• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

LS50 Meta vs LS50 with EQ

Ata

Senior Member
Forum Donor
Joined
May 6, 2021
Messages
388
Likes
334
Location
Adelaide, Australia
As Ive written in my earlier post, the sound difference between my LS50 and a borrowed Meta from my brother in law is big, back to back but uneq-ed of course. A person have to have wet ears to not hear such stark differences. But if eq-ing can bring the OGs a lot nearer to the Metas Im all for it.

That is understandable, as both speakers EQ well, so you will bring the FR flatness to the same level. Since there will not be significant LF differences it will be down to primarily directivity and such. I bet the preference score delta is smaller, too, but don't have Maiky76's numbers on hand.
 

tecnogadget

Addicted to Fun and Learning
Joined
May 21, 2018
Messages
558
Likes
1,012
Location
Madrid, Spain
As far as I'm aware, the metamaterial plate in the new LS50 Metas aims to reduce a peak in the mid-highs and reduce rearward reflections. Would there be any substantial difference between that and simply EQing the older (and cheaper!) LS50s?

I think you are over simplifying from the start. LS50 Meta is not just about that back plate, they have a completely updated 12 generation Uni-Q array. This and little changes in the crossover is what contributes the most to their smoother response than original version.
But I gess if you apply state of the art DSP to copy the Meta response with original LS50, not many people will pass a blind test spotting the differences...
 

KMO

Addicted to Fun and Learning
Joined
Aug 9, 2021
Messages
629
Likes
903
Directivity was much improved by the Meta, so a difference outside of nearfield listening is to be expected.

This is not something KEF have ever touted. If true (I've not checked the spinoramas yet), how was it achieved? There's no obvious things I can think of that would affect it. Crossover? Tweeter gap damper?
 

Ata

Senior Member
Forum Donor
Joined
May 6, 2021
Messages
388
Likes
334
Location
Adelaide, Australia
This is not something KEF have ever touted. If true (I've not checked the spinoramas yet), how was it achieved? There's no obvious things I can think of that would affect it. Crossover? Tweeter gap damper?

Ask and you shall get:

 

KMO

Addicted to Fun and Learning
Joined
Aug 9, 2021
Messages
629
Likes
903
Ask and you shall get:

Yep, watched that - and it's worth reposting. But I don't remember anything about directivity improvements in the Meta. Everything there in the Meta section is described as being about resonances or distortion.
 

phoenixdogfan

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
3,347
Likes
5,291
Location
Nashville
Yep, watched that - and it's worth reposting. But I don't remember anything about directivity improvements in the Meta. Everything there in the Meta section is described as being about resonances or distortion.



From Amir review of the original:

1631024621154.png








From Amir review of the Meta:

1631024368512.png
1631024368512.png
 

KMO

Addicted to Fun and Learning
Joined
Aug 9, 2021
Messages
629
Likes
903
Those charts are hard to compare - the LS50's chart isn't normalised, while the LS50 Meta's is.

Is that why people are saying the Meta is better - just comparing those charts without realising that? Is there an easier comparison anywhere?

But my question was IF the LS50 Meta has better directivity, why? I can't see any directivity-related changes.
 

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
3,060
Likes
4,420
Location
France
Those charts are hard to compare - the LS50's chart isn't normalised, while the LS50 Meta's is.

Is that why people are saying the Meta is better - just comparing those charts without realising that? Is there an easier comparison anywhere?

But my question was IF the LS50 Meta has better directivity, why? I can't see any directivity-related changes.
Look at the directivity index at the bottom of the spins. The why could be related to crossover and/or driver changes.
 

muad

Senior Member
Joined
Jun 8, 2019
Messages
420
Likes
481
Those charts are hard to compare - the LS50's chart isn't normalised, while the LS50 Meta's is.

Is that why people are saying the Meta is better - just comparing those charts without realising that? Is there an easier comparison anywhere?

But my question was IF the LS50 Meta has better directivity, why? I can't see any directivity-related changes.

As stated above the directivity index was improved. It's pretty much linear now. Textbook!

The crossover and likely driver related changes is probably responsible for better directivity marching between the tweeter and woofer.
 

