KEF LS50 Meta Review (Speaker)

[Sancus]

I have read a few but not that paper. I just glanced at it and while there is excellent introductory information, I am not a fan of creating a single number metric like that, especially since there is no tool provided to readily compute it. The criticism there is also about THD. That is the wrong focus. I like to look at the harmonic spectrum and then using psychoacoustics, determine level of audibility in the full spectrum. Louis Fielder has done excellent work in this area with respect to audibility of low frequency distortion. I hope to put this in place at some point but finding the time to do the computation has been hard.

That's true. There was some discussion in this thread about @pkane implementing the calculation for DS metric in deltawave. I don't know if it ever went any further, but it seems to me that that would be a good way to go, conclude on a metric for computation and then cooperate with a software developer to implement it. I have no opinion on the metric chosen, I just want something better

That would also increase the likelihood of persuading other reviewers to use such a computation, which would be the big win I think, since the main benefit of the THD graphs is that they are comparable to other sources, even if they are of limited overall use.

 

[Sancus]

That’s because everyone will have a different perception. You can’t place a human within a test system.

A weird criticism. You can apply this to just about anything in the world, but it doesn't change the fact that you can produce a range of audibility and a likelihood of hearing any particular level of distortion, from "no one tested could ever hear this" to "yes everyone tested was able to hear this". Distortion measurements severely lack information of this type.

 

[sarumbear]

A weird criticism. You can apply this to just about anything in the world…

Yes, you can and that is my point!

…but it doesn't change the fact that you can produce a range of audibility and a likelihood of hearing any particular level of distortion, from "no one tested could ever hear this" to "yes everyone tested was able to hear this".

If it is a fact why there’s no such measurement? It’s only a wish, not a fact.

Besides, do you have an example of any measurement that generates the binary response that you expect? What level is loud and what level is quite? How high a frequency should be to be perceived as high and how low for it to perceived as such? How high is a high temperature? Does sex and race change that perception? Etc.

There is no general measurement for human perception— period.

 

[aarons915]

I have tried. If you are looking for IMD as a single number, I have found it to not make any sense. Your first problem is to decide on which pair of frequencies and what amplitude ratio. I tried tons of combinations and none correlated to anything useful as I swept the level. A coaxial that was supposed to have higher IMD didn't and so on.

Note that THD information just falls out of frequency response measurements. A special "log chirp" is used where harmonic distortion just shows up as shadows. With harmonic distortion, we get the non-linearity at all frequencies at once. There is no such thing for IMD. You have to pick two frequencies per above and pray that they are the right ones, deal with noise floor issues, etc.

This is why the traditional IMD tests have never made sense to me either, you basically have to know what your worst case is in advance and then test those frequencies. We were discussing IMD distortion in the Revel thread a few years back on AVS forum and one interesting thing said from a dealer quoting a Revel tech was that they typically choose a low frequency to modulate the woofer and then run a sweep up to 20K, this makes much more sense because you should be able to see the worst case IMD with this method. As far as the constant frequency, I think 80-100Hz makes a lot of sense for the average speaker since below that point the port starts reducing cone excursion. So 80-100Hz modulating frequency at 90db@1 meter, the typical distortion level, and sweep from 100-20k and plot distortion, not sure if that's possible with the Klippel, I know I can't do it in REW.

Like others have said it would also be cool to maybe do an experiment with distortion and how various high passes affect that distortion. A combination of 2nd and 4th order slopes from 60-120Hz or so would give a nice idea I would think.

 

[phoenixdogfan]

Remember that at higher crossover-frequencys, the distance between a floor-standing subwoofer and a standmounted Kef ls50 becomes critical. At 100 Hz a wavelength is 3,43 meter. At 200 Hz it is 1,71 meter.

According to Linkwitz studies, when using a fourth order crossover, you need to keep the distance between main speaker and subwoofer within a quarter of a wavelength at the crossover frequency.
Meaning: crossing over at 200 Hz = 1,71 / 4 = 43 cm .

The distance between the floorstanding subwoofer membrane should therefore be maximum 43 cm from the standmounted Kef ls50:s membrane if the crossover is at 200 Hz. Almost imposible to realize. And impossible with only one subwoofer.

Edit: using an odd-order crossover att 200 Hz is a little less dependent on exact position main-subwoofer. Meaning a 3:rd order crossover.

But at crossover frequencys as high as 200 Hz, its better to have the subwoofer in the same cabinett as the main loudspeaker.

If we're talking about a time related distortion? If so, why wouldn't it be possible to delay the output of the satellites to bring them into alignment, to allow use of the 4th order L-R filter? And couldn't Dirac or some other DSP fix that as well?

 

[sarumbear]

If we're talking about a time related distortion? If so, why wouldn't it be possible to delay the output of the satellites to bring them into alignment, to allow use of the 4th order L-R filter? And couldn't Dirac or some other DSP fix that as well?

Geometry. You can only do that for one position.

 

[archerious]

Hmm LS50 Meta or R3?

 

[MaxRockbin]

Hmm LS50 Meta or R3?

It seems like the R3 would be the easy call except Erin's Audio Corner AND Amir both basically said... Measures Great! But... Meh.
Amir went so far as to say there are just some things the measurements seem not to capture (the reference to Harman is about his preference for the Revel M16):
"At some point we will have to reconcile these differences, either setting me straight on my subjective evaluations being wrong, or us not knowing all that Harman knows about good speaker sound. Let's remember that they won't release a new speaker unless it passes double blind listening tests against its competitor. No other score allows them to skip this test. Components are tweaked until they achieve this. "

 

[beeface]

If KEF are reading this, I think it’s clear that there’s a market for a step up product from the LS50, something like an R3 Meta using the LS50 design language.

