Magnepan LRS Speaker Review

[mac]

There's something wrong with both measurements, just different things. Stereophile's were made with the microphone too close -- you can't do that with a dipole woofer, because you'll think there's too much bass.

You can with near-field, but you just need to expect that you'll see a 6dB per octave increase as you go lower in frequency. This was well understood by Linkwitz and is documented on his site.

 

[Vasr]

Given the massive differences between those two then it does indicate that something somewhere is wrong somewhere....unless it just means that the speaker is widely unpredictable in terms of listening position & placement (& perhaps hence measurement), and therefore that is it's own set of negative attributes for the speaker right there too.

That overstates the case somewhat conflating too many things. I understand where you are coming from and so this is more for others regarding this. They are no more difficult to set up than a typical subwoofer for optimal placement in a room or even traditional speakers in a difficult room. Availability of room measurement systems make this quite simple.

Magnepans are said to be difficult to get a good placement for or are finicky. This one started long before there were any room measurement setups available for general use. People have done all kinds of strange things to optimize it with heuristics to random walks.

When I started using a mic and REW for room correction a while back, things actually became much easier (just like it does for optimal sub woofer placement after measurement setups became available).

For example, the toe-in to get equidistant to the tweeter and mid panel has been more of a thumb-rule even today. But I did find when I started to make measurements as documented here because of an observed anomaly

https://www.audiosciencereview.com/...s-mid-delay-measurements-for-magnepans.15410/

that you can use the delay measurements driving the tweeter and mid panels separately to make that adjustment optimally rather than a trial and error.

In fact, I also found that the delay was considerably affected by the wall geometry behind it as discussed in the above thread requiring asymmetrical toe-in between left and right.

You can do the same thing with frequency response to find a good spot away from the walls. If the room configuration prevents it, determine what you are going to lose with a sub-optimal set up or find an optimal one within the room constraints.

That shows two things, one that these kinds of speakers are very sensitive to what is around them that makes a lot of difference between two measurements but two that does not translate into difficulty in placing them in a room one-time. These are not mutually inconsistent.

Not any more than placing sub-woofers optimally.

The above is not considering whether a particular measurement set up accurately captures what one is likely to hear acoustically from such a dispersion system. That is a different debate. The measurement of speakers aren't the same as the argument one would make for electronics which are simpler to model in terms of their behavior and hence devise a set of measurements. Amir had a whole thread earlier discussing how one would need to measure this, so it isn't some fixed science formula yet.

To the original point, difficult to measure should not be conflated with difficult to place.

Note that amongst all the reviews of other (even traditional) speakers, Amir's subjective perception is not always correlated with the measurements for speakers. There are some open questions still on the relationship between the two for speakers.

This is independent of whether one feels this speaker (or Maggies in general) are good or bad.

 

[amirm]

There's something wrong with both measurements, just different things. Stereophile's were made with the microphone too close -- you can't do that with a dipole woofer, because you'll think there's too much bass. Amirm's were made off the floor -- and you can't do that with a dipole woofer, because they lose all their bass when you lift them off the floor.

Correct about stereophile measurements but absolutely wrong about mine.

The goal of these measurements is to find out how the speaker radiates sound independent of the room. That way, we find the true physical behavior of the speaker. Once there, we can then apply that to our own situations. The Klippel NFS and measurements in anechoic chamber both produce such data and it is absolutely essential in finding design issues in the speaker.

How a speaker behaves in a room is different and both dipole and regular monopole speakers are impacted by the room. That is why we have secondary measurements. Vertical directivity for LRS for example shows razor thin angle where you get anything close to a full range speaker. This was confirmed in my listening tests and no room can eliminate it.

The measurements also predicted how I heard the speaker. Bass was non-existent despite having a real floor contrary to what you stated. Indeed, measurements tell us the effect of the floor reflections just as well:

The blue line tells you the effect of the floor bounce. It provides some enhancement in low frequencies but there is so much of a deficit there that you are still left with a speaker that simply doesn't have any bass.

Predicted in-room response actually includes the effect of the above reflections and produced this:

This absolutely matched my in-person listening as indicated by the EQ making a massive difference in correcting the ills of this speaker. If you listened with EQ, you would want to run away when it was turned off! It compensated for both lack of bass and too little treble.

So no, the measurements are not at fault. The tell the truth and are a great tool in figuring out how to correct the response of this speaker if you wanted to go that way.

 

[Blumlein 88]

Have you ever seen what happens to the response of a dipole when you lift it a few feet off the ground? That's like drilling a hole in the side of a sealed subwoofer. Seriously. The measurement has nothing to do with how they will measure or sound in your living room. It's just wrong.

