KEF LS50 Meta Review (Speaker)

[Tangband]

While I like and even own the LS50 Meta I often wish this Harman score didn't exist as it oversimplifies loudspeaker preference and choices too much.

Yes thats exactly the point. The Harman score is slightly misleading.

1. In the real world, you always listen to two stereo speakers. That needs some stereo system compensation between 1-2 kHz ( +1,5 dB ) and between 2,5 - 4 kHz ( - 1.5 dB ), and a slightly elevated peak at about 5 kHz . In Erins measurements, we can see that Kef has done some of this.
The loudspeaker has apparently been tuned by ear in a stereo setup.

2. Its easy to draw wrong conclusions from the measurements from Erins measurements of the ls60 = that has a flatter frequency response. Its a very good sounding speaker, no doubt. But is it better sounding than the ls50 Meta?

However - the amplifiers inside ls60 are probably slightly compromised in quality , so If one uses the ls50 meta with a Hypex ncore poweramp or better, it probably gonna sound better because of the higher quality of the amplifier. This was something that I didnt believe in when I had my SAM monitors. I compared those with a simpler passive speaker with a better sounding amplifier , and the simpler speaker sounded better.

3. Because of this knowledge - dont throw out everything you have been learning about ”source first” as a false statement. Listen and compare instead. A better source and amplifier with two ls50 meta might outperform the ls60.

4. One can always try using a digital PEQ ( such as the very good one in WiiM pro ) to make the frequency response flat for the ls50 Meta. In a stereo setup - I doubt it will sound any better, it probably gonna be the opposite.

 

[thewas]

Yes thats exactly the point. The Harman score is slightly misleading.

Wouldn't say its misleading but as a single score metric it has its obvious limitations.

1. In the real world, you always listen to two stereo speakers - that needs some stereo system compensation between 1-2 kHz ( +1,5 dB ) and between 2-4 kHz ( - 1.5 dB ), and a slightly elevated peak at about 5 kHz . In Erins measurements, we can see that Kef has done some of this.
The loudspeaker has apparently been tuned by ear in a stereo setup.

Why wouldn't these be compensated during the mixing stage of the recording?

2. Its easy to draw wrong conclusions from the measurements from Erins measurements of the ls60 = that has a flatter frequency response. Is it better then ?

However - the amplifiers inside are probably slightly compromised in quality , so If one uses the ls50 meta with a Hypex ncore poweramp or better, it probably gonna sound better because of the higher quality of the amplifier. This was something that I didnt believe in when I had my SAM monitors. I compared those with a simpler passive speaker with a better sounding amplifier , and the simpler speaker sounded better.

3. Because of this knowledge - dont throw out everything you have been learning about ”source first” as a false statement. Listen and compare instead. A better source and amplifier with two ls50 meta might outperform the ls60.

While I agree with with your above first sentence I disagree with the rest, namely Tiefenbrun's weighting of the source, even more in the digitial age and hope we keep that long chewed discussion out from the loudspeaker section here.

 

[Tangband]

Wouldn't say its misleading but as a single score metric it has its obvious limitations.

Yes- measuring only one speaker doesnt say what the sound will be with two stereo speakers in a room with walls.

Why wouldn't these be compensated during the mixing stage of the recording?

If you have REAL acoustical instruments in the concert hall as reference , as I do, you can understand me.
If I had the monitors in the studio as a reference, I would have to buy different monitors for every record played.
LTS ( sound technician society ) in Sweden has done research about stereo system corrections using loudspeakers.

Read more here:

faktiskt.io • Visa tråd - Stereosystemfelen, igen

While I agree with with your above first sentence I disagree with the rest, namely Tiefenbrun's weighting of the source, even more in the digitial age and hope we keep that long chewed discussion out from the loudspeaker section here.

A bad sounding amplifier in a system will sound worse than a good sounding amp. Everyone agrees to this. This forum has to much emphasis on the loudspeaker quality.

I dont have any Linn gear in my system.

 

[thewas]

Yes- measuring only one speaker doesnt say what the sound will be with two stereo speakers in a room with walls.

The corresponding research has shown that loudspeaker preference isn't significantly changed by the number of speakers or the room influence.

If you have REAL acoustical instruments in the concert hall as reference , as I do, you can understand me.
If I had the monitors in the studio as a reference, I would have to buy different monitors for every record played.

Welcome to the audio's circle of confusion, neutral measuring devices are the only way out of it.

