• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Linkwitz LX521.4 Review (and measurements!)

MKR

Major Contributor
Joined
Oct 18, 2022
Messages
1,478
Likes
2,160
Location
USA
While certainly not equal to the level of measurements Amir would perform, something better than nothing …


I have extensive experience both with older versions of SL’s designs (including owning the original Orion years ago) and this latest incarnation and they are outstanding loudspeakers. Not many speakers I have heard portray as holographic a soundstage as these do. The only reason they are not on my endgame shortlist is the output falls a bit short for larger rooms.
 

Count Arthur

Major Contributor
Joined
Jan 10, 2020
Messages
2,276
Likes
5,103
At first I thought this looked pretty rough:

1705495189331.png


But they've made the X axis cover only 20 dB, which is exaggerating the peaks and troughs.

Setting the X and Y limits similarly on my own in-room measurements makes them equally lumpy:

1705495489388.png


It looks far better like this: :)

1705495961482.png


Regardless, I've been aware of the Linkwitz speaker designs for some time and I would very much like to hear some. However, open baffle designs are unlikely to work well in my room. :confused:
 
Last edited:
OP
M

MKR

Major Contributor
Joined
Oct 18, 2022
Messages
1,478
Likes
2,160
Location
USA
At first I thought this looked pretty rough:

View attachment 342509

But they've made the X axis cover only 20 dB, which is exaggerating the peaks and troughs.

Setting the X and Y limits similarly on my own in-room measurements makes them equally lumpy:

View attachment 342511

It looks far better like this: :)

View attachment 342514

Regardless, I've been aware of the Linkwitz speaker designs for some time and I would very much like to hear some. However, open baffle designs are unlikely to work well in my room. :confused:
Nice catch on the plot axis!

And yup, these (as with all dipoles) definitely need some room to breath
 
Last edited:

Zambo

New Member
Joined
May 16, 2024
Messages
2
Likes
2
Here is a psychoacoustically smoothed response graph from my listening space. Definitely some issues going on here. Not sure if it's room related, or system related.
I have the latest version of the LX 521.4 MG, being driven by 10×250 W per channel NCore amplifiers, and the latest analogue signal processor. My front end is a benchmark DAC3/ HPA4 being fed by various sources, mainly a MacBook computer for music streaming. It's strictly a two channel system, which I use for music and home theatre. it's a very large open space with ceilings that peak at 15 feet. The speakers are about 14 feet apart, and listening position is about 14 or 15 feet from the speakers. All connections are fully balanced, XLR cables. Speaker cables are Canare star quad 4S11.
I've also included a couple of pictures of my listening space. The home theatre screen drops down from the beam above the fireplace.
I would love to see more room response curves from other LX 521 owners. Our better still have a pair tested on this site
My experience is that they sound incredible on certain music, but not all.
When they are working well, and on the right kind of music, nothing I've heard gives a more magical illusion of the live event.
But it definitely seems to be missing something in the bass/mid bass/mid range area and is a little bright in some of the higher frequencies.
I'm currently looking into a system, which has a more complete and even frequency response, yet at the same time can create that magical illusion of live sound that Sigfried's dipoles have always been great at. I originally built his Orion speakers, which I still feel in some respects, sounded better than the 521's
The cardioid designs in particular have piqued my interest


Screenshot 2024-05-16 at 6.16.23 PM.png
IMG_4771.jpeg
IMG_4776.jpeg
 

ShadowFiend

Member
Joined
Dec 20, 2020
Messages
74
Likes
93
This is the worst way to place LX521 in a room: no toe in, no distance from front wall and big objects near the side.
The place for LX521 in your room and the toe-in angle should be like this picture
IMG_4771.jpeg
 

Zambo

New Member
Joined
May 16, 2024
Messages
2
Likes
2
Yes, I do pull them out for listening. I know Sigfried recommended at least 3 feet from the front wall, but realize many 521 owners bring them much further into the room, which doesn't have a very appealing WAF lol, which is very much at play in my situation. I did pull them into the room for the measurements that I made. That being said, I don't think one should have such wild frequency/SPL response fluctuations.
 

kemmler3D

Major Contributor
Forum Donor
Joined
Aug 25, 2022
Messages
3,573
Likes
7,308
Location
San Francisco
@Zambo welcome to ASR and thanks for sharing! Your house reminds me of a really charming AirBnB I stayed in near Portland, ME one time. We all do what we can to optimize the WAF vs Sound Quality equation.