Here are two response measurements -- compare the bass. Curve 1:
View attachment 83799
Oh my god, way too much midbass, that speaker must sound boomy!

Curve 2:
View attachment 83800

Oh no, no bass at all, *that* speaker must sound really tinny!

Except that they're the same speaker. They're both the LRS. And both were carefully measured by reputable reviewers, John Atkinson in the first case and Amirm in the second.

In both cases, the measurements were invalid, pretty much in opposite ways. Neither has much to do with what you'll hear in your room, which is pretty flat down to 60 Hz.

Comparing the results and methodologies of the two reviews is I think a very informative exercise and again, a lesson in the limitations of measurements in reviews! Compare for example the measurements of vertical dispersion. Totally different scale -- but which *has more to do with what you hear in the room*? I'm guessing Stereophile's, unless you're a spider or a mouse.

Graphs and numbers can fool us into thinking that something is more reliable than it is. How can numbers lie? But if you look at the two graphs above, you'll see that they can, that these two measurements of the same speaker differ from one another more than methodologically consistent measurements of two entirely different speakers would.

I hope everyone who thinks you can "listen with measurements" sees this. That's a dangerous trap to fall into, almost as dangerous as complete subjectivity.

Yes, I've looked at both measurements already. JA typically has a big bass hump on panels which doesn't match what you hear.

What I think about the Klippel, is the principle has been proven to be correct. I'd think unless someone can show otherwise elsewhere that in a real anechoic chamber you get the same results.

I don't think JA or Amir has the right interpretation for how the measurements indicate what the speaker will do in room below 300 hz or so. And that needs some investigation.

OTOH, when tested in a real room at Harman panels haven't fared well. And while perhaps not optimized for panel position the pictures I've seen look in the ball park to be okay. I also don't buy the argument that panels need to be stereo instead of mono. I've been doing Room Correction of one kind or another for about 15 years now. When you correct the measured response of a panel, it does improve the sound as you would expect. So it is not like the quasi-anechoic approach acts weird on panels.

In my mind it is still a conundrum to be solved.

 

[BYRTT]

Based Amir's spindata something to sleep on..

Normalized horizontal line polars Magnespan LRS verse Genelec 8341A, left column is left side right column right side..

Normalized vertical line polars Magnespan LRS verse Genelec 8341A, left column is floor right column cieling..

Normalized horizontal surface polars set to half space Magnespan LRS verse Genelec 8341A, left column is left side right column right side..

Normalized vertical surface polars set to half space Magnespan LRS verse Genelec 8341A, left column is floor right column cieling..

For above battle or measured anechoic reality here their spinorama side by side, curves are, black=on axis, green=listening window, cyan=early reflection, orange=in room response (PIR), blue=power response, red=directivity index, salmon=early reflection directivity index..

 

[josh358]

You can with near-field, but you just need to expect that to see a 6dB per octave increase as you go lower in frequency. This was well understood by Linkwitz and is documented on his site.

Agreed -- you're just adding in the dipole cancellation that occurs when you move beyond the near field. But you have to know to do that, and, unfortunately, 99 out of 100 readers don't!

 

[Vasr]

Correct about stereophile measurements but absolutely wrong about mine.

The goal of these measurements is to find out how the speaker radiates sound independent of the room.

I don't pretend I know anything about these measurements nor will a lot of people be familiar with the methodology.

I am still trying to get my head around how the measurements account for or normalize for the effect of distance from the back wall or alternatively, if they measure (for the metric that is supposed to imply what one hears) as if they were in an open field?

Is the published FR graph specific to a specific distance from the wall or is it good for any distance? It is the latter that I can't get my head around because the tonal balance will change significantly when you move them away or towards the wall. The lower frequencies changes quite audibly based on the distance as any Maggie owner will attest to. What in the measurements reflect this behavior?

Intuitively, these speakers depend on "room gain" so to speak for their final aural characteristic. They have never pretended to be otherwise. So, one can measure without that "room gain" and proclaim them as deficient but is that a good indication of what these speakers are capable of and used as?

The counterpoint to the above would be that a speaker should not be dependent on room topology. But we are not talking about the room angles and symmetry or lack of, just the distance away from the wall which is factored into the design.

 

[josh358]

Correct about stereophile measurements but absolutely wrong about mine.

The goal of these measurements is to find out how the speaker radiates sound independent of the room. That way, we find the true physical behavior of the speaker. Once there, we can then apply that to our own situations. The Klippel NFS and measurements in anechoic chamber both produce such data and it is absolutely essential in finding design issues in the speaker.

