Magnepan LRS Speaker Review

[Burning Sounds]

First of all a big thank you to Amir for a Herculean effort here! And to the owner for sending them in.

As a 30+ years Magneplanar owner some of the results such as the lobing and inconsistent imaging are what I would have expected. I still have a pair of what are the grandaddy of these - MG1s - they are bass shy, and if you stand up the highs disappear - so much for progress! However, I suspect @dreite is correct in his assessment of the poor bass measurements of the LSR - it doesn't really make sense that they are worse than my ancient MG1s.

I also have MG2.5Rs which are completely different with a true ribbon that comes in at 1Khz. Mine originally had a 2nd/3rd order x-over, not a first order as Stereophile's review states. If Stereophile are correct on the unit they measured, then I suspect Magnepan may have changed from the first order x-over as soon as they got reports of destroyed ribbons. Mine are currently active with a LR4 crossover which provides much more protection for the ribbon. With the addition of a Linkwitz dipole bass they can go plenty loud with rock music and provide a real sense of scale - The Who live at Leeds at 105dB peaks sounds pretty good to me. Expensive amps are most certainly not required as long as they provide plenty of current. And they can make less than stellar recordings sound quite pleasant. But, yes they have their flaws - images can be way too large - drummers with 10 foot arms comes to mind. Images can stick to the panels which I find annoying. But they cost me £500 12 years ago. I can't think of any box speaker that comes close at that price. Here's an in-room distortion measurement of my MG2.5Rs from a couple of years ago.

https://www.audiosciencereview.com/.../speakers-distortion.5920/page-14#post-133785

I hope one day Amir gets to measure one of the Linkwitz speakers (or perhaps a Gradient speaker) - certainly the LX521 is a considerable step up in every area from my MG2.5Rs.

 

[Blumlein 88]

Dave Reite solved the problem, these were measured off the floor and you can't measure dipole bass that way, it disappears because of the acoustical short circuit around the bottom of the speaker.

Well the Klippel is giving results of an anechoic chamber. Without being in a chamber. So it should measure as if there were no floor.

I'm not sure if this method has issues with large panels or with dipoles or both. It should not matter. Then again large panel listening experience always seems at odds with any of the common measurement methods.

I remember the first time I heard a panel. MG 1 Maggies. My description was deficient bass with zero impact, and no real treble. A light weight ephemeral sound. But from about 100 hz to 8 khz it's was open and clear and magical. I do think the appearance of something so cool impacted that perception too. Plus it only sounded good over a narrow loudness range with you sitting in The Spot.

I later acquired some MG 2i's. I didn't keep them long. It was a love/hate relationship. I found ESLs much more to my liking even though they've plenty of issues of their own.

I'm not convinced the Klippel is telling us the truth about panels and I'm not convinced that it isn't. The results of Harman blind tests on the panels indicate when you can't see the speaker they sound like they measure. Not very well. I could believe the influence of panel visuals is greater than box speakers. So in real life maybe panels sound great even though they don't.

 

[Feanor]

"Standard" measurements of dipole speakers are not definitive of the sound these speakers can produce with appropriate set-up in the listening room. Amir's measurements and those taken over the years by John Atkinson, et al., support my conclusion, IMO.

I have owned both Magneplanar MMG's and MG 1.6QR's and enjoyed both very much. Both integrated with subwoofers quite satisfactorily. In the former, MMG instance I used an external crossover that excluded all frequencies below 80 Hz, sending these to the subwoofer. Incidentally it's my opinion that tilting the speaker such the the top and bottom of panels are equidistant from one's ears improves the sound at least a bit.

Dipole design for many users produce a soundstage that just isn't equaled by unipole speakers.

 

[Frank Dernie]

By all accounts they don't have the time. They're too busy trying to fill the long back orders for this speaker.

I would be surprised if the person doing measurements was the person flat out filling in back orders!
I would imagine that if sales are satisfactory without publishing measurements that this is an expense they can avoid.

