KEF R11 Meta Tower Speaker Review

[peniku8]

I am not using AP for these tests. It is all done by Klippel. I use the same 50 dB scale as in the frequency response graph so we don't lose sight of what we are looking at. Zooming in makes little sense as the measurements have noise in them and such zooming will exaggerate differences that may not be audible at all.

Ah of course it's Klippel, idk why my mind was on AP there.
We use different scales for different purposes, just as you use a different scale to show distortion, a different scale should also be used for compression. Different scales don't exaggerate the graphs&effects, they make them easier to read. People exaggerate effects when they don't know what they're looking at, which is somewhat what's happening with the DAC measurements sometimes.

 

[Daka]

The R3 initially measured well but had a surprisingly poor subjective review. But these are different?
I wonder how the R11M compare with some of the Revel floor standers?

R3 meta measure great and have great subjective reviews - mostly best in class. I personally like R3 meta much more than original R3s. Enough so to pay double for them.

 

[Morpheus]

The thing Kef seems to consistently underdo subjectively for me is the bass. They allow for to much room gain on their upper echelon speakers for my taste, and so end up sounding a bit wimpy and thin, while their older Ref series (threes and fours) were a bit the contrary..Correcting and extending my KEF Ref 205 using h RME ADI 2 DAC adaptative loudness and Focus Fidelity generated FIR filters in Roon, made them sound so much better it is really an eye opener.I use a modified Harman style curve, with 4 to 5 db bass boost instead of the original 6db, and 1 to 1,5 db less downward high freq tilt past 1Khz. Kef s quality,low distortion, low compression drivers, especially in multy way models, take this Equing with no problem.

 

[dogmamann]

This is a review, listening tests, EQ and detailed measurements of the KEF R11 Meta floorstanding speaker. It was sent to me by the company and costs US $3250 each.
View attachment 359169
The R11 Meta is gorgeous looking with high gloss finish and cabinet that has been shrunk as much as possible to basically hug the drivers. The coax center driver is the star of the show carrying most of the audible band from 200 Hz up (see measurement below). I was impressed with the engineering that went into binding terminal of all things:
View attachment 359170

With a turn of a knob, you connect the bass to the coaxial and vice versa! No jumpers here. Even the cardboard that the speaker came in has clever features like plastic that you pinch and it releases the sides of the box so you can get it out easier. First class execution all around.

KEF R11 Meta Speaker Measurements
As usual we start with our Klippel NFS robotic measurements of frequency response with acoustic center set to center of coaxial driver and grill left out (as well as port plug):
View attachment 359171
I sent out the measurement to KEF and correlation was excellent with their measurements. Their response was a bit smoother than mine but otherwise showed the same dip around 1.2 kHz and dip in bass response. Company explained that the former is diffraction related and goes away off axis and bass shelving was doing was done to accommodate room gain. I will check for this in listening tests later. For now, we can admire the nice directivity which is highlighted in off-axis response that is smooth and sloping down as we want to see it:
View attachment 359172

Very nice. Simulating room response we get:
View attachment 359173
The dip is not as pronounced now which is good.

I forgot to measure the port response but here is one of the woofers and coaxial driver:
View attachment 359175
Even in good speakers I am used to seeing woofer break up/resonances but here, that is so suppressed. Credit goes to the coaxial driver which goes so low, allowing earlier roll off of the woofer response.

Coaxial driver brings uniform directivity and that is precisely what we see:
View attachment 359177View attachment 359178
View attachment 359179

While competing waveguide solutions manage similar behavior horizontally, vertically they are usually a mess. Not so here. Vertical response of the R11 Meta is almost as good as horizontal -- a nice bonus!

Those quad woofers work to bring ease of bass and SPL handling:
View attachment 359181
View attachment 359182

It sounded clean even during sweeps. So I decided to push it to 102 dBSPL:
View attachment 359183
This is why you buy a high-performance tower speaker folks instead of bookshelf. Same amount of floor space but far better handling of music at elevated levels.

Some of you worry about the misnamed speaker "compression" so here are the three responses adjusted to land on top of each other, with proper vertical scale:
View attachment 359184
There is just no audible consequence as a result of going from 86 dBSPL all the way up to 102 dBSPL.

Impedance minimum falls at higher frequencies making it easier to handle as music is not as loud there:
View attachment 359185

There are some resonances as is the case with just about every speaker I measure:
View attachment 359186
And here is the step function for fans of that:
View attachment 359187

KEF R11 Meta Listening Tests and Equalization
Due to heaviness of the speaker, I tested the R11 Meta in our living room as you see in the review picture. This is a massive open floor space with ceiling at some 25 feet. Speaker was away from the rear wall to the tune of 5 to 6 feet (about 2 meters). Stock sound seemed "accurate" for the lack of a better term. I was curious what effect EQ would have on the two things that were visible in on-axis response: dip at 1.2 kHz and bass shelving:
View attachment 359188
I started with the 1.2 KHz (Band 2). This gave female voices more brilliance and pulled them out in front of the speaker a bit. Depending on the clip I played, I could see how someone would prefer it without EQ, while others would want it with that small correction. It was a trade off between sounding a bit bright (with EQ) vs a bit recessed (stock).