How a speaker behaves in a room is different and both dipole and regular monopole speakers are impacted by the room. That is why we have secondary measurements. Vertical directivity for LRS for example shows razor thin angle where you get anything close to a full range speaker. This was confirmed in my listening tests and no room can eliminate it.

The measurements also predicted how I heard the speaker. Bass was non-existent despite having a real floor contrary to what you stated. Indeed, measurements tell us the effect of the floor reflections just as well:

The blue line tells you the effect of the floor bounce. It provides some enhancement in low frequencies but there is so much of a deficit there that you are still left with a speaker that simply doesn't have any bass.

Predicted in-room response actually includes the effect of the above reflections and produced this:

This absolutely matched my in-person listening as indicated by the EQ making a massive difference in correcting the ills of this speaker. If you listened with EQ, you would want to run away when it was turned off! It compensated for both lack of bass and too little treble.

So no, the measurements are not at fault. The tell the truth and are a great tool in figuring out how to correct the response of this speaker if you wanted to go that way.

The problem is that they don't seem to match the response someone else here measured, which is essentially flat to 60 Hz in room. Or Dick Olsher's in room measurements of the MMG, which is the same speaker with wire instead of foil. Also, they don't seem to match John Atkinson's measurements when you correct those by adding in 6 dB/octave dipole cancellation.

I've heard these and believe me, you didn't need or want to run from the room, they had midbass galore. I do know that Wendell Diller told me that they won't work in *large* rooms; dipole bass doesn't couple to a room in a trivial way.

Re the vertical response, you've measured over the entire range, while John Atkinson has measured +15/-10 degrees above and below axis, which I assume is approximately the range from sitting to standing. So the question I would have is which measurement is more meaningful in terms of what one hears? I don't say this because I'm trying to make some kind of point -- I say it because I don't know. It seems to me that the vertical beaming could potentially increase clarity by reducing floor and ceiling reflections. But to the best of my knowledge the main issue here is that the speakers have to be tilted back, and they don't image as well when you do that -- the sound seems to be coming from down near the floor. (Though that was *not* the case at AXPONA, for some reason.)

 

[Blumlein 88]

I don't pretend I know anything about these measurements nor will a lot of people be familiar with the methodology.

I am still trying to get my head around how the measurements account for or normalize for the effect of distance from the back wall or alternatively, if they measure (for the metric that is supposed to imply what one hears) as if they were in an open field?

Is the published FR graph specific to a specific distance from the wall or is it good for any distance? It is the latter that I can't get my head around because the tonal balance will change significantly when you move them away or towards the wall. The lower frequencies changes quite audibly based on the distance as any Maggie owner will attest to. What in the measurements reflect this behavior?

Intuitively, these speakers depend on "room gain" so to speak for their final aural characteristic. They have never pretended to be otherwise. So, one can measure without that "room gain" and proclaim them as deficient but is that a good indication of what these speakers are capable of and used as?

The counterpoint to the above would be that a speaker should not be dependent on room topology. But we are not talking about the room angles and symmetry or lack of, just the distance away from the wall which is factored into the design.

These measurements should be what the speaker does if it were in empty space in all directions or if it were in anechoic chamber. Nevermind the tech behind it, it has been worked out and it can measure in Amir's garage and give you the result if you go and use a full anechoic chamber.

Now there is included directivity index and a PIR (predicted in room response) which does take into account the various reflections in a room. I don't know what size room it models this on.

 

[thewas]

If you want deep bass without a sub and don't care about naturalism, you'll probably get better bang for the buck with a box. If you love acoustical music and value naturalism, planars are a better choice at a given price point.

That's your oversimplified opinion, actually 99,9% of the music you listen and enjoy is mixed and produced with those "non-naturalism boxes".

And most people will hear that right away in an A/B comparison at a dealer's, which is why they're so popular despite being awkward behemoths!

Are they? I know nobody having currently some from my audio friends, except of a friend of mine who sold in the 90s some good large T+A loudspeakers and got a pair of Magnepan being fooled by their audiophile hype and says today it was his biggest mistake buy (he has now a pair of Neumann KH420 "boxes").
Loudspeakers that only sound with specific material good are seriously flawed, which is only ok if the music you listen to is not affected, as the machines should not dictate the art.
PS: I have nothing against planar drivers or open baffles, but only if they are well implemented, like for example the old ESLs were especially considering their time.

 

[Vasr]

These measurements should be what the speaker does if it were in empty space in all directions or if it were in anechoic chamber.

So, if this true, the only thing you can infer from that measurement is don't use them in an open field. And how did Amir hear what the measurement predicted if the listening test wasn't conducted in the acoustic equivalent of an open field? Something fundamentally seems odd here.