 

[Vasr]

Dipole, schmopole...

All I can say is that this set of measurements will certainly leave people polarized into one camp believing that the other is hearing things from these speakers that are not real, and the other camp believing that this measurement setup don't capture what is real about this speakers.

A bit different from the debates on amps and DACs where neither camp is likely to hear anything that is reflected in the measurements.

 

[Robbo99999]

I've not read any of the comments yet, just the review, so this is my quick take on it........to but it bluntly an overengineered disaster of a speaker! I couldn't really see any redeeming factors in anything of the frequency response and indeed in the rest of the review including the subjective listening test.....pretty horrid speaker but an interesting review in light of the extra tests and the quirkeness/rarity of the speakers design.

 

[ruinevil]

You see an indentation toward top. This is not normally visible but the reason for it is a large button screw that holds the panel down (I think)

They are in different positions and numbers in my two channels. They probably help with matching the speakers, altering the distortion and frequency response in some way.

 

[ruinevil]

I have noticed with my LRS, that they lose maximum SPL when trying to play bass below 60 Hz with any other frequency running. Playing something like techno with both a lot of sub-bass and treble makes everything sound flat, whereas with most rock music they do okay full range.

 

[Feanor]

Dipole, schmopole...

All I can say is that this set of measurements will certainly leave people polarized into one camp believing that the other is hearing things from these speakers that are not real, and the other camp believing that this measurement setup don't capture what is real about this speakers.

A bit different from the debates on amps and DACs where neither camp is likely to hear anything that is reflected in the measurements.

It's not complicated, my friend: neither dipole listener nor people in general hear like a Spinorama.

 

[Feanor]

I've not read any of the comments yet, just the review, so this is my quick take on it........to but it bluntly an overengineered disaster of a speaker! I couldn't really see any redeeming factors in anything of the frequency response and indeed in the rest of the review including the subjective listening test.....pretty horrid speaker but an interesting review in light of the extra tests and the quirkeness/rarity of the speakers design.

If you've never listened at any length to dipole speakers, maybe you should shut the eff up ... just as suggestion.

 

[Robbo99999]

If you've never listened at any length to dipole speakers, maybe you should shut the eff up ... just as suggestion.

To be honest it's pretty easy to imagine how they'd sound from the frequency response when it has such large deficiencies within that response. If there wasn't a correlation between frequency response & perceived sound quality then there would be no point in measuring it! But yes, I've never listened to them, and wouldn't plan to for these ones.

 

[ruinevil]

Dipole, schmopole...

All I can say is that this set of measurements will certainly leave people polarized into one camp believing that the other is hearing things from these speakers that are not real, and the other camp believing that this measurement setup don't capture what is real about this speakers.

A bit different from the debates on amps and DACs where neither camp is likely to hear anything that is reflected in the measurements.

The measurements are accurate in how it would sound in a anechoic chamber, but dipoles are designed to use the room. Bose 901 measurements have similar issues.

The directivity problems are definitely true though. It's a problem with the large flat drivers. Magnepan tries to deal with it by changing how their wires are spaced, whereas Martin Logan curves their drivers.

 

[Ron Texas]

Thank you @amirm for your detailed measurements and review. Looks like I will be sticking to monkey coffins.

 

[DonH56]

FWIWFM, my biggest fear in switching from MG-IIIa's to conventional speakers was the loss of soundstage and "presence". I did not find that to be an issue with my Revels, and appreciate the lack of comb filter effects that caused image shifts with frequency until I treated the wall behind the Maggies. The latter was always their Achille's heel to me; bass is solvable, but it always took a lot of tweaking and/or treatment to get a solid image with my Maggies. In a large room it was much easier, but alas I rarely had a large room for them. Over the years the thing that bugged me about many conventional speakers, even some fairly pricey ones, was the transition through the lower to upper vocal region. Too many seemed to have a sort of "disconnect" through that region that was audible and measurable. The big panels and low-order crossover seemed to help. That was also an issue I had with many hybrid ESLs; the woofer crossover to the panel was high enough that the transition from dipole to monopole and change in soundstage was pretty audible.

