A new Kef speaker? The Kef LS Luxe

[ChrisHeinonen]

Here is the New York Times Wirecutter (tech review for normies) on AV receivers. Note the bullet point on room correction.

View attachment 549056
https://www.nytimes.com/wirecutter/reviews/best-receiver/

If I were KEF, I'd sell room correction as a thousand-dollar upgrade. Have a mic input and software built-in, but sell a mic (and access to the software) separately. The software should work like Lyngdorf's Room Perfect, which does not require reading any graphs or doing anything other than moving the mic around the room. Retailers could also sell in-home calibration as an extra-cost service.

You're welcome, KEF. My consultancy bill is in the mail

As the person that used to do this guide, I bought an ABX tester so I could compare receivers without knowing which is which. In the same price range, the only way to distinguish them was the quality of the room correction. Speakers are different as they could be recognized with the ABX tester, but room correction was the most important thing with a receiver when it came to sound quality.

When it comes to a smartphone microphone, Sonos Trueplay only works with iPhones because the microphone response is consistent across models while no Android models are supported since they can vary. It's possible for KEF and others to only support certain devices and require an external microphone like the USB-C one from Parts Express that is more than good enough for this (https://www.parts-express.com/Dayto...SB-C-Calibrated-Microphone-390-813?quantity=1).

 

[CB89]

Manufacturer Nubert offers room correction for its active speakers - even for models priced at €700 apiece. For this, you can use an iPhone’s microphone (since they are standardized) or purchase a USB microphone for Android smartphones for €40.
I would describe the typical Nubert customer as an everyday consumer or someone with a moderate interest in audio.
In my opinion, this is the sort of thing one should expect as a feature these days.

As for the aesthetics... I’m not entirely sure. I like Genelec and the LS50s. The LUXE models, on the other hand, look a bit too egg-shaped. Perhaps a few colored models would have highlighted the shape to good effect.

 

[stren]

Maybe KEF should just opensource the software/firmware that runs on these, then users can modify it to add the PEQ support. They might find and fix a few of the bugs too People aren't buying these for the software - it's often in spite of it.

 

[Gwen 22]

@stren I hope—and think—you're joking!

 

[stren]

@stren I hope—and think—you're joking!

I'm not. Kef doens't have to support it beyond uploading a repo. If you use it and brick your speakers you are on your own Costs them very little, maintains enthusiast goodwill and may prove useful when users create bug fixes and features that they can learn/borrow from in future.

 

[kemmler3D]

I'm not. Kef doens't have to support it beyond uploading a repo. If you use it and brick your speakers you are on your own Costs them very little, maintains enthusiast goodwill and may prove useful when users create bug fixes and features that they can learn/borrow from in future.

I'm for it and would even mess with it ... if there was a very reliable factory reset available. I'm happy to upload a custom firmware but not at the risk of bricking the speaker and having to send it in for repair.

I think if KEF is reluctant to do this (aside from support issues) it might just because it would expose too much in terms of DSP tuning and BOM. On the other hand, if anyone wants to reverse engineer the latter, they just need to buy a pair of speakers and a set of screwdrivers.

 

[markstein]

And most automatic correctios are quite suboptimal, I have myself a WiiM and in the end nothing beats a manual correction from someone with experience in that field.

Let's face it:
- There is no such thing as a "room correction". This wording is as misleading as Tesla's "autopilot".
- You can correct bass response if you know what you are doing. You should do that in any case!
- Automatic full range "correction" does not work and does not sound good IME. I have tried Acourate, Dirac and a Sonos correction (awful).

I can understand - and I like - that KEF does not throw in just another "room correction" algorithm into their products. A manual PEQ (expert mode), however, would be really nice!

 

[Gwen 22]

I'm for it and would even mess with it ... if there was a very reliable factory reset available. I'm happy to upload a custom firmware but not at the risk of bricking the speaker and having to send it in for repair.

I think if KEF is reluctant to do this (aside from support issues) it might just because it would expose too much in terms of DSP tuning and BOM. On the other hand, if anyone wants to reverse engineer the latter, they just need to buy a pair of speakers and a set of screwdrivers.

I don't think KEF is worried about reverse engineering; if Dynaudio or Koisch wants to do it, they can easily do so right now (and so can others), but the risk of seriously damaging the speaker is very high. In a speaker like the LS Luxe, the DSP handles a wide range of critical functions, starting with the tweeter/woofer crossover frequency and the associated filter slope. One wrong adjustment would be enough to destroy the uniQ. The control of the woofer's amplification in the low-frequency range is most likely tailored to the woofer's physical characteristics, and therefore an inappropriate setting could damage it.
I'm not even talking about the VECO—with that, too, you'd need to know the drivers' physical specifications in great detail and have extensive knowledge in the field before you could even begin to understand the settings; messing with it seems like suicide to me! I'll skip over the likely sriver linearization settings, which are probably still active even without the DSP and could damage the speakers if set incorrectly.
Here, I'm just listing the things that immediately come to mind—there must be many others. Allowing access to it seems completely surreal to me, given how high the risk is of causing irreparable damage to the speaker.
To the best of my knowledge—though I could, of course, be wrong—no professional active speaker offers this much freedom to adjust the settings, so in a consumer-grade “live-style” speaker....

