KEF R8 Meta Dolby Atmos Speaker Review

[amirm]

This is a review and detailed measurements of the KEF R8 Meta Dolby Atmos Surround Speaker. It was kindly drop shipped to me by a member and costs US $1,600 for a pair.

The R8 meta is very compact speaker meant to be put on top of your tower surrounds (or bookshelves on stand) to fire up at the ceiling. The reflections then bounce back onto the seating locations giving a (diffused) sense of height. It eliminates the need to cut holes in the ceiling and ability to adjust their aim at will. The R8 meta is gorgeously finished and feels exceptionally solid (the "orange peel" look is reflections from my lightbox, not the speaker).

The binding posts are on the "back" side which is the taller side on the left in above picture:

There is provision for wall mounting which I assume it means surround duty with them pointing down toward listeners from side walls:

I had to do some thinking on how to measure this speaker. Clearly you don't put it on its back side in front of you and listen facing them! I realized that the aim is for you to hear its on-axis response as bounced by the ceiling and therefore, I should measure it with driver 90 degrees to microphone as I would any normal speaker. Imagine this position:

It was tricky to devise a way to securely mount it to the Klippel NFS stand but I managed to do that. Reference axis is naturally the tweeter in the center of the coaxial driver. There is a grill but as usual, I did without it.

I originally wanted to listen to them and then do the review. As you can imagine, testing an Atmos speaker is non-trivial but I have an idea on how to do it. I thought I post the measurements now since there are a few interesting things that apply to this execution and R8 Meta series in general. On the former, Dolby has strict requirement on crossover design of Dolby Atmos speakers as to reduce horizontal bleed from the speaker to the listeners. In an ideal world, all the sound would shoot up the wall and none would come from the sides.

KEF R8 Meta Dolby Atmos Surround Speaker Measurements
Again, keep in mind that on-axis frequency response means the direct sound that is going to shoot up the ceiling and reflect back to one position:

What stands out is the notch between 1 and 2 kHz (crossover frequency is 2.5 kHz). I am 99% sure this is Dolby imposed. If someone has seen the Dolby OEM requirements for the crossover, please post so we can confirm. Putting that aside, we see a lazy but wide (low Q) peaking between 5 and 10 kHz. Is that intentional tool? I don't know. The grill will attenuate that a bit. Directivity is superb (best ever seen?) so equalization should be trivial here for both factors.

Edit: I contacted KEF and they verified that the dip is indeed there but not intended to be. Here is their measurements compared to mine:

We see incredibly good agreement between my and their measurements! The only deviation is in bass where their smaller anechoic chamber is not as accurate as Klippel NFS.

I will give you the early window response even though you have to perform mental gymnastics to figure out how to map them to a speaker that is up firing at an angle:

The perfection in both horizontal and vertical axis cements in your eyes and brains when you look at beam width and directivity measurements:

Puts a smile on my face just looking at them!

This is a sealed and super compact speaker. How will it do in distortion department? Turns out very well thank you:

Yes, bass distortion rises at 96 dBSPL but you will be crossing these over with your sub. During the frequency sweep, even at 96 kHz, what I heard was smooth other than a tiny impairment here and there.

The meta material's job is to absorb unwanted reflections. Although it is mostly there for high frequencies, the entire response seems to be free of resonances:

Finally, here is the step response:

Conclusions
Starting with the "bones" of the speaker, there is clearly some good engineering design and magic is going on with the R8 Meta. Love to test their other speakers. For now, this up firing speaker seems to have characteristics that many other speakers would be jealous of. There are just a couple of frequency response variations that I hope we get to understand as to a) whether they are forced and b) whether the idea behind them is correct. If (b) is right, it would solve a huge problem as far as deploying these upfiring speakers instead of major hassle of sticking speaker in ceiling. More on this if I get a chance to listen to them.

For now, based on what I am seeing, I am going to highly recommend the KEF R8 Meta Dolby Atmos Surround Speakers.

------------
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/

 

[Maiky76]

Here is my take on the EQ.
Please report your findings, positive or negative!

The following EQs are “anechoic” EQs to get the speaker right before room integration. If you able to implement these EQs you must add EQ at LF for room integration, that is usually not optional… see hints there: https://www.audiosciencereview.com/...helf-speaker-review.11144/page-26#post-800725

The raw data with corrected ER and PIR:

Score no EQ: 1.6
With Sub: 5.2

Spinorama with no EQ:

• Not as Flat as i would expect, maybe because the usage?
• Great directivity
• looks free of resonances!
• Would probably make a great satellite in a 2.1 set up if the LF can be helped.

