JBL 4367 Review (was M2 'review')

[amirm]

Well that sounds a tad bit different than 4 subs, 8 mics, ARCOS, etc...and a bit closer to this "M2" review topic. Minus the EQ and plus some type of magic ball props and wiring doo-dahs.

The review was NOT that of M2. The 4367 does not come with built-in EQ. Only the M2 does. Should this have been an M2 review it would have come with calibration service using ARCOS. But it is not.

In my 2-channel music system I use DIRAC (I have ARCOS in my theater). I have it set to correction below 200 Hz and it does a great job of removing the modest amount of boominess that exists without it. This is exactly what I said the M2 can do which you seem to keep arguing against with no evidence presented that there is something wrong with it.

The method has been documented and demonstrated in double blind tests to improve subjective performance: http://www.aes.org/e-lib/browse.cfm?elib=15154

#4 is no EQ. #1 and #2 are ARCOS prototype EQ systems. Clearly subjective performance improved using EQ.

Here is the detail that went into that overall preference:

Coloration was clearly improved in RC1 and RC2 (bottom white rectangle) over doing nothing (RC4). As was boominess which is hallmark of room modes in low frequencies.

Objective analysis backs the same:

Dashed line is no EQ. We see that winning EQ systems smoothed out that curve and added an overall room response that was tilted down (Red, and Green).

So no, there are no "doo-dahs" involved here. EQ works in better implementations confirmed in double blind listening tests. It just does. Trying to screw around with speaker design when the room dominates in low frequencies is the domain of "doo-dahs."

 

[Purité Audio]

I agree I have and will continue to try anything, the Modex panels helped ,a little, perhaps if I had a dedicated room and could install far more of them..
But even with the Modex I still have this huge 28Hz peak, if multiple subs could ameliorate that ?
Keith

 

[amirm]

Well that sounds a tad bit different than 4 subs, 8 mics, ARCOS, etc...and a bit closer to this "M2" review topic. Minus the EQ and plus some type of magic ball props and wiring doo-dahs.


I would suspect most here would already know what localization regarding subwoofers entail...and why it would be a rather poor idea to run 4 subs to 500hz to spatially average out holes....excuse me, nulls, in the frequency response(s), that EQ cannot correct.
Fine, if you'd rather stereo imaging to spatial rendering, I'm ok there also.
I'm going too assume you've done the Harman listener training. Have you heard either M2 or 4367 in this review?

You are still not owning the arguments, keep throwing links around. Answering anyway, we have had the M2 in our theater at work since it came out (12 to 18 months ago?). It is also programmed using ARCOS. It is the most dynamic speaker I have ever heard while bringing the delicacy of Revel line. It can peel the skin off your face without a sub, yet maintains its full fidelity regardless of level. A few second demo would make you want to get it as it has for anyone that we demo it for. I have not had access to 4367 so can't comment on that.

And no, no one said you crossover a sub at 500 Hz. Subs are crossed over at 80 Hz. However to the extent modal density is low in the few hundred hertz, you can let ARCOS do its thing to correct the response. You then listen. If there are improvements there, you leave the filters be. If not, turn them off selectively or modify them in real-time to see if you like another setting better.

No mystery here. No reason to hide behind big words.

 

[amirm]

I agree I have and will continue to try anything, the Modex panels helped ,a little, perhaps if I had a dedicated room and could install far more of them..
But even with the Modex I still have this huge 28Hz peak, if multiple subs could ameliorate that ?
Keith

They can but an EQ would be my weapon of choice first if it is a peak. If you have a PC server for your music, just get a copy of Dirac and let it do its correction there.

 

[Purité Audio]

Amir Hi, yes I have used Dirac , I preferred it just correcting. low bass, now I use an Illusonic IAP processor and Christof Faller of Illusoniccreates the equalisation from my measurements.
This has produced the best results so far.
Keith

 

[AJ Soundfield]

AJ,
My understanding of your philosophy is that if its not perfect, it's not worth it. That seems rather extreme to me.

