KEF LS50 Meta Spinorama and Measurements

[thewas]

The 850 Hz "mess" is really tiny compared to most other 2-way loudspeakers since the port is well designed and at the rear side, but if you cross it higher than 80 Hz you could close the port, as I had also done to my non Meta LS50 in the past.

 

[YSC]

that 850hz "mess" is barely visible and I doubt it's audible.. if blocked the port I believe you will mess up so upper bass FR more

 

[CJH]

Plugged the ports with provided foam plugs and definitely hear a difference with smoother sounding midrange. I am crossing over to subs at 115Hz with 24dB slopes. Subjectively added .5dB in bass to regain proper balance. Hoping Amir (when he gets a Meta to measure) or someone else with proper equipment and knowledge will measure with ports stuffed to objectively confirm.
CJH

 

[jonfitch]

I have blind tested the LS50 and the Revel M105 and felt the same way, because of the very good measurements of the M105, I was certain those were the ones I was preferring but it was actually the LS50 and it was unanimous on every song I played. Coaxials just have a clarity and clearness that I don't think you can get with the typical line source speaker and now that I'm used to the sound it bothers me anytime I listen to any other type of speaker. The other thing I noticed is coaxials sound "bigger" than other speakers and the smooth vertical polars are the only logical explanation since the dispersion in the horizontal direction is narrower.

I liked the LS50 more than the M105 as well, but my belief has to do with the fact that the LS50 has a narrower dispersion and as a result the bass, which is omnidirectional, sounds much heftier than on the M105 even though it doesn't show up in single point FR measurements. The M105 sounds much airier and as a result the highs to me overpower the bass even though the extension is comparable between the two (measurements actually show the M105 has more bass).

I mean doesn't Harman's own research show about 35% of sound quality perception is due to bass? I think when listening to two bass restricted monitors, the one that gives a slightly more sense of bass extension psycho-acoustically should sound "better" if both are somewhat similar in FR.
The other issue is the quality of bass--the M105s have what look like a toilet paper roll inside the port which is creating a somewhat hollow sounding, low-fi bass reinforcement from the port.

 

[thewas]

I liked the LS50 more than the M105 as well, but my belief has to do with the fact that the LS50 has a narrower dispersion and as a result the bass, which is omnidirectional, sounds much heftier than on the M105 even though it doesn't show up in single point FR measurements. The M105 sounds much airier and as a result the highs to me overpower the bass even though the extension is comparable between the two (measurements actually show the M105 has more bass).

John Atkinson compared the LS50 and M106 and both from the anechoic and his listeners position measurement it can be seen that the Revel has a bit more energy above 3kHz which he commented also in the article https://www.stereophile.com/content/tale-two-speakers

Fig.1 Revel Performa3 M106 (red) and KEF LS50 (blue), anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with complex sum of nearfield responses plotted below 300Hz.

Fig.2 Revel Performa3 M106 (red) and KEF LS50 (blue), spatially averaged, 1/6-octave response in JA's listening room.

 

[aarons915]

I liked the LS50 more than the M105 as well, but my belief has to do with the fact that the LS50 has a narrower dispersion and as a result the bass, which is omnidirectional, sounds much heftier than on the M105 even though it doesn't show up in single point FR measurements. The M105 sounds much airier and as a result the highs to me overpower the bass even though the extension is comparable between the two (measurements actually show the M105 has more bass).

I mean doesn't Harman's own research show about 35% of sound quality perception is due to bass? I think when listening to two bass restricted monitors, the one that gives a slightly more sense of bass extension psycho-acoustically should sound "better" if both are somewhat similar in FR.
The other issue is the quality of bass--the M105s have what look like a toilet paper roll inside the port which is creating a somewhat hollow sounding, low-fi bass reinforcement from the port.

Possibly, based on the LW vs ER they really aren't that different to my eyes, though. I'm still experimenting with the type of directivity I like and measurements in general, one thing I don't really like is placing speakers in a narrow or wide directivity category without looking at the actual measurement shape. One of the most impressive speakers I've heard is the KEF Reference One which I know you've heard as well, they are narrow dispersion but didn't sound like it to my ears and the off-axis measurements are basically flat lines that decrease further off-axis. Many designs that people call "wide dispersion" have directivity mismatches between the midrange and tweeter and sound bright to my ears and not as natural sounding. I just bought another narrow dispersion speaker, Neumann KH120, so I'm sure it will give more insight into my preferences once I get some time with them.

My comparison wasn't due to the bass, I always make sure to EQ speakers to be fairly close so it doesn't have as much of an impact on my impressions but I did measure both and the M105 definitely had more bass in room than the LS50. I never had any complaints about the LS50 in my room though, they put a lot of bass for their small size and I ran dual subs as I always do.

 

[Kadent]

How do you guys get the measurements to look so perfect?

 

[napilopez]

How do you guys get the measurements to look so perfect?