That peak in the bass (80 to ~150hz) looks like absolute murder, the rise from 200 to 2khz and the drop from 2khz on could probably also benefit from EQ. Run, don't walk, to your nearest DSP box. :)

Does it look this hair-raising if you do an MMM / Pink noise measurement? I have found that measuring that way often shows good agreement with sweeps, but where they disagree tends to agree with what I hear more.
 

AudioJester

Addicted to Fun and Learning
Forum Donor
Joined
Jan 7, 2020
Messages
962
Likes
1,295
The bass peak is a room mode/issue.

That dip 100-300Hz is almost universal in all measurements I have seen of the LX521, even with the new midrange driver which does not address the problem. This area may be why many people dont like the Lx521 for more "aggressive" types of music.

The problem is the passive crossover network. I use pc - okto 8 dac, but any multichannel dsp eg. Minidsp will be a better solution than the passive controller. Shifting the crossover from the bass bins from 110Hz to 250-300Hz makes a massive difference. And you have better control of drivers and crossovers.
 

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,728
Likes
2,917
Location
Finland
Multichannel DSP is needed to match acoustic delays of signals from each channel. This should be done outdoors without room reflections, but including floor in case of a floorstanding speaker - mic located on ground at 3 meter or more, speaker tilted. indoor at 1m is ok above 400Hz.

Dip between 100-300Hz is often the first floor reflection, a quite wide single dip. It is best seen with 12ms right window gating. It should not be equalized, because it is impossible to do and positive peak eq only makes sound very boomy/chesty. We don't hear that dip like the mic, because our brains have eq setting for it!

Room measurements should be done individually from each speaker and from multiple mic locations. Then start moving the speaker too to see how response changes!
 

AudioJester

Addicted to Fun and Learning
Forum Donor
Joined
Jan 7, 2020
Messages
962
Likes
1,295
image003-3.jpg


This is an old measurement of my lx521. Excuse the green marker, I was trying to set the levels for divers - you can see i didnt have it quite right and the bass level is a bit low - amp gain issue which was easy to fix.
This is raw driver measurement at listening position (3m) in an accoustically terrible room. You can see the room effect below transition zone. The midrange drivers and tweeter looked pretty good to me, but the midrange output below 300Hz is lacking, the bass drivers easily handle this region.

I initially thought that dip was floor/ceiling bounce - but i could not alter it with any form of treatment.
 

JeyB

Member
Joined
Jul 8, 2022
Messages
20
Likes
18
Hi @Zambo

Attached lately some LX521.4MG in-room measurements at my Linkwitz Lounge Santiago Instagram. Take an eye on it if you wish to see how measures my system.


My room is purely orthogonal and this makes a big difference. You seem to have some adjacent open spaces which can act as low frequency resonators and this make for uneven response. I've tested my Linkwitz LX521 in three different rooms during eight years of ownership and no problem in any of those rooms. But they all were orthogonal and closed, ie, rectangular floor plan. If you look through my IG, you may see there is a small entrance to my actual listening room. I always close that space when listening. If not, the low frequencies start to resonate and create a pronounced rising in RT60 times.

In my experience, low energy in 200 Hz region is due to higher than recommended listening position. Try to place the mic lower. For room modes you may rotate individually the LXsub4 and listen. Take a look at my disposition, quite unconventional, but almost a straight line in frequency response up to 130 Hz. No room modes as a result.

My Y axis measurement is calibrated in 10 dBs increments, while yours is calibrated in 1 dB increments. This makes for a visual bigger roughness in your graph. I don't usually measure, unless investigating something, but when I measure I look for tendencies in frequency response and not for correcting some peaks, for sure only below Schroeder frequency. We all listen with our ears and not with our eyes. If you perceive acoustic inconsistencies then you should investigate them. But I suggest you to audition a properly set up LX521 system before any drastic decission. If you allow me to express my humble opinion it is something unique and unequalled.

Best regards
 
Last edited:

MKWingchun

New Member
Joined
Jul 22, 2023
Messages
2
Likes
1
I have DIY Open baffle speakers. They are 5 way, using the 4 way Linkwitz's as a basis (mine are a 12, 8, 6, and a 1, plus a sub on each side) for comparison. Moving the speakers out into the room about 3 feet is sufficient to produce a very good soundstage. I put my speakers on furniture sliders, so they can be moved back and forth, from the wall to out in the room. Having speakers in the middle of the room is inconvenient. That said, when I saw a Linkwitz demo about 15-20 years ago, Linkwitz put them a third of the way in the room, for optimal placement. Really smart nice guy, btw, and visiting his website is a must if you are interested in open baffle speakers.
 