The best I thought mine sounded was in a moderately large room, strategic treatments behind, with a sub, and the panels bi-amped so I could use steeper slopes between Magnepan woofer and sub and also between Maggie woofer and mid/tweeter (I did not tri-amp -- did it for others and never really felt it was worth it).

 

[Vasr]

...to but it bluntly an overengineered disaster of a speaker!

Compared to most competent multi-driver speaker designs, the engineering of these are closer to a string of wires stretched over a hollow gourd than to a Stradivarius. The mystery is how they can actually sound good with that kind of engineering. Except to those that can "hear measurements".

 

[richard12511]

Lack of floor reinforcement is why the NFS measurements are sloping off below 300Hz. (I thought that would have been obvious to everyone.)

Wouldn't that be the same with every speaker?

 

[Wes]

we need some tests of electrostatics

 

[ruinevil]

Wouldn't that be the same with every speaker?

Probably true of floorstanders, but dipoles suffer more because bass frequencies are neutralized preferentially whenever the back wave and front wave touch. The so called acoustic short circuit.

 

[Selah Audio]

Dr Toole told this story about dipole bass.

But Floyd also used Mirage M1's for awhile too? Bi-polar design, but some of the characteristics of a dipole.

 

[josh358]

Given these are in room measurements, I still see a relatively large bass rolloff that seems to agree with Amir's measurements. Other than the peak @70Hz in the right graph, the bass looks equal to the rest of the spectrum, which would be in line with Amir's measurements. Good quality, accurate, and neutral bass should be ~8-10dB higher than the rest of the spectrum. Below (attached) are a couple examples of two different neutral speakers measured in two separate rooms(pre EQ). Obviously, it will depend on the room and listening distance, but as you can see, both of my rooms show the same general trend. In general, you should see a downward slope from low to high for a neutral speaker. I don't see that with those Maggie measurements you posted. If anything, the right graph(ignoring the room mode) actually has less bass energy than it does mid/treble energy, which is in line with the massive bass rolloff that Amir measured anechoically.
View attachment 83765View attachment 83766

Amirm measured a rolloff from 250 Hz down to 80 Hz, which is a measurement artifact. Dave Reite explained it:

Lack of floor reinforcement is why the NFS measurements are sloping off below 300Hz. (I thought that would have been obvious to everyone.)
These speakers do inherently head south below about 60Hz, but they are designed to sit on the floor for normal operation.

I hope people are starting to understand why I suggested this speaker would be difficult to characterize with the NFS system.

Otherwise, in-room measurements are going to vary hugely from room to room and the LRS is tuned for smaller rooms. Magnepan's target curve is I think 0.5 dB/octave, so ideally, they'd be up 5 dB at 20 Hz, 4 dB at 80. (Because it dumps less energy into the room off axis, I don't know that the ideal target curve for a dipole is necessarily the same as the target curve for an omni -- something to look into.) In any case, I'm guessing that Dick Olsher measured with the jumpers rather than with the tweeter resistors in place, a common mistake. The 1 ohm resistor is the normal setting, equivalent to the center position on a tweeter level control. No resistor would be for use in a dead room, and the 2 ohm resistor for a live one.

My impression of the MMG's, which I owned, was that they were bright in my room with the typical commercial recording -- that could be a consequence of low midbass levels, but then the room was also very bright at the time. My impression of the LRS at AXPONA was that it was very well balanced -- but they were very good recordings (the Reference Recordings Minnesota Symphony recordings have become a personal favorite) and I only heard a few tracks. I don't remember feeling that they were bright, or that the midbass level was too low, an impression I have had from some small Maggies in the past.