• 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!

Revel Performa 3 vs ML Electromotion ESL - comparative measurements

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
I ran recently on the russian HiFi site Stereo.ru where I found tests of these 2 speakers. As each of them is a kind of king of it's domain (monopole and dipole) it makes sense to imagine "battle of the titans" to see how they would fare versus each other.

So, which speaker won this monopole vs dipole ultimate fight? Which measured better and which do you think would sound better? :)

Here I will post pictures of the measurement graphs, complete articles are available via links.

Revel Performa 3 F208:


F208.JPG


Martin Logan Electromotion ESL:

ML.JPG
 
Last edited:

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
They both seem to measure well. The revels’ enclosed dynamic drivers of course offer wider dispersion than the MLs (and overall lower distortion).

The distortion measurements indicate that the measurement conditions are not adequate to give accurate data in the bass (likely dominated by ambient noise), although it seems that the MLs are tuned lower than the revels yet also have a higher -3dB point. So the MLs possibly sound leaner and tighter than the Revels (which IME with other Revels already tend in this direction). The MLs’ bass response looks a little odd to me though, assuming the measurements can be trusted down there.

They only provide 3 pretty basic measurements, so it’s hard to say much without more data, eg more detailed horizontal polars, vertical polars, CSD, etc.

I’m also not sure what the measurement conditions are (on my mobile so can’t read the Russian - maybe it’s explained in the text?)

Finally, the MLs will produce a a strong inverted back-wave and greatly reduced lateral dispersion in the operating range of the panel, and will therefore require very different set-up and treatment.
 
Last edited:
OP
Krunok

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
They both seem to measure well. The revels’ enclosed dynamic drivers of course offer wider dispersion than the MLs (and overall lower distortion).

The distortion measurements indicate that the measurement conditions are not adequate to give accurate data in the bass (likely dominated by ambient noise), although it seems that the MLs are tuned lower than the revels yet also have a higher -3dB point. So the MLs possibly sound leaner and tighter than the Revels (which IME with other Revels already tend in this direction). The MLs’ bass response looks a little odd to me though, assuming the measurements can be trusted down there.

They only provide 3 pretty basic measurements, so it’s hard to say much without more data, eg more detailed horizontal polars, vertical polars, CSD, etc.

I’m also not sure what the measurement conditions are (on my mobile so can’t read the Russian - maybe it’s explained in the text?)

Finally, the MLs will produce a a strong inverted back-wave and greatly reduced lateral dispersion in the operating range of the panel, and will therefore require very different set-up and treatment.

Here are manufacturer specs for the bass extension - hard to tell as ML doesn't spec much but to me it seems F208 goes deeper:

F208: -10dB@23Hz, -6dB@27 Hz, -3dB@31Hz
ML ESL: 42–22,000 Hz ±3dB

CSD was not measured. I believe we can safely assume MLs are better in vertical dispersion while Revels hold the ground better in horizontal, although I have to say MLs measrued there better than I expected.

There is no explanation in these 2 articles how exactly they did measurements. I will try to search their site later, maybe there is a dedicated article on that topic.

Although you put your words very carefully I got a feeling your vote would go to Revel. Is that correct? :)
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
Although you put your words very carefully I got a feeling your vote would go to Revel. Is that correct? :)

Haha yes probably. Certainly appears to be the safer choice. It’s also been measured more extensively by Stereophile and those measurements show close to textbook excellent performance so I’m going on that too.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,717
Likes
2,897
Location
Finland
Stereophile has tested and measured many models from Martin Logan, the last one is this
https://www.stereophile.com/content/martinlogan-masterpiece-renaissance-esl-15a-loudspeaker
and this is of Revel
https://www.stereophile.com/content/revel-performa3-f208-loudspeaker

But these measurements don't tell the difference of a dipole vs monopole speaker in a room. They do sound different and after some listening it is easy to decide which you like more!

ML ESL has double 12" sealed woofers with dsp, look at red line of nearfield response. The panel''s peak at 300Hz is an artefact of nearfield measurement - it is needed to compensate dipole loss in farfield.
117MLRenfig1.jpg

Atkinson: I keep returning to the Renaissance ESL 15A's extraordinary in-room response, which is even better than the BeoLab 90's. As I said: "Wow!"
Read more at https://www.stereophile.com/content...-loudspeaker-measurements#6mimLcW74LwZWLD1.99


An this is same for F208, double 8" in bass reflex, so look at both blue and red line
714R208fig3.jpg

Atkinson:
Summing up the Revel's measured performance is easy: In every way, this is textbook loudspeaker design.
Read more at https://www.stereophile.com/content...-loudspeaker-measurements#7wOEQ6k9GCC9RSC2.99
 
Last edited:

Ron Texas

Master Contributor
Joined
Jun 10, 2018
Messages
6,074
Likes
8,906
The two designs are so different that comparing measurements is nearly futile. Because they are sold by Best Buy, ML is one of the most widely available high end speaker brands available in the US, along with B&W and most recently KEF.
 
