A new Kef speaker? The Kef LS Luxe

[Gwen 22]

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.

I agree with you on that.

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

From my perspective, that's already the norm—most competitors already offer it and are highlighting it!

 

[singular]

HighQ ripples in range below 120Hz corrected by highQ IIR means cheesy bass.

Of course you can use FIR with smooth phase, but the measured input is always false as I mentioned...thatswhy I do not see a sense to do that.
But that is my philosophy.

I believe your bussiness will run smooth, guys.

 

[markstein]

Some automatic vorrevtion, such as Dirac Live, use FIR filters—at least in part—to prevent excessive phase shift.

AFAIK, Dirac uses mixed correction filters (both minimum phase and linear phase) to manage latency. I would assume that bass is mostly corrected using minimum phase.

Notice that a high Q minimum phase PEQ exactly compensating a minimum phase room modes does NOT cause additional ringing - on the contrary. Floyd Toole shows a real world example in his book "sound reproduction".
What should be avoided, however, is imprecise bass EQ, e.g., done with graphic equalizers.

 

[markstein]

HighQ ripples in range below 120Hz corrected by highQ IIR means cheesy bass.

No, this is wrong.

 

[singular]

I am curious what my favourite reviewer say about bass quality of Luxe, especialy as Kef fan with blades.
Comparing on youtube channel one could assume it is little bit different phase topology...almost like vented vs. closed enclosure, or something like that.

 

[Gwen 22]

Interestingly enough, while we're debating here whether KEF's decision to do without PEQ and automatic correction is a good one or not, Buchardt has come up with this to improve his active speakers.

MK2 Active Speaker Upgrade

Buchardt MK2 Upgrade Kit for A500, A500 SE, A700, A700 LE & A10

buchardtaudio.com

Among the improvements, note the new DSP with PEQ and automatic corrections!

Just to be clear, I'd like to point out that the Buchardt Anniversary 10 can be considered a direct competitor to the KEF LS Luxe.

 

[kemmler3D]

Interestingly enough, while we're debating here whether KEF's decision to do without PEQ and automatic correction is a good one or not, Buchardt has come up with this to improve his active speakers.

MK2 Active Speaker Upgrade

Buchardt MK2 Upgrade Kit for A500, A500 SE, A700, A700 LE & A10

buchardtaudio.com

Among the improvements, note the new DSP with PEQ and automatic corrections!

Just to be clear, I'd like to point out that the Buchardt Anniversary 10 can be considered a direct competitor to the KEF LS Luxe.

Interesting.

I wonder how much of a factor the dealer / retailer network is in this.

To some extent KEF's support is handled by retailers. KEF support might be easily train-able to handle PEQ questions, but what about the hodgepodge of companies and people that sell KEF?

AFAIK Buchardt is more DTC but I'm talking out of my ear at the moment.

I am still assuming the support burden is a major question for KEF, but it could also be that they have bigger engineering fish to fry, dunno.

 

[Gwen 22]

Interesting.

I wonder how much of a factor the dealer / retailer network is in this.

To some extent KEF's support is handled by retailers. KEF support might be easily train-able to handle PEQ questions, but what about the hodgepodge of companies and people that sell KEF?

AFAIK Buchardt is more DTC but I'm talking out of my ear at the moment.

I am still assuming the support burden is a major question for KEF, but it could also be that they have bigger engineering fish to fry, dunno.

In my opinion, the key question is whether KEF has sufficiently modernized the electronics in the KEF LS Luxe compared to the six-year-old LS 50W2—with an eye toward a potential platform "upgrade"—or whether they've kept the electronics largely the same in order to save money.

In the second case, it's highly likely that they are technically stuck, but if that's the case, they will never admit it!
In fact, admitting that they have to use electronics that meet standards from six years ago—in a high-end model, no less—would be a major communications blunder.
I'm not saying that's the case—I don't know—just that if it is, they won't say so.

 

[kemmler3D]

In my opinion, the key question is whether KEF has sufficiently modernized the electronics in the KEF LS Luxe compared to the six-year-old LS 50W2—with an eye toward a potential platform "upgrade"—or whether they've kept the electronics largely the same in order to save money.

In the second case, it's highly likely that they are technically stuck, but if that's the case, they will never admit it!
In fact, admitting that they have to use electronics that meet standards from six years ago—in a high-end model, no less—would be a major communications blunder.
I'm not saying that's the case—I don't know—just that if it is, they won't say so.

IME you can always trot out the "We maximize quality by not building in anything unused or unneeded - to focus on sound quality and give you the most bang-for-buck". Always worked for me.

It also tends to be true. If the LS50/60 DSP still works up to the bar they're aiming for, why would they put in more expensive parts on the off chance they ship a firmware upgrade someday?

 

[NTK]

How could KEF implement VECO (velocity feedback control of the woofer) without some significant reworking of the electronics?

 

[kemmler3D]

How could KEF implement VECO (velocity feedback control of the woofer) without some significant reworking of the electronics?

I think VECO is analog so if you're just talking about the digital stuff, that could be the same as the older models.

 

[Gwen 22]

IME you can always trot out the "We maximize quality by not building in anything unused or unneeded - to focus on sound quality and give you the most bang-for-buck". Always worked for me.

It also tends to be true. If the LS50/60 DSP still works up to the bar they're aiming for, why would they put in more expensive parts on the off chance they ship a firmware upgrade someday?