A bad sounding amplifier in a whole system will sound worse than a good sounding amp. Everyone agrees to this.

The problem is that it rather takes some "effort" to engineer a poor/different sounding amp, although with the current degenerated audio market such exist too. The big majority of decently engineered devices since the 70s are transparent/similar enough though so correctly

This forum has to much emphasis on the loudspeaker quality.

as electromechanical transducers like loudspeakers or turntable pickups have magnitudes higher frequency response and distortion problems then electronics.

 

[Tangband]

The corresponding research has shown that loudspeaker preference isn't significantly changed by the number of speakers or the room influence.

This is not true if we are talking two channel stereo playback. It is true though with 5.1 channels.

Welcome to the audio's circle of confusion, neutral measuring devices are the only way out of it.

The stereo system is so flawed so there are nothing like a ” perfect sound ” to be had in the first place. For good 2 channel playback illusion with stereospeakers, you need some deviations from perfection using only one speaker, such as stereo system compensation. This will make the illusion of real instruments better , if done right.
Here is more information from Sweden :

faktiskt.io • Visa tråd - Stereosystemfelen, igen

”For the simplest case, we can imagine that we are playing a mono signal (something that is panned in the middle, eg a singer). If we think that she is standing and singing for real in front of listeners, we will have a tone curve deviation right in front of the outer ear, which is primarily due to the shape of the head. We can have this deviation as a reference. That's how it should be when we listen to mono sound. If you let the singer sing through two speakers, as the picture shows, we will not have the same tone curve deviation in front of the outer ears. The situation is symmetrical so we only need to check one ear. Let's select the right ear. What happens? The sound from the right speaker will contain more high-frequency sounds in that it plays more 'straight into' the ear. The sound from the left speaker, however, will lack a bit of treble. If you combine the sound from the left and right speakers and measure at the ear, due to the above differences, the tone curve will not be the same as if the sound had come straight through. Then you can simply make sure that you adjust the direct tone curve from the speaker so that the sound is included as if it had come straight from the front. Then you will get a more 'living' sound image.

Now this is not entirely without complications... When you play sounds that are completely panned either to the right or to the left, you get a deviation that is infected by the compensation to make it sound right on mono sound. It's probably less of a problem because there you have quite a lot of agreement with where the sound should give the illusion of coming from and where it really comes from It will still be right s a s. Although it is still a problem because we often use relative tone curve differences to determine directions in the soundscape. In addition, it is not so easy to just check the difference between center-panned sounds and real straight-ahead sounds. You have to see how it should behave from the whole sound image from the front (perhaps at least +/-30 degrees) and make a balance. There will be a lot of trade-offs! 8)

Another stereo error artifact that Ingvar Oehman fixes in the pi60(s) is the height projection. It also contributes to a more convincing released sound image. Although it does not affect the sound on axis on the speakers.”

The problem is that it rather takes some "effort" to engineer a poor/different sounding amp, although with the current degenerated audio market such exist too. The big majority of decently engineered devices since the 70s are transparent/similar enough though so correctly

This is not always the case , as Amirm has shown. Rega IO is a good example. My testing of different poweramps this summer showed audible differences.

as electromechanical transducers like loudspeakers or turntable pickups have magnitudes higher frequency response and distortion problems then electronics.

The colorations are different in speakers than in source and amplifiers.

 

[Digital_Thor]

A neutral system - as a whole - will definitely always be my choice. Especially after years of testing, reviews and DIY. Maybe a bit more "oomph" in the lower frequencies, simply because I find it a bit more entertaining - generally. But my findings are, that with plenty of power and subwoofer displacement, I mostly have enough lower frequencies for both movies and music - even at low volume - when coupled with multi subs - of course equalized.

Even the standard LS50 Meta, is for me a bit "colored", and need a bit of EQ and higher frequency reaching subwoofers, to be at its best.

 

[MAB]

This is not always the case , as Amirm has shown. Rega IO is a good example. My testing of different poweramps this summer showed audible differences.

How did you show this?

 

[Tangband]

Erins review of the ls50 Meta shows why it might sound better in a stereo setup than a flat measuring speaker.
A slighty elevated response between 1-2 kHz, a dip at 2-4 kHz and a small peak at 5 kHz . Maybe a stereo system compensation. It would be interesting if Kef can comment on this ?
If youre having flat measuring loudspeakers at home with good directivity, you can try this with a digital peq.
Erins own comments during the youtube review was that the stereo image was extraordinary good with the Ls50 Meta.