Why are people complicating this with all kinds of technical jargon? I think we need to understand intuitively what is going on here and what has been measured.

You can design a speaker that is flat in response in LF and use EQ to cut down on room gain to "hear flat". You can design a speaker to cut down on LF so you "hear flat" with room gain. Maggies seem to be the latter (they came into existence long before any roomeq were available in consumer equipment). We can debate which design philosophy is the right one but that is different from what the measurements are supposed to mean.

I see the reflection from the back wall as a fundamental ("room gain") requirement for using these speakers and fairly consistent with distance from the back wall. Without it you aren't really measuring what you would be hearing but everybody is making that leap in terms of LF behavior.

It is like saying we know how to measure a sealed subwoofer but because the ported subwoofer is different, we will measure with the port blocked so we can compare as to how they will sound. Huh?

What have I got wrong here in interpreting the measurements without getting lost in poles and nulls and axis and verticals and horizontals?

 

[josh358]

That's your oversimplified opinion, actually 99,9% of the music you listen and enjoy is mixed and produced with those "non-naturalism boxes".

Are they? I know nobody having currently some from my audio friends, except of a friend of mine who sold in the 90s some good large T+A loudspeakers and got a pair of Magnepan being fooled by their audiophile hype and says today it was his biggest mistake buy (he has now a pair of Neumann KH420 "boxes").
Loudspeakers that only sound with specific material good are seriously flawed, which is only ok if the music you listen to is not affected, as the machines should not dictate the art.

OK, two different issues here.

I've worked in pro audio, and studio monitors are designed for a purpose. They aren't the most realistic transducers you'll ever hear, and I don't think most would want to listen to them at home. Similarly, planars will never be popular as studio monitors since they're SPL limited and big, and since you want to minimize early reflections -- and in recent years, because the industry has shifted to near field monitoring.

I'd have to disagree with "Loudspeakers that only sound with specific material good are seriously flawed." You can't make a loudspeaker sound good with good material and good with bad material at the same time. You just can't, because they have a massively different spectral balance and distortion characteristics, so if a speaker sounds flat with one, it will sound way off with the other. The speakers I like sound good with good recordings and bad with bad ones, though I can understand the appeal of something that sounds mediocre with both (which is basically what you get when you compromise here -- an unhappy average).

 

[GDK]

the only thing you can infer from that measurement is don't use them in an open field

According to the video posted earlier today by @Doug Schneider Axiom test their woofers on top of a 100 foot pole.

https://www.audiosciencereview.com/...ow-soundstage-began-measuring-speakers.16086/

 

[thewas]

I've worked in pro audio, and studio monitors are designed for a purpose. They aren't the most realistic transducers you'll ever hear, and I don't think most would want to listen to them at home.

I know more and more people (locally but also in this forum and other forums, including myself) using studio monitors and saying its the best sound reproduction they ever had. We shouldn't forget that music is mixed and produced with monitors and is mixed to sound nice as the target is to sell as much as possible,

I'd have to disagree with "Loudspeakers that only sound with specific material good are seriously flawed." You can't make a loudspeaker sound good with good material and good with bad material at the same time. You just can't, because they have a massively different spectral balance and distortion characteristics, so if a speaker sounds flat with one, it will sound way off with the other. The speakers I like sound good with good recordings and bad with bad ones, though I can understand the appeal of something that sounds mediocre with both (which is basically what you get when you compromise here -- an unhappy average).

I didn't mean that, as like you say the idea of hifi is to reproduce a poor recording sounding poor.
I mean loudspeakers that cannot reproduce some frequency regions or due to wild dispersion patterns make the own "art" at the reproduction.
Of course if final consumers like this (like also many like their TV colours and contrast turned to max) this is ok for them but the original idea of High Fidelity can be best captured by the words of Floyd Toole:
Audio is engineering, music is art.

 

[Deleted member 14468]

According to the video posted earlier today by @Doug Schneider Axiom test their woofers on top of a 100 foot pole.

https://www.audiosciencereview.com/...ow-soundstage-began-measuring-speakers.16086/

It's true. They use it for calibration for low frequencies for the chamber, too. There's a picture in the video.

Doug Schneider
SoundStage!

 

[josh358]

So, if this true, the only thing you can infer from that measurement is don't use them in an open field. And how did Amir hear what the measurement predicted if the listening test wasn't conducted in the acoustic equivalent of an open field? Something fundamentally seems odd here.

Why are people complicating this with all kinds of technical jargon? I think we need to understand intuitively what is going on here and what has been measured.