OP
Krunok

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
But these measurements don't tell the difference of a dipole vs monopole speaker in a room.

It is absolutely true that nearfield measurements don't tell a complete story about how each speaker sound in a room, but wouldn't you say that dipoles are in a kind of complex situation there because of added reflections? :)

Don't get me wrong, I heard MLs several times and I really like how they sound but strictly speaking wouldn't you say that speakers like Revels are providing more accurate in-room response?
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,403
Likes
5,296
Location
Somerville, MA
It is absolutely true that nearfield measurements don't tell a complete story about how each speaker sound in a room, but wouldn't you say that dipoles are in a kind of complex situation there because of added reflections? :)

Don't get me wrong, I heard MLs several times and I really like how they sound but strictly speaking wouldn't you say that speakers like Revels are providing more accurate in-room response?

The situation with dipoles isn't too hard to compare with monopoles.

Below Schroeder frequency in the bass, there is some debate to the differences between monopole, dipole and cardiod bass. In the case of the ML, it uses an enclosed woofer, so it's just a standard monopole here.

At midrange, the ESL exhibits really narrow directivity due to the large radiating area - equivalent to a woofer of the same size. What's somewhat unique is the sudden and complete drop off at 90 degrees off axis due to the dipole cancellation. This reduces near reflections, which is a good thing in general. What is somewhat more problematic is that in the case of the ESL, the membrane creating the midrange and treble is enormous, leading to very tight directivity. You may know that large drivers 'beam' at high frequencies. Well, if you keep going even higher they get downright chaotic. ML's curvature may help with this for all I know. QUAD systems after the 63 model had another means of radiating treble from a smaller part of the panel.

This would then be a very dark sounding speaker if it weren't for the fact that the power response is filled in with reflections, so there is a good amount of uncorrelated treble energy bouncing around the room. I would recommend placing ESLs far from back walls, but not adding any acoustic absorption to the back walls for this reason.

If you want all the advantages of a dipole, and real dipole bass, see the Linkwitz 521 (RIP SL). This has the advantages of a normal 4-way monopole system with smoothly falling power response, low distortion drivers and wide treble dispersion combined with the lack of sidewall reflections exhibited by dipoles.

It is true that they interact with rooms in very different ways, but the ESL is a poor and somewhat outdated vision of what dipole speakers can be. See the gainphile DIY projects to see probably the most advanced dipoles, which use waveguides for the treble.
 
Last edited:

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
Well the only known and tested correlation between subjective sound quality and measurements is the Spin-o-rama data from Harman. I think we know who has the better end of that.

Also, I've never liked ML hybrids as the woofers don't integrate well unless you use REQ or low end DSP. I never liked ESL's with subs for the same reason, but they can be integrated well in the current state of home audio.

Now based upon what measures there are above, the Revel's are a little better though both actually look fairly good.

I've owned a number of ESL's (have some Soundlabs now) and I've heard some Revels. Have some low end Revels in a video system. I'd think the Revels above are the more accurate speaker all things considered. I'd probably like the ML's anyway. Would I like the ML's enough to pay more for them, to pay more for an amp to run them well etc etc. I might. But I'm probably just conditioned to like ESLs. If someone starting out in this hobby wanted advice I'd advise them to get the Revels. Or I'd invite them to hear some ESL's and see if it tickled their fancy. If not then don't worry, be happy, get the Revels. There is a coolness factor to panel speakers. They do seem to have a sound.

I do believe the Harman research, and yet wonder like many do if the ESL's are given a fair shake in their listening conditions at Harman. Maybe I just need to do the blind test with some in the Harman facility and it would erase the gut feeling I have that ESL's do something that is being problematic in the mono conditions of Harman's testing. Knowing something intellectually and it being in opposition to gut level feelings based upon experience is one of the more interesting and difficult things about life.

EDIT to add:I also think measures with a single point microphone of a panel are problematic.
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,403
Likes
5,296
Location
Somerville, MA
EDIT to add:I also think measures with a single point microphone of a panel are problematic.