What do you mean by 'perfect'?

 

[Kadent]

What do you mean by 'perfect'?

I mean as in no peaks or dips anywhere, and also a very flat line.

 

[napilopez]

I mean as in no peaks or dips anywhere, and also a very flat line.

If the speaker is flat, it'll measure flat

But I suppose you may be asking why the measurements don't look like a typical in-room measurement? Are you familiar with quasi-anechoic measurements?

 

[Kadent]

But I suppose you may be asking why the measurements don't look like a typical in-room measurement? Are you familiar with quasi-anechoic measurements?

My speaker appear to be flat from the manufacturer, but are they? I’m not sure. They definitely aren’t according to my measurements.
What are quasi-anechoic measurements? I’m not familiar with them. Maybe something to do with an anechoic chamber?

 

[napilopez]

My speaker appear to be flat from the manufacturer, but are they? I’m not sure. They definitely aren’t according to my measurements.
What are quasi-anechoic measurements? I’m not familiar with them. Maybe something to do with an anechoic chamber?

Ah okay, that explains it. What is your speaker?

One very important thing to understand is that speakers are expected to measure different in an anechoic chamber and in an actual room. In an anechoic chamber, a good speaker is flat-ish on-axis. In a room, the speaker should be a straightish line too,, but instead of being perfectly level, it should on average tilt down some 8-10dB from 20Hz to 20kHz. (This is assuming a typical living room setup 6+feet away; the closer you are to the speaker, the less the graph will tilt)

The cool thing about anechoic measurements is that using angles all around the speaker, one can generally predict how a speaker will measure/sound in a typical room (hence the "predicfted in-room response" or PIR). However, this doesn't work the other way around, which is part of the reason why anechoic measurements are a lot more important and useful for describing the sound of a speaker. Our ears can usually "listen through" a room to identify the actual sound of the speaker independent of a lot of the room interference. But a typical in-room measurement does not separate the sound of the speaker and the sound of a room because a single microphone and a computer is not the same as two ears and a brain.

So a "quasi-anechoic" measurement is simply a way to effectively approximate a measurement made in an anechoic chamber by dumping room reflections from a data. Using a simple technique called 'gating' or 'time-windowing' you are essentially "cutting off" the reflections from your captured impulse response. In Rew, you do it in the impulse response panel. it looks like this:

See that blip at 7ms? That means that at 7ms after the initial sweep is when the first big reflection hits the microphone. With the purple line I am cutting off the data at 6.5ms in order to ignore everything after that. Once gated ,the data is essentially anechoic -- there are no major reflections in the data you are using for the frequency response.

The big problem with doing this is that you a lot of 'resolution' at lower frequencies, until past a certain point the data is no longer reliable (usually below 200-400Hz, depending on when your cutoff is). So for the bass and low mids, you can instead measure that using other techniques, such as ground-plane or ultra-nearfield measurements and doing a calculation to modify the data to match the farfield data. You the splice the bass and upper frequency data together.

This PDF by the late Jeff Bagby explains how to do quasi-anechoic measurements by mixing far. This wasn't written for REW, but it's not too complicated to translate it for REW.

Note, this is very different from what the NFS Amir and Erin have is doing, a much more complicated process that can actually outperform an anechoic chamber. But whether anechoic or quasi-anechoic, the data more closely approximates the "real" sound of a speaker. It gets to the truth that is usualyl obscured in a typical in-room measurements.

 

[Kadent]

@napilopez Many thanks for the detailed response. This is all new to me and seems very interesting.

I can leave a link here to my speaker: https://www.audiosciencereview.com/...ess-da-wave-speaker-review.20466/#post-676138

I have posts after it showing some various measurements including the one from the manufacturer.

Cheers,
Kaden

 

[phoenixdogfan]

To answer the last question first, it definitely is, below are the measurements you and I wanted to see, all from my normal loudspeaker placement so either quite early gated or MMM.

First the gated on-axis which matches the one of @napilopez well:

View attachment 99751

A MMM based listening window approximation in 50 cm from the loudspeaker which seems to be a bit more decreasing at the highs which could be also due to too large angles at my measurement:

View attachment 99752

A MMM based sound power approximation (rotating the mic the loudspeaker at 2x 360° circles), again very good match, just a similar deviation in the highs:

View attachment 99753

As to be expected than my listeners MMM matches the PIR also well except the high frequencies:

View attachment 99754

A part of the difference can be that he used the 90° orientation and calibration of his mic while I used the 0° one, although my LP MMM usually matches the ASR based one in the highs quite well.

Now to the question of the frequency response difference in the bass of the Meta vs the old Anniversary one which I also own:

View attachment 99756

Low bass seems quite identical, differences are mainly above where the Meta is much smoother, to visualise them better I computed the differences of both L, R and L+R MMM which all match well:

View attachment 99759

From those I computed the average and created manually minimal EQ for people wanting to test Meta tonality on their Anniversary ones, that is minimising their LP MMM differences:

View attachment 99760

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

Please keep in mind that this EQ will only approximate a Meta but not make it identical to it, as both generations have slightly different crossover frequencies, slopes and thus also directivity and also different distortion behaviour but like nowdays several posted EQ settings of the Anniversary, make it sound more neutral.

