one. While having a solid vertical listening window is of importance, reflections are less so due to our horizontally oriented binaural hearing system. What we measure is simply not how we perceive sound. Even manufacturers who follow the research seem to misinterpret this:

Since most research results suggest that what we hear in-room is estimated to be around 12% Direct Sound (from your speakers), 44% Early Reflections (from your room), and 44% Sound Power (how the sound loads your room), we strive to create solutions that'll deliver predictable, consistent, and excellent results both on and off axis – regardless of the living space!

But that's not true. it is not what we hear, it is what we measure. To quote Floyd Toole himself:

With reverberation times of the order of 0.2 to 0.4 s what we hear from loudspeakers is the direct sound, a few early reflections, and not much else.

So what is important? According to Kevin Voecks:

The second-loudest sound, after the direct sound, is the first reflected sound from the loudspeakers. In fact, Harman research has discovered that the first reflection from side walls, both from the wall adjacent as well as the opposite side wall are critically important. The acoustic output of a loudspeaker far off-axis horizontally is very significant, and should match the response of the Listening Window as much as possible. This goal is technically challenging, but is essential for optimum timbre, as well as to provide a sense of seamless coherency

We can actually see this in some well received Harman loudspeakers - while there is usually a bump in the DI - if you isolate the horizontal DI it's usually very good. An in-room measurement however, would not entirely reflect this, and the initial reaction is to correct the dip - if however, the direct sound is neutral - correcting the dip would only make sound worse. I suggest this thread for some good info on the subject.

 

[BrokenEnglishGuy]

I think we need to start looking at vertical directivity more. I mean these speakers are for all intents and purposes identically smooth in response, albeit one slightly wider horizontal radiation pattern that the other, but the main difference I can see is the mid to tweeter XO point, the kii at 2khz and the DD8C at 1500(?)hz. This causes a dip in sound power and this would be the main difference one would hear "around the midrange"

Might also explain why the well measured R3's don't wow people as much as the measurements would suggest. The more I measure conventional 2-way 2 driver speakers the more I find myself enjoying the LR4 xo's above 2khz and I seem to loose intelligibility and focus with 2 ways that are LR4 crossed below 2khz. The horizontal measures are perfect, so I can only speculate on the vertical dispersion. I had R3's but sold them on, I found them too forward around that upper mid region no matter what I did.

I would say its because of their bass.
The slope from R series are weak, and people never realiza that fact in their listening, the speaker extend deep but their slope is in their weak side.
People never listen the KEF R3 + the subwoofer KEF K92.

The wow factor for most people is the bass.



For example People who have the R3 plus 2 subwoofers are very happy.



I have my R7 with a low shelf filter +3dB@125hz, the tower Always make a wow face to my no audiophiles friends and my other friend who is an audiophile also think the R7 needs that low shelf, with that filter the sound became much more dynamic and dense.


Fun enough you founded the R3 too foward, but here my R7 with 20° toe in sounds more in the smooth side.. Never in the foward side