Single point measurements are problematic with all audio transducers, but I'd argue they're actually more useful with ESLs - since you have to listen to them on axis, you want to make sure the axial response is respectable.

Having said that, I have no idea how you would measure a big dipole in a room; I wouldn't trust Stereophile to get it right unfortunately.
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
Single point measurements are problematic with all audio transducers, but I'd argue they're actually more useful with ESLs - since you have to listen to them on axis, you want to make sure the axial response is respectable.

I think @Blumlein 88 was suggesting that panels are likely to exhibit more pronounced response peaks/dips as one moves up/down or left/right by a few degrees. In other words, it would be more useful to average out a number of measurements taken from various positions clustered around the central axis and to average them - something along the lines of a "listening window" response. I don't think he's suggesting taking these measurements at angles greater than say 15° off-axis.

The same is of course true for measurements any speaker as you say, just particularly so for panels.
 

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,717
Likes
2,897
Location
Finland
^RTA room response or ERB-gated sweep tell you the spectral balance and reveal some room modes too. But only listening tells the difference of mono vs. dipole or omni - and easily. That is mostly because reflections of the backwards projected sound have more delay than those from sidewalls, and spectral balance of those diffractions is different. These are tricky to catch in measurements and even more tricky to interprete. Still an open question... And finally, it is a matter of taste which sounds better to you!

This difference and reasons for it were discussed recently in this thread
https://www.audiosciencereview.com/forum/index.php?threads/soundstage.7211/
 
Last edited:
OP
Krunok

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
At midrange, the ESL exhibits really narrow directivity due to the large radiating area - equivalent to a woofer of the same size.

That is true, and response in the region 1-5kHz for 30 and 60 deg is showing that. I wonder how that would compare when listening in the same room..
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
I think @Blumlein 88 was suggesting that panels are likely to exhibit more pronounced response peaks/dips as one moves up/down or left/right by a few degrees. In other words, it would be more useful to average out a number of measurements taken from various positions clustered around the central axis and to average them - something along the lines of a "listening window" response. I don't think he's suggesting taking these measurements at angles greater than say 15° off-axis.

The same is of course true for measurements any speaker as you say, just particularly so for panels.
Yes this is what I had in mind. Distance from edges of the panel vary greatly vs center at any measuring point. Which will give uneven response vs a smaller cone speaker.
 

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,717
Likes
2,897
Location
Finland
I would also be very pleased to see spinorama or even 3D measurements of all panels and dipole systems. Some of them have been measured here. Amateur and magazine measurements are mostly rotated too narrow and done in different circumstances - impossible to compare.

https://www.princeton.edu/3D3A/Directivity.html Essence, Gainphile, Gradient Helsinki, QUAD ESL, Sanders

Acoustat, Essence, Magnepan, ML, Sanders, Quad etc. panels all have different construction and are not totally dipole
Linkwitz, Kreskowsky, Gainphile etc. diy speakers all have different construction and are not totally dipole
Gradient, Jamo, Kyron, Steinway Lyngdorf, DNA, etc. factory-made dynamic "dipole" speakers all have different construction and are not totally dipole

And then the dipole speakers for HT, that we should totally forget
 
Last edited:
OP
Krunok

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Yes this is what I had in mind. Distance from edges of the panel vary greatly vs center at any measuring point. Which will give uneven response vs a smaller cone speaker.

And probably added reflections from the wall behind the speakers wouldn't help either.
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,403
Likes
5,296
Location
Somerville, MA
One wonders how (or if) ML does these measurements. I suspect they do a lot of in room RTA type measurements rather than anechoic or quasi anechoic measurements. The latter are really required to see narrow band resonances.
 

Kal Rubinson

Master Contributor
Industry Insider
Forum Donor
Joined
Mar 23, 2016
Messages
5,270
Likes
9,765
Location
NYC
One wonders how (or if) ML does these measurements. I suspect they do a lot of in room RTA type measurements rather than anechoic or quasi anechoic measurements. The latter are really required to see narrow band resonances.
Why not? They are part of the Paradigm/Anthem group which has a large and modern anechoic chamber of their own.
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,403
Likes
5,296
Location
Somerville, MA
Why not? They are part of the Paradigm/Anthem group which has a large and modern anechoic chamber of their own.
I did not know that. I would love to see polars and how they handle treble dispersion. The last esls I heard sounded dark but good, great bass ( acoustat big guys)
 
Top Bottom