Kef Uni-Q generations and speakers

[Crosstalk]

So much arguing, if you own both then you can tell us your opinion and which ones you prefer!
I like the original and expect to like the meta as well since they improved a lot of things besides the meta material!

For me definitely Meta, the first one never impressed me.

 

[bo_knows]

It is, a quite wide almost 3dB difference in the presence region as shown in the Stereophile measurements exactly does that:

source: https://www.stereophile.com/content/kef-ls50-meta-loudspeaker-measurements

Or if you want from the ASR on-axis window measurements of both

View attachment 186266

ASR early reflexions

View attachment 186267

ASR predicted in room response

View attachment 186268

and my own in room response differences

View attachment 186270

as well as the measurements of Soundstage, Napilopez and others...

You see there is are very consistent differences which are very audible and this only partially and indirectly shows the directivity differences (DI by ASR)

View attachment 186273

and the distortion differences which depending on the listening level can be less audible (my own measurements):

I had even posted an EQ to bring the tonality of the original ones closer to the Meta one

Filter 1: ON PK Fc 55.30 Hz Gain -1.00 dB Q 3.000
Filter 2: ON PK Fc 140.0 Hz Gain -1.00 dB Q 2.000
Filter 3: ON PK Fc 450.0 Hz Gain 1.50 dB Q 1.400
Filter 4: ON PK Fc 1500 Hz Gain 2.00 dB Q 3.000
Filter 5: ON PK Fc 2600 Hz Gain -3.00 dB Q 1.700

especially this 3dB presence difference is really and easily audible and makes the original LS50 sound quite shouty, especially in the nearfield or when toed in.

Wondering if this part of the measurement shows the benefit of metamaterial?

 

[Descartes]

View attachment 186283
Wondering if this part of the measurement shows the benefit of metamaterial?

Or other modifications they did overall!

 

[thewas]

So much arguing, if you own both then you can tell us your opinion and which ones you prefer!
I like the original and expect to like the meta as well since they improved a lot of things besides the meta material!

Without EQ the Meta hands down, with EQ the differences get smaller but still the Meta.

 

[thewas]

View attachment 186283
Wondering if this part of the measurement shows the benefit of metamaterial?

Its the sum of the all the changes, how much of it comes from the metamaterial we won't know.

Here I had posted also the distortion components where the lowering of the more audible 3rd is significant

KEF LS50 Meta Spinorama and Measurements

Great review, as always. Thanks for the spins!

www.audiosciencereview.com

 

[Crosstalk]

View attachment 186283
Wondering if this part of the measurement shows the benefit of metamaterial?

Not necessarily, possible they fixed this with change in crossover too. But in distortion at higher levels meta material can help.

Ls50 and ls50 meta do not have the same crossover if m not wrong

 

[thewas]

As said, except the enclosures and ports everything is different between both models.

 

[BrokenEnglishGuy]

It is, a quite wide almost 3dB difference in the presence region as shown in the Stereophile measurements exactly does that:

source: https://www.stereophile.com/content/kef-ls50-meta-loudspeaker-measurements

Or if you want from the ASR on-axis window measurements of both

View attachment 186266

ASR early reflexions

View attachment 186267

ASR predicted in room response

View attachment 186268

and my own in room response differences

View attachment 186270

as well as the measurements of Soundstage, Napilopez and others...

You see there is are very consistent differences which are very audible and this only partially and indirectly shows the directivity differences (DI by ASR)

View attachment 186273

and the distortion differences which depending on the listening level can be less audible (my own measurements):

I had even posted an EQ to bring the tonality of the original ones closer to the Meta one

Filter 1: ON PK Fc 55.30 Hz Gain -1.00 dB Q 3.000
Filter 2: ON PK Fc 140.0 Hz Gain -1.00 dB Q 2.000
Filter 3: ON PK Fc 450.0 Hz Gain 1.50 dB Q 1.400
Filter 4: ON PK Fc 1500 Hz Gain 2.00 dB Q 3.000
Filter 5: ON PK Fc 2600 Hz Gain -3.00 dB Q 1.700

especially this 3dB presence difference is really and easily audible and makes the original LS50 sound quite shouty, especially in the nearfield or when toed in.

I didn't notice before the LS50M are kind of mid centric, that can also explain why some people can ear easily the '' details '' pushing the mids a bit foward
it seems a slope 1khz->100hz.

While the R3 is more linear

 

[thewas]

I didn't notice before the LS50M are kind of mid centric, that can also explain why some people can ear easily the '' details '' pushing the mids a bit foward
it seems a slope 1khz->100hz.



