• 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!

Magnepan LRS Speaker Review

NTK

Major Contributor
Forum Donor
Joined
Aug 11, 2019
Messages
2,660
Likes
5,820
Location
US East
Sorry to interfere, but which of these on-axis responses is right?

NFS takes measurements on semi-nearfield and extrapolates these at distance wanted. A tall multiway speaker needs more distance to sum right than a typical shoebox two-way tested here at ASR.

I have no comments about LRS, never seen or heard any Magnepans...
That's why Amir provided this plot.

Magnepan LRS Ribbon Speaker Transition from Near to Far field Mesaurements.png
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,538
Location
Seattle Area
Sorry to interfere, but which of these on-axis responses is right?
"Right" according to what? Here is how CTA-2034 dictates the measurements:

1600715536503.png


By default NFS uses the above formula. I can override it but it is not necessary as you saw in those series of measurements and convergence which NTK showed you above.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,538
Location
Seattle Area
This analogy actually works the other way too. The engine technology can be the same, but the output method will make a substantive difference. Dyno an AWD car on a 2 wheel dyno and you won't get accurate results. Dyno one wheel of a car with an open differential and you won't get accurate results. You'll get *results* and you can probably tell something about the vehicle and the engine and even estimate the power in some respect or another, but you can see why someone would question things. I'm not suggesting your measurements are wrong. I'm just in the camp of people wondering if there's more to measuring different kinds of speakers.
And so if someone told you they are using the right dyno for all car configs, please don't keep questioning it by saying, "I don't understand how cars or transmissions work but surely a dyno can't work right for diesel versus gas engine as they generate power different. Heck one has a spark plug and the other does not!!!"
 

Joppe Peelen

Active Member
Joined
Sep 21, 2020
Messages
119
Likes
100
Location
Den Haag , Netherlands
Hi there im new , and just seen this topic posted and thought damned , thats a must read !!! :)
Its in my ally i must admit :) i make planar magnetic speakers for myself and my Youtube Channel.. and i always wondered why everyone was raving about these. and misnamed them TIme after time as a ribbon... well nothing ribbon here but anyways. you do use some weird measurements i must admit. (you treat them as a line source. witch they are not by far, they are to small :( something in between that creates all kinds of weird things)
another thing is you cant use near field measurements to make a a FR response using big OB speakers. like electrostatics or planars. they are all compensated for rolloff + the fact the mediocre line source still will create much more low end compared to top end. (like far less 16Khz compared to lets say 3khz) i dont know how klippel does compensate for far field.. since it is not a point source and open baffle most common tactics dont work... like minus 6db doubling distance... for a point source... but it is only minus 3 for a line source. and it is a line source for some frequency.. not the perfect one i must admit but stil. if you derive your measurements from close measurements. you might have shown here something that does not resembles a planar or electrostatic speaker of this size ? maybe you did fix that, i just could not find it.

But i did do a tiny victory jump since i noticed the 300-500 hz hump is common even in this version, in there SMGA a speaker that, in my believe is the grandfather this one is coming from.... (same budget to) looks like they just changed out wire for foil and changed housing (less wide) and filtering (but left in the 500 hz hump that states there efficiency.......standing 4-5 db more proud of the rest) dont get me wrong i can enjoy Magnepans for sure even still!. but since i make these kind of speaker myself as well i do know there limitations. and i must say its a fresh breath of air to see someone doing some proper measurements besides the epic high end .... people all raving. without any measurements !! and just say you need a better amp.... since really depending on how loud you play.. i always managed to play my speakers with the cheapest gear a guy can find. 3 or 4 ohm is not all that insane for a normal amplifier unless you use some tube stuff. that is the weirdest choice for these kind of speakers to begin with.

my channel about planar, rubanoides, esl, ribbons etc is https://www.youtube.com/channel/UCeBox1lGM29f72KL2KrDm7A

although i build allot. there is still sooooo much to learn. and i really enjoyed the insane in depth measurements of a commercial planar for a change ! still got some questions about the method used but any how , nice work !!!!!! and hell im no expert !
 

ahofer

Major Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
4,952
Likes
8,698
Location
New York City
Lord knows I've seen it a lot on other forums, but it never ceases to amaze how angry people get at a bad review or bad measurement of equipment they know and like.

