KEF R8 Meta Dolby Atmos Speaker Review

[Blumlein 88]

I wonder about these bounce off the ceiling height speakers. I know even back with Q-sound they could do height with a pair of stereo speakers up front partly/mostly thru filtering to mimic the comb filtering that causes us to hear sound as from above around the 8-12 khz range. That has made me wonder if such processing wouldn't allow a 2nd pair of side speakers to mimic overheads without being overhead. I wonder if along with the Dolby mandated ripple in response in that 8-12 khz range they don't add to it or vary it to vary where in the overhead area one hears the sound in their object oriented channel processing.

I've also wondered if an up firing speaker with bass rolled off couldn't feed into a tube that had an angle reflected at the top of the tube so one could actually put much of the sound traveling up the tube and exit at height even though the speaker is mounted on the floor.

 

[HarmonicTHD]

Considering the frequency range, very few overhead or height speakers can reach 40 Hz by themselves. I actually wonder if that is even necessary having subwoofers.

This speaker is not made for 40Hz. It is ca 20dB down at 40Hz plus look at the distortion at 40Hz and 86dB SPL. And how could it, it is a mid/upper driver and meant to be integrated into a surround system with subs.

 

[GXAlan]

@amirm, can you test this pointing to the tweeter axis of the imaginary main speaker and at the distance that is approximately where the Atmos reflection should be?

 

[tktran303]

Neither the original LS50 nor LS50 Meta or other recent 2-way Uni-Qs have such a wide and deep dip there, so I am sure its a conscious tuning choice from KEF, would be interesting to know the reasons.

Interesting that we see different things. I see that mid band dip in the Q150, Q350, LS50, LS50 Meta, even the recent ceiling/in wall specific model reviewed here. All just slight variations of the same thing,

When you say tuning choice: if you mean it wasn’t optimised for perfectly flat on axis response; then I agree- The application is not designed for listening on axis- so it would not make sense to optimise it for on axis. The design target would be to (total) sound power.

If you mean tuning choice as in the was a deliberate choice to add extra passive crossover components in there to create a dip, no I don’t think so.

My guess is that it is designed for total sound power, or predicted in-room response.

But then again I think we need a little more science than this. Does the PIR correlate with with the LP when the speaker is mounted on a front speaker? Or mounted in the wall?

Bring a coaxial and having nearly equal horizontal and vertical off axis responses; and with such well controlled directivity, my guess is that it quite immune to placement considerations. I don’t think the preference score makes sense here.

Take any other your KEF LS50/LS50Meta/UniQ and sit it on its back, perhaps place hard foam or hardcover book as wedge to tilt it and then listen to your favourite music. Sounds surprisingly spacious, but not quite right for stereo music,
But it works well for multichannel / surround/Atmos.

 

[Jukebox]

For my future home cinema room I was debating for front/rear heights (Auro 3D) between these and LS50Meta.
For whatever reason I thought that R8A Meta will be to small/non-linear/high distortion at bigger volumes compared with LS50 Meta.
But now that I saw the measurement, they do make a lot of sense for me; being smaller and with built in wall-mount.

 

[Deleted member 48726]

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

1850 USD /Scandinavia.

 

[exm]

Thanks Amir for testing these speakers. I have 3 pairs of these, replacing Q50As. The ones I have sent to Amir will be my front heights. The others are mounted as can be found in the picture below - I am using these are height channels. From reading the graphs, it seems like a 150Hz crossover makes the most sense?

One interesting thing about the driver. Here are the specs for the R8 Meta and... The LS50 Meta: MF/LF: 130 mm (5.25 in.)
The other R Meta units (R3,R7,R11,R2,R6) all have a slightly smaller driver: MF: 125mm (5in.)

This makes me believe that the R8 Meta is using the LS50 Meta driver and not the R Meta driver; it seems to basically be a tuned closed-box version of the LS50?

 

[AwesomeSauce2015]

I voted "Not terrible", and here's why:
The frequency response and directivity are very good, except for the dip and peak which could be compensating for their intended use case, so I am not knocking them for that.
What I am knocking them for, is the insane distortion rise in the low frequencies. Such distortion is, in my opinion, unacceptable for a speaker from KEF.
Yes. I do understand that the height channels aren't super important, but I believe that having good output capacity across all channels in a surround system is very important to achieving an engaging and non-fatiguing experience, and that bass distortion is just too much for me.