I then dialed in the bass response with that boost. I was prepared for some distortion but nothing remotely was audible in that front. Instead, I was greeted with glorious, deep bass that was substantially more rewarding than stock response. What's more it helped to balance the overall tonality with the 1.2 kHz filter, no longer having that tad brightness effect.

I cranked up my amplifier to 0dB reference and started to play track after track. Every piece of music was glorious. Deep, deep bass that was clean as a whistle. Upper range response was delightful while not being accentuated at all. With my wife and dogs around, my testing especially at these playback levels is usually limited but I was enjoying the speaker so much I kept going. Next thing I know, our female dog is worried, running to my wife to hold her. And the male dog coming to me giving me that look of: "what are these loud sounds???" After they did this three times I decided to sadly quit.

Let me summarize it for you: the stock tuning is designed to not remotely offend. The shelving in bass will mean even if there are significant room modes, speaker will not get boomy as I routinely hear from flat response speakers. And the small dip at 1.2 kHz means it will never sound sharp either even if the recording is such. In that regard, I would call the R11 Meta tuning "conservative." There may indeed by something to this tuning as starting point. Since we must measure the response and EQ in bass anyway, we can make the correction I did. But for those who don't, they will get a better response. So the choices here seem wise even though I like very much preferred the filtering I applied (especially in bass).

Conclusions
We expect excellence, objectively optimized response from KEF speakers and we have that in R11 Meta. My experience with budget coaxial designs is that they give up power handling which to me is a poor trade off. Not here. The R11 Meta has excellent bass handling with very low distortion allowing me to EQ it with no degradation as far as distortion of playback ability. There is a bit of room left here in there for enthusiasts who want the optimal performance to get there with EQ. Result was that even in our living room with many hard surfaces and large space to boot, a single R11 Meta roared to action, delivering optimal and super enjoyable response on every reference track I threw at it. Science and excellent engineering works!

I am happy to recommend the KEF R11 Meta speaker. Not only does it perform well, it is well priced as well.

------------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Why were they measured without the feet? I think that’s also an integral part of the speaker which should have some effect on the sound.

 

[DMill]

Another great review. Another great KEF product. I can’t be the only one thinking $6500 for a pair is not chump change though. What’s sad about this hobby is how many $6500+/pair speakers don’t deliver. At least this one does. It’s sometimes the little things that speak to how thoughtful the overall product is. In this case the jumper bar solution is really smart. They probably are worth the money. One could easily argue you’d need nothing more for at least the next 20 years.

 

[CleanSound]

Another great review. Another great KEF product. I can’t be the only one thinking $6500 for a pair is not chump change though. What’s sad about this hobby is how many $6500+/pair speakers don’t deliver. At least this one does. It’s sometimes the little things that speak to how thoughtful the overall product is. In this care the jumper bar solution is really smart. They probably are worth the money. One could easily argue you’d need nothing more for at least the next 20 years.

If you were buying these in the US, I assure you, you can get a discount.

The second thing is, while $6,500 is no chump change, in the grand scheme of things, it's "reasonable" for those in this hobby. Speakers will last you at least 10 years, most likely 20+. How many new computers and new phones you have to buy in this period?

And even if you can't swing this price, there are so many options less than $3k out there for phenomenal speakers.

 

[Willyspu]

The vertical response is remarkably good for a passive array of this size. That little coax is doing a ton of heavy lifting, to keep distortion low on that little unit operating from 200hz up is very nice work by KEF.

$6500 for two of these strikes me as a pretty good deal - isn't this their biggest/best speaker before the ls60/blade things?

The Reference 5 Meta is the next level up at $21,999.98 for the pair.

 

[amirm]

Why were they measured without the feet? I think that’s also an integral part of the speaker which should have some effect on the sound.

They come separate and you have a choice of using them or not. It won't make a difference in measurements anyway. As I explained, the measurements highly correlate with KEF's anechoic room measurements. Let's agree they know how to test them!

 

[amirm]

We use different scales for different purposes, just as you use a different scale to show distortion, a different scale should also be used for compression.

We do use different scales but such scales need to be chosen carefully. The CEA-2034 standard specifically mentions 50 dB vertical range and proper aspect ratio:

"Data shall be reported in graphical form as illustrated in Figure 4. This chart follows the guidance of IEC 60263, which stipulates that the horizontal length for a 10:1 frequency ratio be equal to 50 dB on the vertical scale. "

And importantly:

"All measurements have errors, including measurements in anechoic chambers and outdoors, so it is important to validate your own techniques by comparing the results to a measurement made in a free field environment. A good way to do this is to submit one of your speakers to a recognized testing laboratory. An agreement of ± 1.5 dB is considered to be good."