I had Maggies (IIIa), and enjoyed them for several years. Then Thiels (CS3.6), now Harbeth (SHL5+AE). All of these speakers have devoted fans and some significant design trade-offs relative to the Olive ideal that is favored here. So what? I don't run off my mouth as if my pleasure (and that of many others') listening to Harbeth proves Olive & Co. wrong. I also 'get' the room experience of Maggies, although they had drawbacks/distractions with some of my favorite listening material that sent me searching for something else.

Even if you are full on subjective, it should still be interesting what might be in your head or your own preferences. You don't have to be in a rush to chase it out just because someone thinks it might be one of those things after some measurements.

Cripes!
 

TheGhostOfEugeneDebs

Member
Forum Donor
Joined
May 21, 2019
Messages
62
Likes
66
And so if someone told you they are using the right dyno for all car configs, please don't keep questioning it by saying, "I don't understand how cars or transmissions work but surely a dyno can't work right for diesel versus gas engine as they generate power different. Heck one has a spark plug and the other does not!!!"

We're not talking about sound output material. It's not about the difference between beryllium drivers vs aluminum and things the machine would never notice. The distinction is that one outputs sound in one direction, the other outputs sound in two directions. You can *say* you have the right dyno for all configurations of engines, but a dyno doesn't measure *spark plugs* it measures output *at the wheels*. So, different wheel outputs = different dyno configurations. It's a totally reasonable question to wonder if the device used for measurement is designed with esoteric output models in mind. It's weirder, in my mind, to *not* question things like that. Nobody is going to make a dyno for my one off 5 wheel off-road concept vehicle with 3 open differentials because that would be a stupid waste, so I'm going to question your magic all-purpose dyno.

I mean, for goodness sake, I'm not even trying to defend the LRS or shit on your review; I've never heard it, don't want one, and will continue to fund your experimentation through patreon. You could very well be totally right: it has no bass and sounds wonky in exactly the way it's measured! But we've got only one panel speaker put to the test. Why am I being mocked for the entirely reasonable curiosity if this is an appropriate methodology for panel speakers - considering they are, after all, a totally esoteric, oddballish and rarer speaker design?
 

Joppe Peelen

Active Member
Joined
Sep 21, 2020
Messages
119
Likes
100
Location
Den Haag , Netherlands
Going above and beyond :D. Results line up with most reviews when reading between the lines. I suspect a lot of the problems emanate from the mid and tweeter sharing the same membrane. Issues kinda shared by coax dynamic drivers.
sharing the membrane is not a big deal. at all. the tweeter plays on the side where there is hardly any movement. and the tweeter most certainly wont move the foil much
 

Joppe Peelen

Active Member
Joined
Sep 21, 2020
Messages
119
Likes
100
Location
Den Haag , Netherlands
There is a fundamental misunderstanding here on the part of people who are arguing that the measurements aren't valid because it is a dipole speaker and the measurement approach doesn't work for dipole speakers.

If a dipole speaker does not perform well when measured as a normal front-radiating speaker, how would placing it in a room and measuring the room response eliminate or compensate for the problems identified when measuring the speaker as a normal front-radiating speaker? The only way this could possibly happen would be if the speaker were set so far back into a deep room that all the acoustic energy radiated into the horizontal space surrounding the speaker mixed together in a homogeneous fashion such that the direct response did not dominate the sound you heard. Maybe if you had a room that was twice as deep as wide and you placed these speakers and one end and you sat back at the other end to listen, this might happen, and it might then make sense to think that the problems that Amir identified would not matter.

This speaker has a wide tweeter and a wide woofer located side-by-side, using a crossover frequency at 1 kHz, where the wavelength is large in relation to the horizontal distance separating the woofer and the tweeter. There just isn't any way that a speaker of this sort will exhibit uniform tonality across the forward radiation space.

Does anyone think that Linkwitz's designs would have the same kinds of problems? I can't think of any reason why they would. To the best of my knowledge none of his designs place the tweeter side-by-side with the woofer and with the crossover wavelength only 1/3 meter. I would be stunned if Linkwitz had ever design a speaker like this, or would have ever considered something like this. I said this before and I'll say it again: For anyone who likes the psychoacoustic effect that Linkwitz strongly preferred, where room reflections contribute greatly to what you hear but do this in a very particular way, you need to look for speakers where the woofer and tweeter are not located side-by-side. It is preposterous for the woofer and tweeter to be placed side-by-side in this manner. It is a total non-starter. Linkwitz understood this perfectly well, and this is fundamentally the reason that he designed his dipole speakers using conventional drivers.