Also maybe a stand/sub that matches.

 

[aarons915]

It seems like the R3 would be the easy call except Erin's Audio Corner AND Amir both basically said... Measures Great! But... Meh.
Amir went so far as to say there are just some things the measurements seem not to capture (the reference to Harman is about his preference for the Revel M16):
"At some point we will have to reconcile these differences, either setting me straight on my subjective evaluations being wrong, or us not knowing all that Harman knows about good speaker sound. Let's remember that they won't release a new speaker unless it passes double blind listening tests against its competitor. No other score allows them to skip this test. Components are tweaked until they achieve this. "

Just a few points but after Amir EQ'd the room mode he mentioned that he did like the R3 I believe, he's just never amended that part of the review. Also the only evidence I've seen is that the M16 was compared to the KEF Q300, that was also the speaker that is the basis for claiming all KEF speakers are plagued by IMD distortion even though the newer ones are much better and the 3 ways don't have the issue at all. Remember the R3 came out after the Performa series and M16 so they were never benchmarked against it.

 

[JohnYang1997]

It seems like the R3 would be the easy call except Erin's Audio Corner AND Amir both basically said... Measures Great! But... Meh.
Amir went so far as to say there are just some things the measurements seem not to capture (the reference to Harman is about his preference for the Revel M16):
"At some point we will have to reconcile these differences, either setting me straight on my subjective evaluations being wrong, or us not knowing all that Harman knows about good speaker sound. Let's remember that they won't release a new speaker unless it passes double blind listening tests against its competitor. No other score allows them to skip this test. Components are tweaked until they achieve this. "

Seems this just ain't going away then. Amir has explained numerous times that with room modes EQed out the sound is much more convincing. And you can clearly see in the measurements the bass is quite a bit lacking and we all know by now he takes bass very seriously.

 

[mononoaware]

Recent Audioholics video on room treatments discussed bass traps with tuned/weighted panels only a few inches deep. Could that be it? I don't have the link.

No it was specifically about speaker's in a demo room, and there was a link to an article explaining how. . .

If I remember correctly PSI Audio have developed some "Active" bass traps which was interesting.

 

[Tangband]

I have tried. If you are looking for IMD as a single number, I have found it to not make any sense. Your first problem is to decide on which pair of frequencies and what amplitude ratio. I tried tons of combinations and none correlated to anything useful as I swept the level. A coaxial that was supposed to have higher IMD didn't and so on.

Note that THD information just falls out of frequency response measurements. A special "log chirp" is used where harmonic distortion just shows up as shadows. With harmonic distortion, we get the non-linearity at all frequencies at once. There is no such thing for IMD. You have to pick two frequencies per above and pray that they are the right ones, deal with noise floor issues, etc.

Thanks for the answer.
Regarding the Kef uni-q drive unit:
I looked in erins audio corner where he tested one of the uni-q drivers, Kef q100, with the cone at + 3 mm, 0 and - 3mm, with the help of a battery for the bass unit.

This simulates a variable waveguide for the centre mounted tweeter, such as you will have when you play loud bass frequencys in music .
The difference in frequency response is big . Over 8 dB difference at 7,5 khz .
When the cone moves, it will modulate the tweeter response.

Maybe thats the biggest drawback to have a tweeter mounted in the middle of a bass-unit, playing fullrange?
And maybe that will have some effect on the perceived sound quality ?

 

[Doodski]

Maybe thats the biggest drawback to have a tweeter mounted in the middle of a bass-unit, playing fullrange?

That and some of those designs can sound a bit honky with certain voices.

 

[Puddingbuks]

If KEF are reading this, I think it’s clear that there’s a market for a step up product from the LS50, something like an R3 Meta using the LS50 design language.

Also maybe a stand/sub that matches.

Why not a LS80 or LS100 with a 8 or 10 inch coax and larger cabinet?

Now, these would be interesting.

 

[phoenixdogfan]

If KEF are reading this, I think it’s clear that there’s a market for a step up product from the LS50, something like an R3 Meta using the LS50 design language.

Also maybe a stand/sub that matches.

How about a super LS 50 with a coaxial 6.5 inch woofer/tweeter instead.

 

[beeface]

ya that makes more sense

 

[abdo123]

Why not a LS80 or LS100 with a 8 or 10 inch coax and larger cabinet?

Now, these would be interesting.

yes please!

 

[MarkWinston]

Came in here to read about measurements of the Metas as I have thought about an upgrade from the original LS50s for quite sometime now, read about crossing over at 200hz with 2 subs. Dayuuuuuuuum, looks like I need to rethink my upgrade choices. Any speaker that needs to be crossed over at 200khz with 2 subs at volume is not a speaker which I would actually go for.

 

[Sancus]

How about a super LS 50 with a coaxial 6.5 inch woofer/tweeter instead.

There are 6.5" uni-q speakers like the Kef Q350, which was reviewed. It doesn't measure quite as well as the LS50 Meta, but it's also older. Perhaps they will update it and improve it as they did the LS50.

Why not a LS80 or LS100 with a 8 or 10 inch coax and larger cabinet?

Going larger than 6.5" probably causes problems because the woofer beams too much at the crossover. Kef probably thinks you should move to the 3-ways if you want that much power.