SS55

Member
Joined
Aug 8, 2021
Messages
49
Likes
58
Sorry to disagree, but the jump from LS50, an already very good speaker, to the LS50 Meta, a great speaker, is not as small or incremental as you say.

Since this is a scientific review site, we should look up the preference scores for both, pre-EQ. That is 4.6 for LS50 and somewhere between 5.7 (Amir's review) and 6.0 (napilopez's review) for the LS50 Meta. This is significant in two ways:

a) 1.1 to 1.4 difference in preference score is very significant and audible. A great majority of people will prefer LS50 Meta to LS50 (but I can't remember how to convert the absolute score difference into a statistical measure);
b) The next "step up" in preference scores at this pace would be around the 7.0 mark, which is beyond the current state of the start for passive bookshelf speakers (and especially two-way ones), or at least those that have been measured today. KEF R3, the current champion, is three way at 6.5 IIRC.

A very good reference when comparing speakers based on objective measurements: https://pierreaubert.github.io/spinorama/index.html

On the other hand, agree on the KC62 vs SB-1000 Pro, have the latter and can't see how KC62 can outdo it with smaller woofer area.
If you read my post above where I mentioned the eq by Maiky76, I have also mentioned the score after optimisation which is 6.36
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,448
Likes
7,957
Location
Brussels, Belgium
a) 1.1 to 1.4 difference in preference score is very significant and audible. A great majority of people will prefer LS50 Meta to LS50 (but I can't remember how to convert the absolute score difference into a statistical measure);

this is not really the big a deal that you make it out to be.

My speakers go from 5.0 to 6.2 with EQ and the difference is very subtle.
 

phoenixdogfan

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
3,347
Likes
5,291
Location
Nashville
this is not really the big a deal that you make it out to be.

My speakers go from 5.0 to 6.2 with EQ and the difference is very subtle.
I own BOTH the Ls 50s and the LS 50 Metas, use the Maky76 eqs for both, and the difference between them is not subtle IMHO.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,448
Likes
7,957
Location
Brussels, Belgium
I own BOTH the Ls 50s and the LS 50 Metas, use the Maky76 eqs for both, and the difference between them is not subtle IMHO.

i'm not saying there isn't a difference between the LS50 and the LS50 Meta, I'm saying that once you're within +/- 2dB tolerance any further EQ is very subtle.
 

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,148
Likes
8,724
Location
NYC
Those charts are hard to compare - the LS50's chart isn't normalised, while the LS50 Meta's is.

Is that why people are saying the Meta is better - just comparing those charts without realising that? Is there an easier comparison anywhere?

But my question was IF the LS50 Meta has better directivity, why? I can't see any directivity-related changes.

Ugh, polar maps. Weird that the meta is normalized, Amir doesn't normally do that. In any case doesn't matter too much as the Ls50 meta is quite linear on axis.

Anyway for the sake of consistency, here are the polar maps for both speakers in both normalized and un-normalized form, based on Amir's data:

LS50 Un-normalized:
1631126952134.png


LS50 Meta Un-normalized:
1631126995171.png


Normalized LS50:
1631127069061.png


yikes mcbikes

LS50 Meta:
1631127094912.png



Now in a much more cultured data presentation...

LS50 Old:
1631127198401.png


LS50 Meta:
1631127274867.png


I prefer the un-normalized SPL charts because I do think it's important to consider both the relative off-axis performance and the actual frequency response of the off axis performance. Especially as some speakers, including the LS50 both old and new, are designed to heard slightly off-axis. When you take this into account the old LS50 isn't quite as problematic. Still, for reference sake, here's the normalized version:

LS50 Original:
1631127396030.png


Now LS50 Meta:
1631127427995.png


A rather dramatic improvement.

Even if you consider that the speakers are designed to be heard off axis, and that both LS50s are flattest around 20 degrees or so, the LS50 Meta remains significantly better. Referencing to 20 degrees off axis we get:

LS50:
1631127626023.png


LS50 Meta:


1631127575184.png


Anyway you cut it, without even considering the metamaterials enhancements, the LS50 meta is a significantly better speaker.
 