If Linkwitz' Orion or LX521 were measured by Amir using the same setup he used for this small Maggie, the Linkwitz' speaker would in all likelihood measure exceptionally well without the difficulties that Amir encountered in taking measurements of this speaker.

except for nearfield :)
 

Vini darko

Major Contributor
Joined
Jun 1, 2020
Messages
2,280
Likes
3,395
Location
Dorset England
sharing the membrane is not a big deal. at all. the tweeter plays on the side where there is hardly any movement. and the tweeter most certainly wont move the foil much
Hi joppe I'll take your word on this one. Your channel has been slowly encouraging me to have a go making a planar. Good to see you here.
 

Joppe Peelen

Active Member
Joined
Sep 21, 2020
Messages
119
Likes
100
Location
Den Haag , Netherlands
I agree that, in most regards, it's foolish to comment on the sound of something one has never heard. Though there isn't really any reason for you to assume it's the case beyond the negativity of my remark, I indeed have not heard maggies.
But, what's crucial is that I was commenting in context of Amirm's review. While Amirm disliked the LRS, he did notice how different it sounded. I just think it's important to make light of how it's not uncommon in the audiophile world for people to mistake different for better, because they don't really understand what better means. Most tube amps are great example of such. People like hearing the "warmth" of their music through tubes. But, the reality in most cases is that they don't specifically like the distortion of the sound, they like the difference in it, because the same thing over and over again tending to be boring is part of the human condition.

It is not about hearing, the guy you quoted has a really good point. you cant measure huge speakers nearfield.. normally size wise you could... but in this case they are mid/tweeters that are 40 times longer then they would in a semi point source. besides that it is compensated for lows dropping off according to the width of the panel... so you double up on NOT having top end :) does not explain by the way why there is no low end either.....

i do believe i measurements so just listening is not enough to me. and you can measure a speaker and say something about them . if you know how to measure them. and still i dont know maybe he did compensate for all the things i dont know. waiting for an answer. overall it was a really in depth measurement, maybe he made a mistake maybe he did not.
 

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,717
Likes
2,897
Location
Finland
That's why Amir provided this plot.

View attachment 84129

Fine, but I still don't know how to read it. The text says "3m at 400Hz and above" Reliable at 3m for over 400Hz or what does it mean?

Amir said in next post that NFS by default extrapolates response at 2m (and sets dB corresponding to 1m). Well, 2m is less than 3m...hmmm As well the "on-axis" definition is very important as mentioned in earlier posts, wide radiating area and wide separation between sources makes that yes.

I'm fond of dipoles and in my avatar you can see my diy 4-way speaker's horizontal dispersion. I don't even try to make dipole to go below 100Hz! My upper mid and tweeter are B&G magnetoplanars by the way! I love the soundfield and semi-diffuse wide imaging they provide in my symmetric setup along the long wall of the room. When I walk in the room spectral balance remains constant and I must put my ear to only a few inches from the midrange to hear the lateral nulling - brains use longer gating than you would guess!
 

Joppe Peelen

Active Member
Joined
Sep 21, 2020
Messages
119
Likes
100
Location
Den Haag , Netherlands
Welcome to ASR.

There is a thread on the principles behind the Klippel NFS.
https://www.audiosciencereview.com/...nderstanding-how-the-klippel-nfs-works.13139/


Ok so i stopped reading:) major problem i got is... its not a sphere... a sphere indicated a point source and extrapolates from there. a big planar speaker is far from a point source.
its so different in many ways. you cant use a measurement system based on a point source. first of, a point source drops 6 db every doubling of distance an line source does only 3. so your low efficient line source 80dB will be as loud as a 83dB point source at 2 meters. after that the line source will lose 3 dB less over the point source when you double the distance. and that adds up allot !! at 4 meter the low eff line source is 3 db louder...
(if it is a true line source)

true line source is the next problem interfering with measurements derived from a spherical speaker. a line source tweeter measures up close maybe straigh. a little further there is much more low end compared to top end since only the direct sound reaches you and the rest is delayed . some is out of phase and canceled out. so you end up with much more lower frequencies. it also created a a nasty sort of comb filtering if you dont listen exactly in the middle of that semi line source shown in the picture here from Bolsersts (diyaudio) there are 3 line sources. one of 20cm 60cm and 180cm
then on the right you can s ee what happens if you listen on axis or 10 cm up or down, or 20 cm up or down.... a semi line source gives rahter pooo results. a 180cm one does rather nice. so if the LRS are measured like a point source close by you get some major weird results.