And if you look at my post history, you will find that I do tend to favor speakers with good output capability, as I just like the way they sound. So for someone who has a KEF base layer in their system, and who won't really crank it, this speaker would probably work well. However, for someone like me, who does enjoy turning it up (on occasion), I would want a speaker that can actually get loud.

If KEF had addressed the bass distortion, then I would have voted "Fine", since it is still a good speaker (when you ignore the response issues which may not even be issues).

 

[thewas]

Interesting that we see different things. I see that mid band dip in the Q150, Q350, LS50, LS50 Meta, even the recent ceiling/in wall specific model reviewed here. All just slight variations of the same thing,

In my view in this case it is much deeper and wider than from the rest so not really comparable.

When you say tuning choice: if you mean it wasn’t optimised for perfectly flat on axis response; then I agree- The application is not designed for listening on axis- so it would not make sense to optimise it for on axis. The design target would be to (total) sound power.

If you mean tuning choice as in the was a deliberate choice to add extra passive crossover components in there to create a dip, no I don’t think so.

We agree here, I also don't think they a RLC added suction circuit but that the overall tuning was optimised for off-axis listening.

Anyway, we will see what the KEF ASR members will write about it.

 

[phoenixdogfan]

So is their footprint small enough to fit atop my LS 50 Metas? And how would they perform if mounted on the ceiling? Would be nice if they could be mounted either way and maybe just have some kind of built in eq setting to adapt them for either mounting location. Very pricey as well. I'm also very skeptical of the "just bounce it off the ceiling" approach for Atmos height speakers.

 

[Morpheus]

Neither the original LS50 nor LS50 Meta or other recent 2-way Uni-Qs have such a wide and deep dip there, so I am sure its a conscious tuning choice from KEF, would be interesting to know the reasons.

Could it be that drive unit trim piece design, or not correctly positioned/dislocated in transport peaking out or too pressed in..? The R3 had that problem with the shadow flare trim at its correspondent frequency as shown in previous measurements..

 

[thewas]

Could it be that drive unit trim piece design, or not correctly positioned/dislocated in transport peaking out or too pressed in..? The R3 had that problem with the shadow flare trim at its correspondent frequency as shown in previous measurements..

Good thought, guess it's not impossible, although on the photo at the first post it looks to be quite in position.

 

[HarmonicTHD]

I voted "Not terrible", and here's why:
The frequency response and directivity are very good, except for the dip and peak which could be compensating for their intended use case, so I am not knocking them for that.
What I am knocking them for, is the insane distortion rise in the low frequencies. Such distortion is, in my opinion, unacceptable for a speaker from KEF.
Yes. I do understand that the height channels aren't super important, but I believe that having good output capacity across all channels in a surround system is very important to achieving an engaging and non-fatiguing experience, and that bass distortion is just too much for me.

And if you look at my post history, you will find that I do tend to favor speakers with good output capability, as I just like the way they sound. So for someone who has a KEF base layer in their system, and who won't really crank it, this speaker would probably work well. However, for someone like me, who does enjoy turning it up (on occasion), I would want a speaker that can actually get loud.

If KEF had addressed the bass distortion, then I would have voted "Fine", since it is still a good speaker (when you ignore the response issues which may not even be issues).

You didn’t understand the purpose of this speaker. Nobody in their right mind will operate them below 100Hz probable around 200 or so. These are no full range speakers and meant to be integrated in to a surround system with subs.

 

[Andysu]

You didn’t understand the purpose of this speaker. Nobody in their right mind will operate them below 100Hz probable around 200 or so. These are no full range speakers and meant to be integrated in to a surround system with subs.

those kef for that price is , no comment .

my JBL 8330 mkII THX can do some surprising surround ( with LFE.1 merged into them at present and not had issue yet ) or front or below surround , overhead surround what's that , something new to sell speakers within last 10 years ? overhead surround has been around since at one cinema , gamount odeon screen 1 and screen 2 before it was split up in 1989 , screen 1 remained and had overhead surround x6 , no not what your thinking , since 1969 and CIC/UCI tower park had overhead or experimental overhead surround since 1989 , yet most public today wouldn't believe that ?

gamount odeon 70 cinerama screen 1 , can you see the flush original overhead surround , the other surrounds are JBL 8330 mkI installed 1999 , last time overheads was working was around early 1998 then ?
the overheads were all wired working with Dolby Stereo CP50 , the original 5 screen reminded behind screen three of which would be used , simple wire the x6 overhead

