MAG Theatron M12-C Speaker Review

[amirm]

There are no real complaints about the measurements themselves, except that they were taken from a damaged speaker.

How do you explain that their measurements -- as filtered as they are -- correlate well with mine?

Notice that their measurements don't go below 50 Hz, indicating some kind of gated measurements and hence, lower resolution than mine:

Same resonance exists near 20 kHz. Overall sensitivity is almost perfect match. Same boost around 4 kHz followed by a dip. I actually show a bit more bass energy at 50 Hz than they do.

 

[amirm]

Testing an individual speaker outside of the full system is like listening to a solo violin without the orchestra; it doesn’t reflect the true capabilities of a unit designed to operate as a component of a single auditory organism in a specific environment.

Your own measurements I assume are for a single speaker outside of a system as well. This is the only fair way to compare products. If you sell an integrated system where its components can't be separated, then sure, I will test it that way. Here is an example from Tony Grimani: https://www.audiosciencereview.com/...tems-rixos-l-review-active-dsp-speaker.35171/

This is its directivity:

Again, far superior. System is sold and priced as a package. Yours has individual speaker pricing which makes it fair to test individually.

 

[Soundman UA]

Nothing was measured at "1 meter." Klippel near-field scanner, as the name indicates, actually scans closer than that. But by computing the full radiation pattern of the speaker, it can then present the results at any angle or distance. For the same of the review, and as I noted in the other thread, I show horizontal directivity at 3 meters:

We are in essentially far field of the speaker and measuring at longer distances would show little to no difference, as evidenced by your own measurements showing similar results:

View attachment 466459

The extremely uneven directivity will mean large seat to seat variations which cannot be fixed with any signal processing as it would affect all axis/distances. The tweeter is also beaming, making the sweet spot quite narrow -- the opposite of what you want in a home theater.

Compare your measurements to another Pro speaker aimed at this market, the JBL 4309: https://www.audiosciencereview.com/forum/index.php?threads/jbl-4309-review-speaker.27255/

This is its directivity:

As you see, it is in entirely different class than your speaker. Extensive research shows that smooth directivity correlates well with listener preference. We also have much wider directivity, maintained all the way nearly 20 kHz.

The complex and optimized geometry of the JBL compression driver is responsible for this excellent performance.

Nobody’s talking about measurements at 1 meter — the point is that no one listens to this speaker at that distance. In most cases, it’s placed behind a screen, in niches, with listeners at 3+ meters away, so the dispersion angle is more than sufficient. The point isn’t that the measurements are bad — on the contrary, they’re accurate. But they don’t reflect the real listening experience of a properly calibrated cinema system.

Attempting to interpret the sound based on raw graphs leads to incorrect and often negative conclusions. Try measuring line array speakers from recognized leaders in professional audio, like d&b or L-Acoustics, without presets — you’ll see frequency response swings of 20 dB. Yet, the M12 stays within ±3 dB even without a preset.

The problem is that you still don’t seem to understand what you’re talking about, comparing a JBL 4309 studio monitor to professional cinema audio. These are fundamentally different things from different worlds — they’re simply not comparable!

MAG can design directivity however they want, as they develop the horns for their systems themselves. But they make it exactly as required by the task and the installers.

Why are you highlighting the wider directivity at 1 kHz? This peak has no impact on the listening experience.

Look at how the sound spreads on your JBL diagram. It’s uncontrolled up to 1 kHz, meaning it’ll bounce off walls and create a ton of standing waves. In contrast, the M12 has excellent dispersion control, even at 200 Hz! The JBL’s characteristics are terrible for the tasks the M12 is designed for — it’ll make the whole room hum, yet it still can’t match the M12’s level of sound pressure, dynamics, or coverage accuracy.

 

[Soundman UA]

Your own measurements I assume are for a single speaker outside of a system as well. This is the only fair way to compare products. If you sell an integrated system where its components can't be separated, then sure, I will test it that way. Here is an example from Tony Grimani: https://www.audiosciencereview.com/...tems-rixos-l-review-active-dsp-speaker.35171/

This is its directivity:

Again, far superior. System is sold and priced as a package. Yours has individual speaker pricing which makes it fair to test individually.

Of course, speakers are measured individually, in line with standards — that’s the only possible measurement method, so your measurements align accordingly. There were never any complaints about them.

But what does this have to do with interpreting the results or the sound outside the context of a complete system, installation, and calibration? And how can it be overlooked that the M12’s dispersion is far better suited for cinema applications than what you show for JBL or Grimani Systems, which fill the room with hum up to at least 1000 Hz & 400 Hz ?

Let me repeat: designing a speaker and horn with a wider pattern is not an issue. But what will you do about the hum from reflections in a center speaker, which is supposed to deliver a maximally intelligible vocal channel?

 

[bachatero]

But what does this have to do with interpreting the results or the sound outside the context of a complete system, installation, and calibration?

If the speakers aren't specifically required to be used with DSP effects, then the default is to use no effects, yielding the frequency response shown.

 

[Soundman UA]

If the speakers aren't specifically required to be used with DSP effects, then the default is to use no effects, yielding the frequency response shown.

The MAG Theatron M12 acoustic systems require the use of DSP, as do most professional audio systems (or, at the very least, a parametric equalizer configured according to the presets provided on the manufacturer’s website). That’s why they aren’t sold in regular stores but are offered exclusively through distributors/installers as part of a complete cinema solution.

That said, the M12’s sound is quite satisfactory even without DSP — not fantastic, but within acceptable limits. However, I haven’t seen a full-fledged cinema system without a processor in a long time, so I see no reason to forgo one.