Directivity:
Better stay at tweeter height
Horizontally, better toe-in the speakers by 10deg and have the axis crossing in front of the listening location, might help dosing the upper range.

EQ design:

I have generated two EQs. The APO config files are attached.

• The first one, labelled, LW is targeted at making the LW flat
• The second, labelled Score, starts with the first one and adds the score as an optimization variable.
• The EQs are designed in the context of regular stereo use i.e. domestic environment, no warranty is provided for a near field use in a studio environment although the LW might be better suited for this purpose.
• In this case the Score is fully representative of the perfromance because the LF extension does even reach 100Hz which skew the PIR regression for example...

Score EQ LW: 4.0
with sub: 7.5

Score EQ Score: 4.5
with sub: 8.0

Kef R8 Meta DB Atmos APO EQ LW 96000Hz
May092023-170552

Preamp: -3.4 dB

Filter 1: ON HPQ Fc 86.75,    0.00,    1.15
Filter 2: ON PK Fc 184.56,    -2.84,    1.88
Filter 3: ON PK Fc 947.17,    -1.13,    1.73
Filter 4: ON PK Fc 1360.98,    3.83,    2.23
Filter 5: ON PK Fc 2527.16,    -1.03,    4.25
Filter 6: ON PK Fc 7433.37,    -3.23,    1.17
Filter 7: ON PK Fc 13039.93,    1.54,    1.05

Kef R8 Meta DB Atmos APO EQ Score 96000Hz
May092023-170552

Preamp: -3.5 dB

Filter 1: ON HPQ Fc 86.25,    0.00,    1.15
Filter 2: ON PK Fc 179.02,    -3.03,    1.59
Filter 3: ON PK Fc 1010.95,    -2.04,    2.48
Filter 4: ON PK Fc 1302.22,    4.40,    2.58
Filter 5: ON PK Fc 2514.16,    -2.00,    3.25
Filter 6: ON PK Fc 7518.11,    -4.11,    0.98
Filter 7: ON PK Fc 11528.50,    1.11,    2.59

Spinorama EQ LW

Spinorama EQ Score

Zoom PIR-LW-ON

Regression - Tonal

Radar no EQ vs EQ score
Nice improvements

The rest of the plots is attached.

 

[amirm]

Here is a useful short video on design of R8 meta in general:

 

[JSmith]

It eliminates the need to cut holes in the ceiling and ability to adjust their aim at will.

This is surely welcome in many peoples homes who wish for Atmos, but don't want to install ceiling speakers.

It's reassuring to know these measure well too, so thanks for testing this speaker.

The R8 Meta can serve as a surround or rear speaker with the integrated wall mount feature, or be placed upon another speaker for Dolby Atmos height to bounce sound off the ceiling, bringing Dolby Atmos to the room with no need for custom installation. It has a 12th generation Uni-Q® array with MAT™ to steer the sound for genuine 3D results.

At the heart of the latest R Series is Metamaterial Absorption Technology (MAT), a revolutionary innovation that takes the performance of the 12th generation Uni-Q driver array to another level in terms of accuracy and clarity.

MAT is a highly complex maze-like structure, where each of the intricate channels efficiently absorbs a specific frequency. When combined, the channels act as an acoustic black hole, absorbing 99% of the unwanted sound that comes from the rear of the tweeter, eliminating the resulting distortion and providing a purer, more natural acoustic performance.

JSmith

 

[Nordling]

@amirm by 96 "kHz" you mean 96 "Hz" right .

 

[amirm]

@amirm by 96 "kHz" you mean 96 "Hz" right .

Thanks for catching that. I meant 96 dBSPL. Corrected.

 

[martijn86]

Yes, bass distortion rises at 96 kHz but you will be crossing these over with your sub. During the frequency sweep, even at 96 kHz, what I heard was smooth other than a tiny impairment here and there.

Below the cross-over point where directional sound becomes room pressurisation I don't even think it matters for Atmos speakers. The idea is to have object based surround sound so any frequency that you cannot localise is useless to try and push through a height channel. You're either listening to them with (30-35hz) capable front speakers or (a) subwoofer('s).

Edit: Excuse my lack of English here. I can't recall what the correct terminology is for beaming sound vs. pressure. I think it happens at around 350Hz??

 

[DualTriode]

@amirm ,

Thanks for another review.

You made an interesting comment about the angle of the test microphone relative to the speaker.

I have several, mostly GRAS test microphones in Sparkey's Lab. (Sparkey is the cat.)

All of the microphones are calibrated with the same 90 degree angle to the speaker. Seems like most times we ever see a photo of a test microphone and speaker the microphone is directedly facing the speaker, absent the 90 degree angle.