Talking about distortions, the KEF paper again doesn't lie being reduced at the Meta noticeably both in the bass, mids and treble, both measured at the same, high SPL level, in absolute THD:

View attachment 99762

In relative THD:

View attachment 99763

And their components, Anniversary:

View attachment 99764

Meta, above 100 Hz the dominating component is the 2nd (3rd and above are reduced compared to the Anniversary) which is less audible and less disturbing:

View attachment 99765

Detailed listening impressions and differences I would like to post in a later point as I want to listen more to them.

How high could I cross these over with a single sub (SVS Sb2000). I'm sure I could do 100 Hz without many, if any problems with localization, but could I get away with 140hz which would take a lot of burden off that 4.5 inch driver, and take it out of the range where it seems to have its highest distortion?

 

[jonfitch]

I'm starting to think KEF should experiment some way to improve the bass response of the LS50 without affecting the midrange, maybe for a higher end Reference model refresh or something in order to maintain a curved cabinet and point source coherency. I'm guessing Genelec may have a trademark on the slot hidden woofers, but what about a passive radiator like the buchardt s400?

 

[thewas]

How high could I cross these over with a single sub (SVS Sb2000). I'm sure I could do 100 Hz without many, if any problems with localization, but could I get away with 140hz which would take a lot of burden off that 4.5 inch driver, and take it out of the range where it seems to have its highest distortion?

Yes, try crossover frequencies between 100-150 Hz which I also used in the past for my Anniversary ones.

 

[thewas]

I'm starting to think KEF should experiment some way to improve the bass response of the LS50 without affecting the midrange, maybe for a higher end Reference model refresh or something in order to maintain a curved cabinet and point source coherency. I'm guessing Genelec may have a trademark on the slot hidden woofers, but what about a passive radiator like the buchardt s400?

A passive radiator is acting very similar to bass reflex which they already have so no improvement there, I was thinking of an 3-way active version with a woofer on the back (like the D&D 8c) and still a similar great curved cabinet.

 

[phoenixdogfan]

A passive radiator is acting very similar to bass reflex which they already have so no improvement there, I was thinking of an 3-way active version with a woofer on the back (like the D&D 8c) and still a similar great curved cabinet.

And cardioid! And l.t. $7k.

 

[aarons915]

How high could I cross these over with a single sub (SVS Sb2000). I'm sure I could do 100 Hz without many, if any problems with localization, but could I get away with 140hz which would take a lot of burden off that 4.5 inch driver, and take it out of the range where it seems to have its highest distortion?

I made this post a few years ago after compiling all of the studies I could find on localization. The 1st study shows that a single sub located in between the mains could be crossed over up to 140Hz without localization while duals subs were shown to be better in the other studies. I do think your bass preferences would play a big part in this though, these studies seem to assume that the mains and sub SPL are even but if you like a "house curve" or boosted bass then I would think the crossover would need to be lower.

The other thing to consider is your slope, I had the original LS50 for a few years and was always trying different crossovers and slopes to get the best integration and I found 4th order worked the best at attenuating the signal going into the LS50 while keeping the crossover low enough to avoid localization. A 90 Hz 4th order high pass is still down about 3 db at 120Hz but the sub is less likely to be localized since it's rolling off at 90Hz.

https://www.avsforum.com/threads/subwoofer-localization-studies.3058756/

 

[napilopez]

I'm starting to think KEF should experiment some way to improve the bass response of the LS50 without affecting the midrange, maybe for a higher end Reference model refresh or something in order to maintain a curved cabinet and point source coherency. I'm guessing Genelec may have a trademark on the slot hidden woofers, but what about a passive radiator like the buchardt s400?

A passive radiator is acting very similar to bass reflex which they already have so no improvement there, I was thinking of an 3-way active version with a woofer on the back (like the D&D 8c) and still a similar great curved cabinet.

Honestly, imo all KEF really needs to do is just glue an LS50 Wireless II on top of a KC62 and it would have a near-perfect active bookshelf.

Even as separate cabinets, if you literally just sit an LS50 Wireless 2 on top of a KC62, the dimensions are still just 22x10x12.

That's just a little bigger than a Genelec 8351B(18x11x11), and presumably KEF might be able to shave a few inches if, you know, it actually tried to keep everything in the same cabinet.

By comparison, a pair of the 8351B costs ~$8,000, while a pair of LS50WII + dual KC62s (again just as separate devices) would be $5,500. As tuned out of the box, at least, the KEF combo would have far more bass (and lets face it, most people are not doing fancy DSP tricks to extend the bass extension of their speakers).