 

[bachatero]

That’s why they aren’t sold in regular stores but are offered exclusively through distributors/installers as part of a complete cinema solution.

? I don't think this was the case because you can buy one right here: https://srnd.store/collections/mag-theatron/products/mag-theatron-m12-c

It's a similar idea with PA speakers and actually my Neumann and Kali studio monitors, since you can only get those through distributors/installers too.

 

[amirm]

In most cases, it’s placed behind a screen, in niches, with listeners at 3+ meters away, so the dispersion angle is more than sufficient.

A center speaker needs to provide coverage for all the seats and width of the screen. To get the latter, you need side reflections as to widen the perceived width of the soundstage. In addition, such dispersion needs to be similar to on-axis without major discontinuity as the M-12C is showing. Such directivity errors will create colorations as they add to the on-axis response of the speaker.

Attempting to interpret the sound based on raw graphs leads to incorrect and often negative conclusions. Try measuring line array speakers from recognized leaders in professional audio, like d&b or L-Acoustics, without presets — you’ll see frequency response swings of 20 dB. Yet, the M12 stays within ±3 dB even without a preset.

You say measurements lead to incorrect conclusions but then go on and use measurements to make your point???

The numbers you use are indeed useless. We rely on 5 decades of research which tells us which measurements correlate well with listener preference in controlled listening. That research says that on-axis response needs to be flat, and off-axis, smooth copy of the on-axis. This research is formalized into the CEA-2023/CTA-2034 standard which I use, and MAG also quotes in its literature. Actual measurement distance is 2 meters, but reported at one meter. This is selected as to be in the far field of speakers tested. This sequence of measurements -- not just one set of numbers as you use -- allow us to make quick predictions of the neutrality and fidelity of speakers:

We see that on-axis response is not flat but both rough and wavy. Directivity (orange line) has major discontinuity around 1 kHz, indicating poor integration of the woofer and tweeter. We also have a strong resonance around 19 kHz which should have easily been identified and fixed. Bass extension is also not good for a speaker in this price class and size. Clearly the target was maximum sensitivity at the expense of fineness and accuracy.

 

[amirm]

MAG can design directivity however they want, as they develop the horns for their systems themselves. But they make it exactly as required by the task and the installers.

The latter is false and making your own waveguide doesn't mean you know how to accomplish this task. JBL for example, has a patented simulation tool that allows them to specify what directivity they want, and the software produces the shape of the waveguide. You can see this in the detail of their waveguide:

I see nothing in MAG portfolio that says they can do that with a snap of a finger. That said, it can be done with modern simulation tools (COMSOL, etc.). They just need the motivation which hopefully my review has provided them.

 

[amirm]

Let me repeat: designing a speaker and horn with a wider pattern is not an issue.

It clearly is or we would have a wide and controlled dispersion.

But what will you do about the hum from reflections in a center speaker, which is supposed to deliver a maximally intelligible vocal channel?

Side reflections from a center speaker increase the total energy that reaches your ear and with it, improves intelligibility. There is a ton of research here to back this. From Dr. Toole's book, Sound Reproduction Loudspeakers and Rooms:

"Following the pattern set by studies involving single reflections, Lochner and Burger (1958), Soulodre et al. (1989), and Bradley et al. (2003) found that multiple reflections also contribute to improved speech intelligibility. The most elaborate of these experiments used an array of eight loudspeakers in an anechoic chamber to simulate early reflections and a reverberant decay for several different rooms (Bradley et al., 2003). The smallest was similar in size to a very large home theater or a screening room (13 773 ft3, 390 m3). The result was that early reflections (<50 ms) had the same desirable effect on speech intelligibility as increasing the level of the direct sound. The authors go on to point out that late reflections (including reverberation) are undesirable, but controlling them should not be the first priority, which is to maximize the total energy in the direct and early reflected speech sounds. Remarkably, even attenuating the direct sound had little effect on intelligibility in a sound fi eld with sufficient early reflections."

Side reflections also widen perceived image of the center speaker, covering more of the screen as opposed to sound coming from a point in the middle of the screen. This is why MTMs in horizontal configuration make poor center speakers despite their popularity.

All of this calls for a center speaker to have wide and accurate off-axis response/directivity. These are the same speakers that also do well in normal 2-channel listening, making our job of evaluating them simple.

 

[Steakhouse]

Imo the M15 speakers look mich better with regard to their measurements. Supposedly because the diffraction critical range around 700Hz now falls into the midrange driver which is in a waveguide that does not „see“ the edges of the enclosure.

 

[Matt_Holland]

Hi, I have some connection to MAG, so I’ll comment personally.

Thanks for your contributions to this thread. May I ask what your connection MAG is? You say you are answering personally but your comments seem to me like you are speaking on behalf of the manufacturer.

I think it’s a shame that MAG don’t plan to contribute any further to this thread. Misunderstandings about the intended customer and the sale channel aside I think there have been some positive comments about the M12-C and a general growing understanding that in the context of flush mounting and use of prescribed EQ presets, the performance of the speaker has much merit.

The directivity blip at 1KHz absolutely deserves further debate. And an answer to the question many have asked as to why the bass driver is recessed in a shallow waveguide (not sure it will be guiding much!) because its sharp edges might be hurting the frequency response, directivity and even causing secondary sound sources due to diffraction.

I have been in many home cinemas that are overly treated acoustically with side wall reflections absorbed. In this scenario, and with suitable seating area inside the speaker’s horizontal beam, I can imagine the directivity discontinuity being a non-issue, or minor at least.