 

[kemmler3D]

I don't think KEF is worried about reverse engineering; if Dynaudio or Koisch wants to do it, they can easily do so right now (and so can others), but the risk of seriously damaging the speaker is very high. In a speaker like the LS Luxe, the DSP handles a wide range of critical functions, starting with the tweeter/woofer crossover frequency and the associated filter slope. One wrong adjustment would be enough to destroy the uniQ. The control of the woofer's amplification in the low-frequency range is most likely tailored to the woofer's physical characteristics, and therefore an inappropriate setting could damage it.
I'm not even talking about the VECO—with that, too, you'd need to know the drivers' physical specifications in great detail and have extensive knowledge in the field before you could even begin to understand the settings; messing with it seems like suicide to me! I'll skip over the likely sriver linearization settings, which are probably still active even without the DSP and could damage the speakers if set incorrectly.
Here, I'm just listing the things that immediately come to mind—there must be many others. Allowing access to it seems completely surreal to me, given how high the risk is of causing irreparable damage to the speaker.
To the best of my knowledge—though I could, of course, be wrong—no professional active speaker offers this much freedom to adjust the settings, so in a consumer-grade “live-style” speaker....

If I had access to the firmware I would certainly leave all the limiters and crossovers where they are! I think what would be interesting is seeing if someone could get FIR in there, then some more advanced room correction becomes possible, etc.

 

[Gwen 22]

If I had access to the firmware I would certainly leave all the limiters and crossovers where they are! I think what would be interesting is seeing if someone could get FIR in there, then some more advanced room correction becomes possible, etc.

Even for someone who is careful, a handling error is always possible.
There will certainly be a small number of people who will try anything and everything—what about the warranty in that case?

 

[kemmler3D]

Even for someone who is careful, a handling error is always possible.
There will certainly be a small number of people who will try anything and everything—what about the warranty in that case?

I think custom firmware would have to void the warranty for any damage caused by overloading the speaker.

Ideally they could honour the warranty if a part failed due to reasons unrelated, but it's probably hard to tell in many cases.

 

[aarons915]

Let's face it:
- There is no such thing as a "room correction". This wording is as misleading as Tesla's "autopilot".
- You can correct bass response if you know what you are doing. You should do that in any case!
- Automatic full range "correction" does not work and does not sound good IME. I have tried Acourate, Dirac and a Sonos correction (awful).

I can understand - and I like - that KEF does not throw in just another "room correction" algorithm into their products. A manual PEQ (expert mode), however, would be really nice!

Exactly, if they were to support it, it should be automated and up to the transition frequency of 200-300Hz, there is already built in DSP in the speakers which is why they measure near perfect already.

 

[samysound]

Having an option to add something like Dirac or Lyngdorf RP to correct up to 500hz-ish would be amazing

 

[singular]

Manufacturer Nubert offers room correction for its active speakers - even for models priced at €700 apiece. For this, you can use an iPhone’s microphone (since they are standardized) or purchase a USB microphone for Android smartphones for €40.
I would describe the typical Nubert customer as an everyday consumer or someone with a moderate interest in audio.
In my opinion, this is the sort of thing one should expect as a feature these days.

As for the aesthetics... I’m not entirely sure. I like Genelec and the LS50s. The LUXE models, on the other hand, look a bit too egg-shaped. Perhaps a few colored models would have highlighted the shape to good effect.

Nubert's X-Room calibration from 20-160Hz...it is range where you could change mainly perceptual feeling for subbass.
This range in domestic enviroment is always tricky to measure, and listeners almost always prefer to boost it - which is not about quality of bass, nor sound, rather say about windows, street, kitchen, etc.

Also you want to put a shelf in that range, which is not always possible with bassreflex system.

They also mention low cutoff frequency 38Hz which in my language is high pass filter at Fc 38Hz...could be 1st order, 2nd order or more...this means you will get less excursion for bassreflex system below Fb, and perceptualy more subbas in real time.

I cannot see in Nubert's anything more than what does Kef in their IOS app, except ability to use microphone and automatization in that range 20-160Hz.

Nothing special really, maybe more difficult and less control in the end.