KEF LS50 Meta Bookshelf Speaker Review

Kef LS50 Meta

www.erinsaudiocorner.com

 

[thewas]

This is not true if we are talking two channel stereo playback. It is true though with 5.1 channels.

Do you have a source of this contradiction to Tooles research?

The stereo system is so flawed so there are nothing like a ” perfect sound ” to be had in the first place.

A "perfect sound" was never stated or the asked, the reference for any recording can only be what was heard in the studio and therefore

For good 2 channel playback illusion with stereospeakers, you need some deviations from perfection using only one speaker, such as stereo system compensation. This will make the illusion of real instruments better , if done right.

above is not true as for example a stereo recording is already mixed and mastered with and for the playback with stereo loudspeakers.

This is not always the case , as Amirm has shown. Rega IO is a good example. My testing of different poweramps this summer showed audible differences.

You mean your anecdotal claims of non blinded testing?

Erins review of the ls50 Meta shows why it might sound better in a stereo setup than a flat measuring speaker.
A slighty elevated response between 1-2 kHz, a dip at 2-4 kHz and a small peak at 5 kHz .

Erin criticised in his video review the 2-4 kHz dip and preferred it equalised.

If youre having flat measuring loudspeakers at home with good directivity, you can try this with a digital peq.

Have done that and while some times I have some lust for such voicings (especially with some older/poorer redcordings) I usually keep coming back to flat in the lorn term.

 

[MAB]

Do you have a source of this contradiction to Tooles research?


A "perfect sound" was never stated or the asked, the reference for any recording can only be what was heard in the studio and therefore

above is not true as for example a stereo recording is already mixed and mastered with and for the playback with stereo loudspeakers.


You mean your anecdotal claims of non blinded testing?


Erin criticised in his video review the 2-4 kHz dip and preferred it equalised.


Have done that and while some times I have some lust for such voicings (especially with some older/poorer redcordings) I usually keep coming back to flat in the lorn term.

Thank you for this excellent and simple summary.
As usual, no source, all anecdote.
So many fallacies in Tangband's statements, glad you organized his ideas for him.

I also don't mind a bit extra in reproduction from time to time. I don't get irritated every time I hear a oddly voiced speaker, even enjoy since I like music. But I do always come back to enjoying flat response. And agree, the perfect sound argument is a red herring, most of us are just trying to get good sound out of (stereo, mono, multichannel, etc.), not reproduce an exact event.

 

[Sancus]

While I like and even own the LS50 Meta I often wish this Harman score didn't exist as it oversimplifies loudspeaker preference and choices too much.

Without it, everybody who is not willing to spend literally hundreds of hours reading about loudspeakers would have no real objective way to compare them in even a loose way. Reviewer impressions and buyer impressions are 100x more misleading than the score but I never see anyone say "I wish everyone would stop posting their impressions of a speaker".

 

[thewas]

Without it, everybody who is not willing to spend literally hundreds of hours reading about loudspeakers would have no real objective way to compare them in even a loose way. Reviewer impressions and buyer impressions are 100x more misleading than the score but I never see anyone say "I wish everyone would stop posting their impressions of a speaker".

I agree that for someone new or not willing time to spend understanding loudspeaker measurements its a not bad "tool" (although it wasn't designed for such), also nowadays thankfully good reviews like from Armin and Erin combine both and even show and explain the correlation of measurements and listening impressions.

 

[MAB]

Without it, everybody who is not willing to spend literally hundreds of hours reading about loudspeakers would have no real objective way to compare them in even a loose way. Reviewer impressions and buyer impressions are 100x more misleading than the score but I never see anyone say "I wish everyone would stop posting their impressions of a speaker".

You are absolutely correct.
But I vacillate between liking and not liking scores.
And, I actually do ask for people to stop posting their impressions. Which may be rude, but this is the internet, which can be similar to a refreshing drink from a fire hydrant. It's different than live interactions and I do need to turn off the spigot of opinions and ideas from time to time.

 

[Tangband]

Im sure that Kef didnt choose the frequency response for ls50 Meta because it sounded bad
So why did they choose a non flat response ?

 

[Digital_Thor]

As a compromise to make an overall tiny midrange play a bit of bass. The dedicated midrange from the R3 seems more smooth with its less obstructive suspension profile

Im sure that Kef didnt choose the frequency response for ls50 Meta because it sounded bad
So why did they choose a non flat response ?