You can design a speaker that is flat in response in LF and use EQ to cut down on room gain to "hear flat". You can design a speaker to cut down on LF so you "hear flat" with room gain. Maggies seem to be the latter (they came into existence long before any roomeq were available in consumer equipment). We can debate which design philosophy is the right one but that is different from what the measurements are supposed to mean.

I see the reflection from the back wall as a fundamental ("room gain") requirement for using these speakers and fairly consistent with distance from the back wall. Without it you aren't really measuring what you would be hearing but everybody is making that leap in terms of LF behavior.

It is like saying we know how to measure a sealed subwoofer but because the ported subwoofer is different, we will measure with the port blocked so we can compare as to how they will sound. Huh?

What have I got wrong here in interpreting the measurements without getting lost in poles and nulls and axis and verticals and horizontals?

FWIW, the reflection off the front wall (the one behind the speakers) is actually deleterious, both for planars and dynamic speakers, because the reflected wave cancels with the direct wave creating a null. This happens at different frequencies for a planar and a dynamic. In a planar, the null will typically be at about 125 Hz, depending on position.

What is crucial is that the speaker be sitting on the floor. A dynamic speaker will sound OK on a stand; a planar will lose all its bass. I won't go into the technical details (two pi, four pi, dipole cancellation, yada). The point is that if you don't measure the speaker as it interacts with a real room, you won't measure what you hear.

The size of the room also matters. A bigger room requires a planar with a bigger baffle. This is because of the way a dipole couples to the room. There is no "one size fits all" solution here, since a planar that sound right in one size of room will sound wrong in another. The LRS is tuned for smaller rooms.

OK, so flat vs. curved. As you know, two channel stereo requires a "house curve," e.g., a diminishing response to sound flat. Some speakers are designed to be flat and the curve is added with equalization, some are designed with the equalization built in. The Maggies are of the second type -- they're designed to sound flat without EQ (though Amirm says they fail in that). Which I think is what you were saying.

 

[Deleted member 2944]

Just for reference for those that can't grasp the inherent testing issue here. The proper way to characterize the bass response of a speaker like this is to tip it over on the side and make a ground-plane measurement from a couple meters away. THAT will yield a measured response that is a pretty good indication of how it will perform in normal usage......bass wise.
I've used that approach for many years testing planar speakers.

For those that are still questioning the results here.....the basic speaker here (the MMG/LRS) has been shipping since 1997. Earlier versions of a smallish Magnepan speaker have been shipping since well before that.
There are literally thousands of pairs of these in the field all over the world providing enjoyable music reproduction. You might want to consider that before labeling these speakers "not a product finished and fit for use by a consumer."

Dave.

 

[josh358]

I know more and more people (locally but also in this forum and other forums, including myself) using studio monitors and saying its the best sound reproduction they ever had. We shouldn't forget that music is mixed and produced with monitors and is mixed to sound nice as the target is to sell as much as possible,


I didn't mean that, as like you say the idea of hifi is to reproduce a poor recording sounding poor.
I mean loudspeakers that cannot reproduce some frequency regions or due to wild dispersion patterns make the own "art" at the reproduction.
Of course if final consumers like this (like also many like their TV colours and contrast turned to max) this is ok for them but the original idea of High Fidelity can be best captured by the words of Floyd Toole:
Audio is engineering, music is art.

I've been retired for some time and I haven't heard the most recent generation of monitors. The ones with which I was familiar had much better dynamics than most consumer speakers -- they played loud without a hint of stress -- but inferior to the best consumer speakers in other regards. I like to think that they've improved.

I don't think much art went into the design of the LRS -- it's engineering, pure and simple, with just a touch of dark art. What they have, as with any speaker designers, is an understanding of what matters to what they're trying to achieve and what doesn't. This is not something you can get just from measuring some dynamic speakers or from a textbook -- the relationship between measurement and sound is a lot fuzzier than some believe, particularly when you step beyond the well-known world of dynamic speakers.

But, you know, ask someone what planars sound like and they'll say "amazing sense of space and detail without a boxy sound." Also, "the bass sucks" or something like that. And "amazingly natural." And that's basically what they *do* sound like. They do some things worse than dynamics and some things better, and, in general, they have better bang for the buck because they're cheaper to make.

 

[ruinevil]

In my room, the LRS has a hump starting at 100 Hz with a 6dB peak at 500 Hz, that tapers down around 1000 Hz... where the directivity problems start. No problems with midbass at all... it's all in the sub-bass and treble.

 

[SKBubba]

I don't know about all this, but Audyssey set the crossover at 40hz for my nearly 30 year old Klipsch Quartets. To quote Seinfeld, they're real and they're spectacular.