JeyB

Member
Joined
Jul 8, 2022
Messages
20
Likes
18
Hi again @Zambo

I've been thinking a little more about your particular case. 14 feet speaker separation is way too much for proper phantom scene recreation. So I would recommend you to adhere to 7 to 9 feet distance suggested originally by SL. 15 feet to each loudspeaker is also way more than optimum. Please, try 6 to 8 feet and for sure the holographic properties will begin to show adequately. The distance to front wall is what defines the depth properties of the LX521. I mean that if your distance is 4 feet you will have approximately 4 feet phantom scene depth. Consider my own set up, I have 8 meters (26 feet) to the front wall, so imagine how can it sound when all that space to the front wall is populated by multiple virtual sound sources corresponding to instruments or voices. This is what is unequalled by any other speaker I have experienced: the ability to fill the room at any depth and at any width, and even height perception, with pin point virtual sources corresponding to the placement of the microphones around instruments in the recording venue. I purposely quoted height perception because the LX521 precision is of such caliber that it allows to clearly distinguish in a duet if one singer is taller than the other one singing to his sideway. And this has been corroborated by more than one astounded fellow at my listening room.

On the other hand, think about this, the closer the LX521 is to your ears (within limits) the more you are increasing the direct to reverberated ratio. With this transducer is of utmost importance to set up distances long enough to allow the precedence effect to take place (Haas effect), thanks to the fact that the reflected sound is a spectral replica of the direct sound. The bigger those distances the easiest the listening room acoustics to disappear from the acoustic horizon. When this happens, then the acoustics of the recording venue superimposes the local acoustics and dominates the listening experience. So, in consequence you only perceive the encoded acoustics in the recording and the teletransportation trick magically happens even if you don't desire it to happen. If the encoded acoustics corresponds to a moderately dead ambience (like the typical recording studio) then the musicians seem to be placed on your room at different positions, but enveloped with the studio acoustics, which may be more or less similar to that of the listening room. But when the encoded acoustics corresponds to a symphony hall, much different to the listening room, then your room disappears and you are placed psychoacoustically in the original venue with all the sound sources in front of you, spread over a three dimensions space.

In your actual situation with your LX521, the measurement is what it is, no matter which speaker you use. It is probably not a problem of the speaker itself but a very gross setup problem. When the direct sound of the speaker arrives to your ears, the reflected waves probably have arrived earlier and this is suboptimal in every sense. You should register RT60 measurement, spectrogram and decay times also.

Recently I have assisted to two demos, one of ATC SCM100 and another one to Kef Blade One Meta. The result, I couldn't be most dissappointed. I really thought the Blade was going to blow me and reveal like an endgame speaker because I have watched Erin's review and the magnificent Klippel measurements, so I felt predisposed to be blown away. What a horrible sound it produced. I felt very annoyed by such a stomach revolving bass and ultra detailed and thin high frequencies. Any resemblance to a real instrument playing in the room was absolutely absent, let alone there was basically no soundstage anywhere to talk about. But people around me seemed to like it a lot (!). This leads me to think that people should go more to real live events at symphony halls, be it symphonic, soloist, lyrical or chamber music. If that's not your favourite musical genre neither acoustic music, well, then why a LX521?

This situation also makes me question what is the excellence in measurements valid for. There is no way to measure how a set of transducers will excite the innards of the ear-brain cognitive system. So, this refutes the theory that only listening on a periodic manner to acoustic live music keeps your database to date and positively allows the comparison of reproduced sound to that elusive reality. Only honestly recorded music is the recipe to feed a system like LX521 and be rewarded with true realism at your home. Feed it with Dire Straits, Depeche Mode, The Cars, Michael Jackson and almost any similarly recorded content, like in Erin's Spotify playlist for example, and you will only receive garbage sound. I'm not talking now about musical qualities, only about sound qualities. It's necessary some kind of sensibility to talk about and listen to music, not just "listen to that rasp-rasp-rasp sound how distinctly it sounds". Just a reconsideration, from here on out I will stop watching those crappy YT channels, no matter anechoic data presented or not. I will employ the time to listen to more music instead.

The LX521, and I will add as is set up in my room, is the only system I have heard to date which is capable to completely reconstruct a coherently recorded voice and piano in my room, with the absolute credibility of a real singer and a real pianist performing in front of me at about 8 m away from the listening position. It's even more, it let me feel being in the hall with the artists. Please, play this short song and discover if your system is up to the task.


Maybe you own a gem and you still don't know it.

I would only take the in room measurements as a guide to optimize the positioning in the room, not to alter its response. SL solved his part.

Best regards
 
Top Bottom