 

[Tangband]

As a compromise to make an overall tiny midrange play a bit of bass. The dedicated midrange from the R3 seems more smooth with its less obstructive suspension profile

Is Kef the source of your statement ? Or is it your opinion ?

 

[Digital_Thor]

Is Kef the source of your statement ? Or is it your opinion ?

I have both drivers, measured them, and compared with spinorama.org. My next project is with the dedicated midrange, since I find it hard to hear an improvement in the meta design. I surely wished so... But I think I'll be happier in the long run with a smoother midrange and add woofers that can cross at 400Hz

 

[Tangband]

I have both drivers, measured them, and compared with spinorama.org. My next project is with the dedicated midrange, since I find it hard to hear an improvement in the meta design. I surely wished so... But I think I'll be happier in the long run with a smoother midrange and add woofers that can cross at 400Hz

Thanks for the answer.

 

[RobL]

Im sure that Kef didnt choose the frequency response for ls50 Meta because it sounded bad
So why did they choose a non flat response ?

There is a fair bit of information on the development of the LS50 in Newell and Holland’s book “Loudspeakers For Music Recording And Reproduction”. This is a quote from page 288 (emphasis mine):

An interesting dilemma faced the designers at KEF Loudspeakers in 2012 when they were asked to trecreat' a modern version of the "legendary' LS3/5A (initially as a limited edition) to celebrate the company’s 50th anniversary. Although the LS3/5A had originally been designed by the BBC as a monitor loudspeaker for male-voice recording in outside-broadcast vehicles, it subsequently achieved a great following in the domestic hi-fi world, where it became much revered. The question therefore arose as to whether the designers at KEF should try to create as close a copy as possible of the LS3/5A, or whether they should attempt to 'correct' the known defects.



Considerable technical advances had taken place at KEF (and at sister company Celestion) in the 35 years or so since the original LS3/5A had been designed, including significant developments in driver technology, cabinet design, electroacoustic measurement, and mathematical modelling. Despite this, problems such as port turbulence, edge diffraction, and cabinet resonances were all considered to be part of the classic' sound of the LS3/5A, but it still did not seem to make sense to copy the 'deficiencies' if it was thought that the original designers would have avoided them if they could have done. It was also unknown how a modified design would be accepted by professionals who trusted the original model, and it was not impossible that a newer version which 'solved' some of the long-standing problems may be rejected for not having the original' sound. Nevertheless, the decision was finally made to try to design the best loudspeaker that the engineers could make, whilst still respecting the general objectives of the original LS3/5A design.


After much computer-modelling, different drivers were used, a new port geometry was designed, and the cabinet was made with a more rigid construction and less edge diffraction. In addition, the passive crossover was also considerably redesigned. The results are shown in Figures 8.24 and 8.25, which represent the on-axis responses and the power responses, respectively.? It should also be noted that in the balancing process, the priority was on the subjective performance: not obtaining the flattest response' with reference being made to both the original LS3/5A and the then new KEF Blade. Not surprisingly, the combination of the technical improvements by skilled engineers, and the subjective refinement guided by experienced listeners, led to a product with what must be considered to be a superior overall performance. The smoothing-out of the power-response peaks of the LS3/5A (visible in Figure 8.25 and resulting from the directivity anomalies shown in Figure 8.23) also enhanced the consistency of the sound from the LS50s in rooms with different acoustic characteristics. Eventually, the demand for the LS50 from the domestic market was so great that the limited production was extended into full production.

 

[Tangband]

Interesting!

”It should also be noted that in the balancing process, the priority was on the subjective performance: not obtaining the flattest response' with reference being made to both the original LS3/5A and the then new KEF Blade. Not surprisingly, the combination of the technical improvements by skilled engineers, and the subjective refinement guided by experienced listeners, led to a product with what must be considered to be a superior overall performance. The smoothing-out of the power-response peaks of the LS3/5A (visible in Figure 8.25 and resulting from the directivity anomalies shown in Figure 8.23) also enhanced the consistency of the sound from the LS50s in rooms with different acoustic characteristics. Eventually, the demand for the LS50 from the domestic market was so great that the limited production was extended into full production”

The LS50 was developed using technical improvement AND subjective listening refinement guides by experienced listeners .

So they probably didnt use a single speaker during those subjective listening tests, they used a pair for stereo listening.
Sounds very sensible to me.