The DSP doesn’t have to be more expensive; for the same price, a 2026 chip performs much better than a 2020 chip, and let’s not forget that there is, after all, a substantial price increase when it comes to the speaker.
I think we can all agree that the KEF Connect app’s user interface is rather dated (it takes me back to the 90s when I open it – just after MS-DOS).
Whilst this may well be acceptable for a €2,500 speaker released in 2020, for a €4,000 speaker released in 2026, failing to allow for sufficient power headroom to accommodate future upgrades to the platform would be a mistake.

 

[NTK]

I think VECO is analog so if you're just talking about the digital stuff, that could be the same as the older models.

From this post by Dr Jack, I doubt VECO is a simple analog feedback system (e.g. PID. But who is doing that these days anyway? Even PID controllers are done digitally.). In his post he talked about state variable monitoring and comparing sensor readings to system/driver models, which suggests KEF is employing some kinds of state-estimators, e.g. Kalman filters, and probably in something like LQR control, which can only be done (practically) digitally.

Regarding the software on the streaming side or EQ, they most likely are just reusing the common stuff across different platforms.

Yes, that's right. We, and many others in the industry, have products that have a DSP model of the driver+cabinet and then calculate feed-forward correction to reduce the distortion in the radiated sound. The approach works quite well but there are some aspects of the driver behaviour that are difficult to predict. Some such systems are fixed at production and others are adaptive and monitor a state variable, such as voice coil current, and minimise the error between the internal driver model and the real behaviour. One of the motivations for the sensor was to superseded this approach, although the sensor can also be used to augment this approach by providing another state variable to the model.

 

[Gwen 22]

From this post by Dr Jack, I doubt VECO is a simple analog feedback system (e.g. PID. But who is doing that these days anyway? Even PID controllers are done digitally.). In his post he talked about state variable monitoring and comparing sensor readings to system/driver models, which suggests KEF is employing some kinds of state-estimators, e.g. Kalman filters, and probably in something like LQR control, which can only be done (practically) digitally.

That’s interesting, I didn’t know that – so who’s in charge of it, the DSP?

Regarding the software on the streaming side or EQ, they most likely are just reusing the common stuff across different platforms.

To be honest, I really hope not – 2020 is a long way off in the world of electronics!

 

[Colonel7]

Erin H. Has a pair to review. Not sent by KEF

 

[Goldhamster916]

Erin H. Has a pair to review. Not sent by KEF

Niiiice

 

[Gwen 22]

Erin H. Has a pair to review. Not sent by KEF

Yesss !!!! That’s good news!
Do you know how long it usually takes him from receiving the speaker to the publication of his review?

For those of you who might think this question shows a certain impatience... Well, you’re right

 

[Rõlnnbacke]

From this post by Dr Jack, I doubt VECO is a simple analog feedback system (e.g. PID. But who is doing that these days anyway? Even PID controllers are done digitally.). In his post he talked about state variable monitoring and comparing sensor readings to system/driver models, which suggests KEF is employing some kinds of state-estimators, e.g. Kalman filters, and probably in something like LQR control, which can only be done (practically) digitally.

Regarding the software on the streaming side or EQ, they most likely are just reusing the common stuff across different platforms.

In this video, Mr. @pthakkar24 says it stays analog:

If I'm not mistaken, the sensor is actually part of a magneto (best known in the form of a bicycle 'dynamo'). Maybe the current that it generates is near perfect enough analog to the voice coil movement to keep it analog. So a model would't be needed for that. It would be interesting to know if it would be beneficial if possible irregularities of the rest of the driver would be compensated for with a digitalized feedback signal, with the models you mentioned. Maybe Dr Jack means something like that in the last sentence of the quote.

I am not an expert at all, so maybe, for experts, I stated the obvious or missed the point.

 

[kemmler3D]

In this video, Mr. @pthakkar24 says it stays analog:

If I'm not mistaken, the sensor is actually part of a magneto (best known in the form of a bicycle 'dynamo'). Maybe the current that it generates is near perfect enough analog to the voice coil movement to keep it analog. So a model would't be needed for that. It would be interesting to know if it would be beneficial if possible irregularities of the rest of the driver would be compensated for with a digitalized feedback signal, with the models you mentioned. Maybe Dr Jack means something like that in the last sentence of the quote.

I am not an expert at all, so maybe, for experts, I stated the obvious or missed the point.

OK, I knew I heard that somewhere, at least I am not crazy. I mentioned it because I thought it was suprising in 2026 to have an analog control system of any sort, but there you have it. Maybe there is also a digital component... in their marketing, I think KEF does try to walk the tightrope between taking advantage of modern tools, and giving something to comfort the retro-subjectivist-analog-is-sacred crowd.

 

[Rõlnnbacke]

OK, I knew I heard that somewhere, at least I am not crazy. I mentioned it because I thought it was suprising in 2026 to have an analog control system of any sort, but there you have it. Maybe there is also a digital component... in their marketing, I think KEF does try to walk the tightrope between taking advantage of modern tools, and giving something to comfort the retro-subjectivist-analog-is-sacred crowd.

Some of them would know that this speaker can't be 100% analog anyhow.

It could be that the ls luxe is 'perfect enough' for now, they already have the veco-patent and are busy developing new products. Like veco subwoofers. Combined with that Unicore(?) system it would quite complicated.. Or a large speaker with veco, possibly with a port tuned at 20 Hz!