picture 2 result of the same 'line sources' but what it does to distance.

so if they where measured near field....... you get the result you see in pic 2


it also makes clear that small planar speakers are far from ideal, maybe cheap but most certainly shit compared to huge ones :) at least long ones
 

Attachments

  • Line_Trend_Height (1).png
    Line_Trend_Height (1).png
    246.4 KB · Views: 89
  • Line_Trend_Distance.png
    Line_Trend_Distance.png
    253.9 KB · Views: 101
Last edited:

KaiserSoze

Addicted to Fun and Learning
Joined
Jun 8, 2020
Messages
699
Likes
592
Quite true - this review reminds me of people trying to "take the laboratory into the field" in certain areas of physiology. It has been done with heat exchange studies, energetics, and endocrinology (in the latter, samples of tears could be taken right after an animal is captured with no harm to the animal and analyzed in the lab).

While I appreciate the work to do these tests, their applicability to use in a room is not known, and likely low. I'd like to see larger panels measured to see the differences too.

The "take it on faith" post is a canard. You need to listen to Maggies. People often take special trips to hear expensive speakers, tho not for ones at this price range. A few years ago I urged Wendell to take a pair of speakers and let customers listen to them for a fee, then ship to the next customer. The large size of maggies is a real detriment to their use and evaluation.

I appreciate the fact that you wrote this in a perfectly polite, mature tone, not expressing indignation over the methodology as others have done. Thank you for that.

I am sympathetic to your perspective, but not entirely so. The reason is that it seems to me that a dipole speaker should measure well when measured as a conventional forward-radiating speaker, and that if it doesn't measure well when treated as a conventional forward-radiating speaker, it isn't going to be a good dipole speaker. Said differently, I don't see how the fact that it is a dipole speaker and was measured as a conventional forward-radiating speaker would make the issues that Amir identified any less problematic than they are. If a dipole speaker is mounted outdoors on a tall pole and measurements are taken of just the front hemisphere, shouldn't it measure much the same as a good forward-radiating speaker? Perhaps the vertical dispersion would be more narrow, perhaps the same would be true for the horizontal dispersion, but the need for smoothness in the off-axis response should apply the same to a dipole speaker as it does to a forward-radiating speaker. But if you were to measure this speaker in this manner, mounted on a tall pole outdoors, you would almost certainly encounter the same issues that Amir encountered, and the tonality in the forward horizontal space would display the same lack of uniformity that Amir discovered.
 

Joppe Peelen

Active Member
Joined
Sep 21, 2020
Messages
119
Likes
100
Location
Den Haag , Netherlands
I appreciate the fact that you wrote this in a perfectly polite, mature tone, not expressing indignation over the methodology as others have done. Thank you for that.

I am sympathetic to your perspective, but not entirely so. The reason is that it seems to me that a dipole speaker should measure well when measured as a conventional forward-radiating speaker, and that if it doesn't measure well when treated as a conventional forward-radiating speaker, it isn't going to be a good dipole speaker. Said differently, I don't see how the fact that it is a dipole speaker and was measured as a conventional forward-radiating speaker would make the issues that Amir identified any less problematic than they are. If a dipole speaker is mounted outdoors on a tall pole and measurements are taken of just the front hemisphere, shouldn't it measure much the same as a good forward-radiating speaker? Perhaps the vertical dispersion would be more narrow, perhaps the same would be true for the horizontal dispersion, but the need for smoothness in the off-axis response should apply the same to a dipole speaker as it does to a forward-radiating speaker. But if you were to measure this speaker in this manner, mounted on a tall pole outdoors, you would almost certainly encounter the same issues that Amir encountered, and the tonality in the forward horizontal space would display the same lack of uniformity that Amir discovered.