View attachment 186382
View attachment 186383

While the R3 is more linear
View attachment 186386

The R3 has also a small lower mid accentuation in the listening window

but let's better compare the PIRs of both, with the anchor point at 200 Hz like Amir does (and makes sense as its ground tone region which decides how we perceive the tonality spectrum of the overtones):

The biggest deviations of the LS50 Meta is the little bump in the lower mids which makes it a bit more "forward" sounding and the presence dip around 3 kHz while the R3 has a wide mid dip between 800 - 2000 kHz which makes its mids sounding a bit "recessed".

They are two quite different characters which from what we read also in the opinions of many listeners appeal to different people, but wouldn't say one is more neutral than the other as their overall kind of "RMS" deviations seem similar, they are just at different regions.

 

[BrokenEnglishGuy]

The R3 has also a small lower mid accentuation in the listening window

View attachment 186393

but let's better compare the PIRs of both, with the anchor point at 200 Hz like Amir does (and makes sense as its ground tone region which decides how we perceive the tonality spectrum of the overtones):

View attachment 186397

The biggest deviations of the LS50 Meta is the little bump in the lower mids which makes it a bit more "forward" sounding and the presence dip around 3 kHz while the R3 has a wide mid dip between 800 - 2000 kHz which makes its mids sounding a bit "recessed".

They are two quite different characters which from what we read also in the opinions of many listeners appeal to different people, but wouldn't say one is more neutral than the other as their overall kind of "RMS" deviations seem similar, they are just at different regions.

I guess Kef like to put a bit of coloration in their speakers, it's very small coloration. I like it.

My goal was never get a perfectly flat speakers to enjoy my music

Thank for these graph, are very useful to see the differences


Anyway, the coloration is not too much, but actually the rooms can change a lot how they sounds

For example if you got some in-room gain issues between 200hz-500hz with the R3 the mids are going to sounds too much resseced/laid back and kind of blurry, having the same problem with the LS50M are going to be less harmful in comparason to R3

I guess if you put these in a very good room, these spakers are going to sound very similar but the ls50m a bit foward and the R series a bit laid back

 

[Crosstalk]

Yep, but that's why i also wanted bass directivity control, like the d&d 8c or kii three.

Personally i have the R7s in my smalish room, with EQ and placement i got very good sub.
I saw some people measurements from the R series, if you let these to gain some dB in the sub bass and then correct them using in-room EQ, these kind of towers can extend pretty low.


For example this guy bought the KEF R7 + K92, but he prefered the R7+ dirac result without sub, but he got 25hz sub bass.. but in-room, of course.

(This was Dirac, right?) I can't remember.
Here you can see the measurements:
KEF R7 ( yellow+pink ) and also the subwoofer ( light blue line )

How much distortion was there at 25hz ? If we look at R7s graph, the distortion already overlaps the main signal at lower frequencies. That means this speaker should be used only with a high pass at those frequencies to avoid distotion therorctically. But at these frequencies we don’t hear distortion like in high freq. so, that’s fine, but boosting them with a DSP takes distortion to another level. Again, this is possible with most floorstanders. I think the on R 11 or reference 5 it works better.

 

[Mnyb]

There must be a host of other small differences in the uni Q drivers for each price tier.

Comparing ref 3 and R7 max spl 113,5 vs 111 dB ( measurment method not speced for the R nb )and xover for the uni Q 350HZ and 2,8kHz vs 400Hz and 2,9kHz .

So they seems to be able to make the ref series uni-Q go a bit louder and lover in frequency at the same time .

That would make sense for the blade or muons , where people will have huge rooms and still just one 2” dome tweeter .

I’m not sure that I ever play that loud and I will probably ever be able to afford R7

 

[bo_knows]

St

There must be a host of other small differences in the uni Q drivers for each price tier.

Comparing ref 3 and R7 max spl 113,5 vs 111 dB ( measurment method not speced for the R nb )and xover for the uni Q 350HZ and 2,8kHz vs 400Hz and 2,9kHz .

So they seems to be able to make the ref series uni-Q go a bit louder and lover in frequency at the same time .

That would make sense for the blade or muons , where people will have huge rooms and still just one 2” dome tweeter .

I’m not sure that I ever play that loud and I will probably ever be able to afford R7

Stay positive and anything is possible.

 

[BrokenEnglishGuy]

How much distortion was there at 25hz ? If we look at R7s graph, the distortion already overlaps the main signal at lower frequencies. That means this speaker should be used only with a high pass at those frequencies to avoid distotion therorctically. But at these frequencies we don’t hear distortion like in high freq. so, that’s fine, but boosting them with a DSP takes distortion to another level. Again, this is possible with most floorstanders. I think the on R 11 or reference 5 it works better.

I don't know
But even with DSP in my R7s they can go very very loud, bookshelf tend to have their 100% of distortion much earlier in sub bass
This is a comparason r7 vs r3, anechoic chamber, im not sure
at 86dB kef told me the intetion with the towers was lower the 20hz-200hz distortion, but also because towers have 2 ports you can add one port lower, this is clearly represent here:

The distortion start to rise at very super loud volume, If you wanna play super loud the R11 are going to perform better at super loud volumes, but if you wanna super loud volume it's better add 2 subs.. if not, then the R11. But you can add 2 subs into the R11 anyway, lol.
Towers can play very loud..