 

[markstein]

Nubert's X-Room calibration from 20-160Hz...it is range where you could change mainly perceptual feeling for subbass.
This range in domestic enviroment is always tricky to measure, and listeners almost always prefer to boost it - which is not about quality of bass, nor sound, rather say about windows, street, kitchen, etc.

Also you want to put a shelf in that range, which is not always possible with bassreflex system.

They also mention low cutoff frequency 38Hz which in my language is high pass filter at Fc 38Hz...could be 1st order, 2nd order or more...this means you will get less excursion for bassreflex system below Fb, and perceptualy more subbas in real time.

I cannot see in Nubert's anything more than what does Kef in their IOS app, except ability to use microphone and automatization in that range 20-160Hz.

Nothing special really, maybe more difficult and less control in the end.

That range is the most important to correct in-room, so your comment does not make sense to me.

Edit: I don't know what Nubert x-room is doing - hopefully it is reducing peaks caused by room resonances, using PEQs.

 

[singular]

Luxe reminds me shape of bubble leaving bubble blower.

 

[singular]

That range is the most important to correct in-room, so your comment does not make sense to me.

Edit: I don't know what Nubert x-room is doing - hopefully it is reducing peaks caused by room resonances, using PEQs.

You may be surprised what is enough for your brain to percept good bass, and how easy is to get there.

When brutte force is used to correct measured peaks of highQ in that range, ( 20-160Hz ) then hi-fi is going to hell.

Also that range, mainly from 120Hz down is hugely affected by bulding construction resonances ( walls highQ sharpy peaks and dips ), and by apartment fundamentals.

Also you cannot get exact bass with DC in AC line...could be nice/free/airy...but not exact.

It works the same for passive loudspeakers...You could not get better result of acoustic phase in that range doing so in digital domain...because all you measure in that range is masked in enviroment noise, and measure in top levels of SPL get false gain results.

Companies like KEF know that and use only what is neccessary to keep phase smooth.

Of course it could be done well and worse...it depence on many details.

 

[markstein]

When brutte force is used to correct measured peaks of highQ in that range, ( 20-160Hz ) then hi-fi is going to hell.

I disagree. High Q PEQs to precicely pull down room resonances up to approx. 200Hz are our friend. We have to be careful with high Q PEQs to fill dips though, and avoid filling dips caused by reflections (watch peaks in excess group delay).

Also that range, mainly from 120Hz down is hugely affected by bulding construction resonances ( walls highQ sharpy peaks and dips ), and by apartment fundamentals.

Exactly. Many of these phenomenons are minimum phase and can be corrected with minimum phase PEQ.

Also you cannot get exact bass with DC in AC line...could be nice/free/airy...but not exact.

I do not understand this statement.

It works the same for passive loudspeakers...You could not get better result of acoustic phase in that range doing so in digital domain...because all you measure in that range is masked in enviroment noise, and measure in top levels of SPL get false gain results.

Wrong.

Companies like KEF know that and use only what is neccessary to keep phase smooth.

Phase will not be smooth any more after the sound has passed the room. It can be smooth(er), however, after proper use of PEQ.

 

[Gwen 22]

You may be surprised what is enough for your brain to percept good bass, and how easy is to get there.

When brutte force is used to correct measured peaks of highQ in that range, ( 20-160Hz ) then hi-fi is going to hell.

Also that range, mainly from 120Hz down is hugely affected by bulding construction resonances ( walls highQ sharpy peaks and dips ), and by apartment fundamentals.

Also you cannot get exact bass with DC in AC line...could be nice/free/airy...but not exact.

It works the same for passive loudspeakers...You could not get better result of acoustic phase in that range doing so in digital domain...because all you measure in that range is masked in enviroment noise, and measure in top levels of SPL get false gain results.

Companies like KEF know that and use only what is neccessary to keep phase smooth.

Of course it could be done well and worse...it depence on many details.

Except that the human brain tolerates the phase shift in the low frequencies quite well—up to a certain limit, it's true. As a result, after automatic correction or correction via PEQ, the audio quality is perceived as improved.
Some automatic correction, such as Dirac Live, use FIR filters—at least in part—to prevent excessive phase shift.
It always comes down to the same thing: automatic calibration does lead to a greater improvement than people realize, and the fact that KEF is skipping it in 2026 at this price point is a major mistake.

 

[CB89]

I cannot see in Nubert's anything more than what does Kef in their IOS app, except ability to use microphone and automatization in that range 20-160Hz.

That is exactly what I’m getting at. Automatic correction via microphone. Making it as simple as possible for customers who don’t want to, or can’t, delve too deeply into the technical details. And simply removing the peaks between 20 and 150 Hz.
The icing on the cake would be an "Advanced Mode" to modify the PEQ settings or enter them manually.

The advantage for us customers is that the competition might view it as a selling point and incorporate it into their models as well.