CIC/UCI tower park , speakers was all EV THX approved in all screens ( all screens none THX ) large 2 screens x8 overhead surround and all smaller screens x7 overhead surrounds , i was there . overhead surround back then wasn't published like the craziness of atmos cash cow today , selling small pair speakers at £1k branded atmos , daylight robbery . go ebay get some JBL 8330 mkI mkII or A version , cheaper and lot satisfactory . the angle of the speakers , at tower park , easy and for Dolby Stereo CP55 with SRA5 they were " immersive storytelling 1989 "

 

[AwesomeSauce2015]

You didn’t understand the purpose of this speaker. Nobody in their right mind will operate them below 100Hz probable around 200 or so. These are no full range speakers and meant to be integrated in to a surround system with subs.

I do understand that. However, I still believe that having low distortion in the bass area is important, since even with a high pass filter, the distortion is still there, just a bit quieter.
In the dolby guidelines linked above, it is stated that the surround speakers should be able to achieve 17db above a reference level of 79-85 dbC, without adding significant distortion.
17+79 = 96db. When we look at the 96db distortion graph, the distortion below about 500hz is quite high.
Dolby also specifies that the subwoofer should be crossed no higher than 80hz. They do not differentiate for the height speakers.

So while yes, most people will probably run these with a 120+ hz crossover, and will also probably never run them super hard, I still believe that they are not that great since they lack the ability to hit the reference SPL targets, especially given how much they cost.

 

[Andysu]

View attachment 284446@amirm, can you test this pointing to the tweeter axis of the imaginary main speaker and at the distance that is approximately where the Atmos reflection should be?

at £1600.99 kef r8 to reflect sound from the ceiling . sorry i have to laugh . sound is going to reflect from what ever speaker is used in a room a speaker facing towards a wall or away from wall makes 0 difference . well maybe some highs due to be off , way off axis ? well i think not .

get some ebay JBL 8330 mkI mkII 8330A , way cheaper they actually been used in dub stages to cinemas and THX cinemas and cost far less than some , kef that has never been actually used in real THX cinemas . this pay THX to get THX rubber stamp today is really , well if rich people like to spend £999.99 for pair go heard if the price vs specs makes you happy ,, personally if rich can spend that they can spend $50 per gallon for petrol . see my point of logic ?

i tried that aim it at ceiling and its total bs for home atmos scam speaker . may as well use an speaker , it doesn't work . its only diffusing the sound

Sooty not impressed with dolby atmouse - it doesn't work like a magical wand , home atmos speakers should be the hall of snake oil shame .

 

[rvsixer]

Now I know why Atmos (real Atmos anyway...not bouncy bouncy or sound bar kind) is so misunderstood. Even after the specs are presented, keeping it mysterious is more fun .

 

[Billy Budapest]

These speakers are clearly well designed.
However, there are obvious issues with this bouncy sound.

Placed on-top of a speaker they can interfere with each other.

There has to be loss for distance to the ceiling and loss for bouncing.
I don't think 6 dB is out of the question, so this speaker needs (at least IMO), 4x the power handling, not tested here.

It is reasonable to measure speakers, but up firing speakers are in the snake oil category.

Dolby Atmos is a marketing machine with very little regard to implementation.
For $$ you get the Atmos label. This cannot be argued as my ThinkPad has the Atmos logo, clearly ridiculous.

If mounted to aim at the listener, I'd give them a fine/good.
If mounted to bounce, then the wide dispersion is a negative. I'd give them poor because the concept is bad.
Bouncy sound for Atmos is where the product becomes all marketing and deserves to be called Crapmos.

- Rich

I use up firing speakers for Atmos and I can tell you that they are able to reproduce overhead effects very well. Not sure where the “snake oil” comment is coming from because the effect is convincing.

 

[GXAlan]

get some ebay JBL 8330 mkI mkII 8330A , way cheaper they actually been used in dub stages to cinemas and THX cinemas and cost far less than some

+1 and preaching to the choir. That said, I am scientifically interested/curious to see how it measures when done that way.

My home system is the JBL S/2600 which has the 4660’s horn and compression driver for the top end. I have add the UT-405 to fill in the sound above 16 kHz which is the 2405 slot tweeter.

 

[pierre]

Score 1.8 goes up to 4.4 with an EQ. Nothing special to report: bass are a bit boosted, curves gets flatter.
The 1kHz-2kHz dip is mostly gone but it maybe here for other reasons as Amin mentioned.