Having that much distortion at lower frequency's it's ok, most speakers don't do 20hz.. it's pretty normal have a high distortion in that region
Even References or that magico you posted have that kind of distortion, your R11 have 4 woofers, and apparently that help a lot more that a bit better woofer
I guess the R11 can go very loud and play everything with ease

 

[Descartes]

I don't know
But even with DSP in my R7s they can go very very loud, bookshelf tend to have their 100% of distortion much earlier in sub bass
This is a comparason r7 vs r3, anechoic chamber, im not sure
at 86dB kef told me the intetion with the towers was lower the 20hz-200hz distortion, but also because towers have 2 ports you can add one port lower, this is clearly represent here:
View attachment 186627
The distortion start to rise at very super loud volume, If you wanna play super loud the R11 are going to perform better at super loud volumes, but if you wanna super loud volume it's better add 2 subs.. if not, then the R11. But you can add 2 subs into the R11 anyway, lol.
Towers can play very loud..



Having that much distortion at lower frequency's it's ok, most speakers don't do 20hz.. it's pretty normal have a high distortion in that region
Even References or that magico you posted have that kind of distortion, your R11 have 4 woofers, and apparently that help a lot more that a bit better woofer
I guess the R11 can go very loud and play everything with ease
View attachment 186628
View attachment 186630

Adding two subs and crossed over at 80HZ will allow less distortions!

 

[bo_knows]

I don't know
But even with DSP in my R7s they can go very very loud, bookshelf tend to have their 100% of distortion much earlier in sub bass
This is a comparason r7 vs r3, anechoic chamber, im not sure
at 86dB kef told me the intetion with the towers was lower the 20hz-200hz distortion, but also because towers have 2 ports you can add one port lower, this is clearly represent here:
View attachment 186627
The distortion start to rise at very super loud volume, If you wanna play super loud the R11 are going to perform better at super loud volumes, but if you wanna super loud volume it's better add 2 subs.. if not, then the R11. But you can add 2 subs into the R11 anyway, lol.
Towers can play very loud..



Having that much distortion at lower frequency's it's ok, most speakers don't do 20hz.. it's pretty normal have a high distortion in that region
Even References or that magico you posted have that kind of distortion, your R11 have 4 woofers, and apparently that help a lot more that a bit better woofer
I guess the R11 can go very loud and play everything with ease
View attachment 186628
View attachment 186630

Hello,
Could you please provide the link to the website where Reference 3 was tested and this graph was obtained?
Thanks,
Bo

 

[BrokenEnglishGuy]

Hello,
Could you please provide the link to the website where Reference 3 was tested and this graph was obtained?
Thanks,
Bo

Hello,
The graph is from Stereoplay.de
And the link is here:

Testbericht über 1 Standlautsprecher in stereoplay 11/2014 | Testberichte.de

Testergebnisse der Fachzeitschrift stereoplay zu Lautsprecher: „R wie Revolution“

www.testberichte.de

 

[bo_knows]

Hello,
The graph is from Stereoplay.de
And the link is here:

Testbericht über 1 Standlautsprecher in stereoplay 11/2014 | Testberichte.de

Testergebnisse der Fachzeitschrift stereoplay zu Lautsprecher: „R wie Revolution“

www.testberichte.de

Hmmm, it seems they are charging money for their review(s). Do you have the whole review that you can share?

 

[KMO]

Wondering if this part of the measurement shows the benefit of metamaterial?

The LS50 Meta white paper shows a similar distortion comparison and attributes it to the drive system changes:

Looking at figures 18-22 together, we see that the new design gives less modulation of the magnet flux density in the gap, more linear drive, lower voice coil inductance and virtually no change in inductance with voice coil position. The upshot of all this analysis and redesign is a significant reduction in total harmonic distortion (THD) in the frequency range where the ear is at its most sensitive (figure 23).

Although it's an overall comparison, so we have to just take their word for it being mainly attributable to the drive system.

The Metamaterial Absorption Technology strikes me as an odd hybrid of genuine engineering innovation with shiny marketing but not much actual final impact. I think the metamaterials are an important technology, and we'll see more of them, but in this particular product, unlike, say, the Dan Clark Stealth headphones, I don't think it's doing that much apart from saving space and manufacturing cost. But that's not nothing - money and space saved gives more scope for other improvements. (You could say the same thing about integrated circuits...)

 

[Crosstalk]

Adding two subs and crossed over at 80HZ will allow less distortions!

But we don’t have measurements of any subs anywhere in terms of distotion. Also, if you are not used to subs getting them time aligned at listening spot is a PITA. With a tower the manufacturer had already done that almost right for us.