:) most dipoles are not as large as a planar magnetic and i think there is where it goes wrong to be honest. if you would not need 4 times the amount of surface area to get the same low end you could measure them up close... if not.... you cant measure them up close. for instance you could measure a dipole woofer nearby if it has 20mm xmax but what if the woofer had only 4 mm ? they might have used 5 of them to achieve the same. but that does mean you cant measure them up close anymore and have a decent picture since it does not resemble a point source very much anymore for allot of frequencies.
 

Joppe Peelen

Active Member
Joined
Sep 21, 2020
Messages
119
Likes
100
Location
Den Haag , Netherlands
Fine, but I still don't know how to read it. The text says "3m at 400Hz and above" Reliable at 3m for over 400Hz or what does it mean?

Amir said in next post that NFS by default extrapolates response at 2m (and sets dB corresponding to 1m). Well, 2m is less than 3m...hmmm As well the "on-axis" definition is very important as mentioned in earlier posts, wide radiating area and wide separation between sources makes that yes.

I'm fond of dipoles and in my avatar you can see my diy 4-way speaker's horizontal dispersion. I don't even try to make dipole to go below 100Hz! My upper mid and tweeter are B&G magnetoplanars by the way! I love the soundfield and semi-diffuse wide imaging they provide in my symmetric setup along the long wall of the room. When I walk in the room spectral balance remains constant and I must put my ear to only a few inches from the midrange to hear the lateral nulling - brains use longer gating than you would guess!

extrapolates :) well thats the thing you cant do that if the thing is based on point sources. they lost some dB there if they did. maybe they compensated i dont know.
 

KaiserSoze

Addicted to Fun and Learning
Joined
Jun 8, 2020
Messages
699
Likes
592
The crux of the matter is the design intent. For any dipole or bipole radiator, the ROOM IS PART OF THE DESIGNER'S INTENDED SOUND. By discarding the room sounds and measuring only the anechoic response, you discard almost half of the expected acoustic. That's why dipole and bipole radiating speakers come with specific instructions as to room placement. If the speakers are to be measured, both the distance from the speaker and the distance from the speaker to reflecting walls, ceilings, and floors are critical. To measure a panel speaker near field or in an anechoic environment is the equivalent of placing a "bookshelf" speaker in a corner and on the floor for measurement.

For any planar-magnetic or electrostatic panel, the width of the radiating surface is also an issue. Although Amirm uses multiple averaged measurements across the panel dimensions, this (again) ignores the designer's intent. Specific peaks and dips in the frequency response ARE inherent in the physical geometry of the panel, but those irregularities are also taken into account (by the designer) in the room placement. To measure the panel characteristics in isolation (ignoring the effect of the room, again) distorts the design intent of the speaker.

And finally, the distortion characteristics of the panel speaker have been measured in isolation. In the case of the LRS, yes, the distortion may rise precipitously above 10 watts, but the average listener uses the speaker at one or less watts. Want lower distortion at higher power levels? Buy bigger panels with more radiating area!

I concede that Amirm's measurements are accurate within their (gross) limitations, but I strongly dispute, due to the way that they were taken, that the measurements are at all useful in evaluating the actual sound of the speaker in situ. Any conclusion drawn from these measurements is a disservice to any reader trying to understand the speakers' sound. You're free to argue this if you wish, but the designer would almost certainly agree with me - the ROOM is an inherent part of the design and any attempt to measure the speaker in acoustic isolation is grossly misleading.

There is some merit and truth to what you have said, but on the whole it does not hold up. In particular, you wrote: "Specific peaks and dips in the frequency response ARE inherent in the physical geometry of the panel, but those irregularities are also taken into account (by the designer) in the room placement." It would be nice if this were realistic, but I just don't believe that it is. Because they are dipole speakers, the radiation to the sides is weak. The radiation to the rear of the speaker (the front of the room) is as strong as the radiation to the front of the speaker, but I just can't see any way that the reflected sound from the wall in back of the speaker is going to even out the response in the speaker's front. For one thing, this couldn't possibly happen unless the speaker were placed very close to the wall behind it, so that the reflected sound from the wall behind wouldn't be perceived as a reflection per se. And whatever that distance is, the comb filtering effect will be strong throughout most all of the treble region, probably the upper midrange as well. The opposite is what needs to happen, i.e., these speakers need to be used in a very large room such that the first reflections are psycho-acoustically separate from the direct sound. And if they are used in this way, it is obvious that the reflected sound is not going to correct the inconsistent tonality in the forward radiation space.
 
Top Bottom