If you would, will you share the reasoning of how/why it works that way?

Thanks DT

 

[abdo123]

Below the cross-over point where directional sound becomes room pressurisation I don't even think it matters for Atmos speakers. The idea is to have object based surround sound so any frequency that you cannot localise is useless to try and push through a height channel. You're either listening to them with (30-35hz) capable front speakers or (a) subwoofer('s).

Edit: Excuse my lack of English here. I can't recall what the correct terminology is for beaming sound vs. pressure. I think it happens at around 350Hz??

I would not say 150Hz is not localizable but you can easily use 'Directional Subwoofers' (the channel would play from the subwoofer closest to the speaker with regard to layout) and that should solve this concern for people worried about this.

PHENOMENAL SPEAKER OTHERWISE. Looking forward to Amir getting their hands on more traditional designs like the R3 Meta.

 

[sweetchaos]

Special thanks to @exm for sending these and thanks to @amirm for this excellent review.

These were highly anticipated, slightly behind the R3/R6 Meta, which I hope to see on the test bench soon.

 

[amirm]

If you would, will you share the reasoning of how/why it works that way?

The calibration for the microphone is almost always given at 0 degrees and sometimes at one other. These are "omni" microphones so the variation due to angle is small. Still, for the most accurate on-axis response, it is good to point the mic at the reference axis.

 

[amirm]

What are the prices in Europe/UK for those of you who live there?

 

[sifi36]

What are the prices in Europe/UK for those of you who live there?

£1400/pair in the UK, €1600 in Europe.

 

[amirm]

£1400/pair in the UK, €1600 in Europe.

Oh, so not cheaper there.

 

[Jukebox]

The R Meta line is something else; if we get this kind of distortions from the baby R8a, the bigger brothers should be almost textbook... and not for uber expensive prices.
Thanks for this Amir! hope you will get the R7 or R11 to see what more we can get in terms of performance

 

[amirm]

Thanks for this Amir! hope you will get the R7 or R11 to see what more we can get in terms of performance

You all need to pester KEF to send in samples. I don't know why their marketing department hasn't taken the initiative yet to do this. Surely measurements are the best way to highlight their performance than some random subjective review.

 

[Koeitje]

I'm sure that dip isn't there by accident or is an oversight. Maybe @jackocleebrown can explain why its there?

 

[Vacceo]

I wonder how "creative" you can get with placement. These speakers seem to be competent on both bouncing sound from the ceiling and working as surround, but it would also be interesting to consider how they'd work as height speakers directly radiating the Atmos part of sound.

 

[Basjebas]

Oh, so not cheaper there.

We have 19-21% VAT included in our prices. So 1322€ ex. VAT, which would be around 1450$ sort of MRSP like you have in de US, which would make them 10% cheaper over here.

 

[milosz]

I've tried this Atmos "aim a speaker at the ceiling" idea with some Elac atmos speakers ( designed by Andrew Jones ) - They are a decent sounding speaker on their own, and I don't see any real resonance peaks in the measurements I made. So, the premise is that they are at least "OK" speakers, possibly better than "OK." They have this same angle built in to their cabinets. I find the sound doesn't appear to be coming from any kind of height dimension from these things. They just sound like off-axis bookshelves sitting on my towers, or even placed on shelves that are about 5 feet off the ground which flank my listening / viewing area. The Atmos height effect from these doesn't seem- well - effective. It doesn't seem like the sound is bouncing off the ceiling very much, most of what I hear is direct from the speakers.

Makes me wonder if any of these up-firing Atmos speakers will really do the effect justice. I have some small speakers I can attach to mouldings on the left and right side walls at about 7 foot height and aim them up to "splash" over the ceiling, I want to see if that gives a better effect. They haven't got any low end to speak of, but then I wonder how well low frequencies can be localized in a room by listeners....the wavelength at 200 Hz is 150 cm, which is 59 inches., so it would seem to me that a person ought not to be able to sense the location at that frequency any more precisely than ± 59 inches. Could I really tell if a 200 Hz sound was coming from Atmos "up" in my room or would I just localize the sound based on higher harmonics? I think the latter; makes me wonder why you'd need bass from an Atmos speaker.

It wouldn't surprise me to find that Dolby applies some filters to the Atmos "up" channels, reducing the low end. Information on what Dolby Atmos does is kind of sparse. I know it's an object-based surround paradigm, with the decoder deciding which sounds go where based on the sound mix and the number and type of channels in the configuration of the playback system; but I know nothing of what, if any, DSP is applied. Filters? Delays? I'd like to know.