The underlying measurement system is subject to noise and variability. No way you want to zoom in beyond the accuracy level of the data itself.

There are also complex micro-level variations in measurements which are not understood and so not possible to label.

 

[alont]

The shadow flare is the one big issue I have with the R series. I was hopeful it would be resolved with the meta refresh but alas it seems like that's not the case.

Thanks for the measurements!

 

[vkvedam]

KEF never ever disappointed me! Thanks for the review @amirm, where's the panther?

 

[peniku8]

And importantly:

"All measurements have errors, including measurements in anechoic chambers and outdoors, so it is important to validate your own techniques by comparing the results to a measurement made in a free field environment. A good way to do this is to submit one of your speakers to a recognized testing laboratory. An agreement of ± 1.5 dB is considered to be good."

The underlying measurement system is subject to noise and variability. No way you want to zoom in beyond the accuracy level of the data itself.

There are also complex micro-level variations in measurements which are not understood and so not possible to label.

This accuracy concern is valid in an absolute sense, but when you're taking compression sweeps, this is a relative comparison.
You're changing nothing but the input volume of the DUT. An accuracy of ± 1.5 dB in this context would be terrible.
Also, I'm not sure I follow why you quote the guidelines on CTA-2024. CTA-2024 doesn't include a section on linearity/compression.
I don't understand why you think presenting this data in a more digestable format is a bad idea, while you present DAC linearity zoomed in to a 10dB scale. I don't think obsucring data in fear of misinterpretation is the way to go, rather, we should educate those who misinterpret said data and I think this forum is doing a great job in achieving this in general.

 

[milosz]

Very nice. That low distortion characteristic will really help bass sound deep. Kind of amazing that they were able to keep the THD down like that all the way to the bottom.

How's the group delay in the bass? That's usually where some problems arise with ported systems.

 

[NTK]

CTA-2024 doesn't include a section on linearity/compression.

Yes it does. Just no one I know of publishes test results according to it, likely due to the difficulty of measuring speakers at high output for long durations without an anechoic chamber. See section 8 of the procedure. Below is the procedure for "maximum usable continuous output SPL" (SPLmuco).

 

[Somafunk]

Good test and performance, @amirm.

I like coaxial acoustic systems with good sound coherence.

Some of my coaxial Dual Concentric Tannoy. These DC Tannoys have been updated and improved.

TANNOY 615
Peak Power channel = 300W
FR: +/-3dB = 41hz - 30kHz
Sensitivity (2,83V 1m) = 92dB

I bought the Tannoy 605’s back in 1992 as I thought they looked cool and I liked the sound, used them “very enthusiastically“ in my system as any 20yr old should and I gave my entire technics separate’s setup and 605’s to a mate in 2007, all still going strong and still working perfectly fine to this day 32 years later.

 

[douede]

My current speaker is the R900 (with EQ 41 Hz -10db).
I have always found it a bit slow, lacking slam, and with a kinda "plastic" treble.
Wondering if this R11 meta is significantly better...

 

[Robbo99999]

Expensive speaker, but not controversial, voted it Fine. I'd probably look at other options given the cost, would I find a speaker which has been tested here on ASR that has such low distortion and similar or better frequency response & directivity for a lower price - probably, but I'm not looking!

 

[peniku8]

Yes it does. Just no one I know of publishes test results according to it, likely due to the difficulty of measuring speakers at high output for long durations without an anechoic chamber. See section 8 of the procedure. Below is the procedure for "maximum usable continuous output SPL" (SPLmuco).

View attachment 359386

Fair enough, but this is a different process and a different metric. And of course the data presentation of giving a single number (SPLmuco) does not help us in the scope of the prior discussion.
Nobody publishes this data because it's useless for broadband speakers imo. It's somewhat useful for speakers with limited bandwidth, such as subwoofers, but there, CTA-2010 is much more interesting.

 

[NTK]

Fair enough, but this is a different process and a different metric. And of course the data presentation of giving a single number (SPLmuco) does not help us in the scope of the prior discussion.
Nobody publishes this data because it's useless for broadband speakers imo. It's somewhat useful for speakers with limited bandwidth, such as subwoofers, but there, CTA-2010 is much more interesting.

If you've read both the CTA-2010 and CTA-2034 standards, you'd realize that both standards use the same methods of testing for max continuous and max peak SPL outputs.

 

[peniku8]

If you've read both the CTA-2010 and CTA-2034 standards, you'd realize that both standards use the same methods of testing for max continuous and max peak SPL outputs.

View attachment 359392

Yes, sorry for not clarifying. I was refering to the peak output figures, since that's the data that you'll find when looking for CTA-2010, not continuous output data.