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,148
Likes
8,724
Location
NYC
One last, simpler visual I like to use these days. The Listening Window, Horizontal Early Reflections, and Horizontal Early Reflections Directivity Index.


LS50:
1631128600747.png


LS50 Meta:

1631128734114.png


The improvements are more subtle in this case, but still present. The Meta's graph suggests basically textbook soundstage performance, the other suggests a slight discontinuity in the presence region.

Edit: I find these lines tend to correlate best with my with perceived impressions of soundstage than the more dramatic differences shown off in other graphs. The difference will be subtle to most listeners, but it's there. And the overall spinorama shows the more balanced tonality of the meta.
 
Last edited:

phoenixdogfan

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
3,347
Likes
5,291
Location
Nashville
After EQ-ing what are the major differences between both?
The OG LS 50s have a mismatch in directivity in the 2-5 khz range which makes for a more forward peaky upper midrange, lower treble. To my ears, it could be ameliorated with eq but really not eliminated b/c the mid/tweeter crossover created a directivity mismatch. To my ears, it always sounded slightly like two different speakers playing the same notes in that frequency range.

And the thing I'd like to emphasize, is that that discontinuity or lack of smoothness in speaker directivity is a tell that you're listening to a speaker. While I'm not certain directivity needs to be uniform throughout the frequency range, I'm very certain that a mismatch especially in the region where the midrange crosses over to the tweeter, should always be considered a design flaw.
 
Last edited:

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,338
Likes
6,710
Ugh, polar maps. Weird that the meta is normalized, Amir doesn't normally do that. In any case doesn't matter too much as the Ls50 meta is quite linear on axis.

Anyway for the sake of consistency, here are the polar maps for both speakers in both normalized and un-normalized form, based on Amir's data:

LS50 Un-normalized:
View attachment 152160

LS50 Meta Un-normalized:
View attachment 152161

Normalized LS50:
View attachment 152162

yikes mcbikes

LS50 Meta:
View attachment 152163


Now in a much more cultured data presentation...

LS50 Old:
View attachment 152164

LS50 Meta:
View attachment 152165

I prefer the un-normalized SPL charts because I do think it's important to consider both the relative off-axis performance and the actual frequency response of the off axis performance. Especially as some speakers, including the LS50 both old and new, are designed to heard slightly off-axis. When you take this into account the old LS50 isn't quite as problematic. Still, for reference sake, here's the normalized version:

LS50 Original:
View attachment 152167

Now LS50 Meta:
View attachment 152168

A rather dramatic improvement.

Even if you consider that the speakers are designed to be heard off axis, and that both LS50s are flattest around 20 degrees or so, the LS50 Meta remains significantly better. Referencing to 20 degrees off axis we get:

LS50:
View attachment 152170

LS50 Meta:


View attachment 152169

Anyway you cut it, without even considering the metamaterials enhancements, the LS50 meta is a significantly better speaker.

Wow, didn't realize the directivity was that much better with the new model. I would have expected it to be very similar. So it's probably not all that feasible to try to EQ the original to sound exactly the same as the Meta.
 

Dave-Oh

Member
Joined
Mar 31, 2019
Messages
17
Likes
26
Amps Im looking at now are the Peachtree Nova 500 and especially the Anthem MCA 225 Gen 2 in particular. That should bring the best out of the LS50. Im dying for a Rega Osiris and McIntosh MA8900 but its way way way out of my budget.
I just bought the Anthem 225 MCA Gen 2 and its phenomenal. Everything in the bass range is more articulate and dynamic. The upper bass even sounds cleaner. I used to own a big class AB amp with a similar power rating but I don't recall it sounding nearly this good.
 

MarkWinston

Addicted to Fun and Learning
Joined
Jul 13, 2021
Messages
736
Likes
579
I just bought the Anthem 225 MCA Gen 2 and its phenomenal. Everything in the bass range is more articulate and dynamic. The upper bass even sounds cleaner. I used to own a big class AB amp with a similar power rating but I don't recall it sounding nearly this good.
Its great to hear that because that is on the top of my list now to power the stubborn ls50. What are you pairing it with?
 
Top Bottom