You misunderstand. No "perfect" here, nor would I believe any such thing exists. I've been following the research for 20+ years. Toole, Jones Eureka/Archimedes, Salmis bypass tests, JJs PSR, Griesinger/Logic7, etc, etc, etc.
You know Toole because he is by far the most recognizable. But there is a great deal more out there. I follow it all.
There is nothing wrong with what you are doing if it satisfies your need for fidelity. The Toole/Harman method does not meet mine for music.

I personally use many of these well known, tested and proven playback techniques. I even use four subs in the Welti midwall configuration. I use speakers with excellent and controlled polar response. I was careful not to put my listening position into a major LF SBIR ("null").
I probably use more room treatments than Amir is used to seeing on a stereo playback environment. But the treatments I use are not usual fiberglass types so I bet he would be surprised if he ever heard my system.

Maybe I would be too, but what Amir or I think should be irrelevant, if it pleases you. There is no escaping preference in this hobby. My speakers clearly wouldn't work in such an environment, because I haven't put that excess power there to begin with and the delayed decorrelated indirect radiation needs reflective, not absorptive environment. And that's just the front channels.
"Stereo" doesn't meet my needs, well explained in what I've linked previously (PSR synopsis and here). Too much missing info.

These recommendations are well known and not based on black magic. There are many AES papers which support these strategies. I'm one of people who is willing to try anything and I have. I've found this approach to be the most valid approach to playback.

If you haven't noticed, mine are all based heavily on AES work, just not by the same one or two "well known" authors. Actually most of it is rather obscure. Yes, I'm a total geek.
But that makes it no less valid. Just less well known.
As I'll say ad infinitum, we've only got one set of ears to please.

cheers,

AJ

p.s. You attending LSAF this year?

 

[AJ Soundfield]

Well that sounds a tad bit different than 4 subs, 8 mics, ARCOS, etc...and a bit closer to this "M2" review topic. Minus the EQ and plus some type of magic ball props and wiring doo-dahs.

The review was NOT that of M2. The 4367 does not come with built-in EQ. Only the M2 does. Should this have been an M2 review it would have come with calibration service using ARCOS. But it is not.

Sigh. Yes Amir, not an M2. Yes Amir, the reviewer did not use EQ. Hence "minus" the EQ.

So no, there are no (wiring) "doo-dahs" involved here. EQ works in better implementations confirmed in double blind listening tests. It just does. Trying to screw around with speaker design when the room dominates in low frequencies is the domain of (wiring) "doo-dahs."

Ok Amir, perhaps all just a misunderstanding of what was said.

 

[AJ Soundfield]

I agree I have and will continue to try anything, the Modex panels helped ,a little, perhaps if I had a dedicated room and could install far more of them...

What did they help with?

But even with the Modex I still have this huge 28Hz peak, if multiple subs could ameliorate that ?

Yes, absolutely, as could EQ. I use both <50hz. More headroom, lower distortion with multi-sub. Mono is fine <50hz, as 40Hz is the very edge of lateralization thresholds and certainly well below localization.

cheers,

AJ

 

[fas42]

It can peel the skin off your face without a sub, yet maintains its full fidelity regardless of level

Ahhh ... you mean it's a competent system! Hmm ... I knew there had to be at least one or two out there ...

 

[TitaniumTroy]

I like my speakers forged from the finest Damascus steel, by master blacksmiths.

 

[amirm]

I like my speakers forged from the finest Damascus steel, by master blacksmiths.

With an alias like yours, it better be made out of some kind of metal.

 

[AJ Soundfield]

Welcome Doug, hope all is well.

Answering anyway, we have had the M2 in our theater at work since it came out (12 to 18 months ago?). It is also programmed using ARCOS. It is the most dynamic speaker I have ever heard while bringing the delicacy of Revel line. It can peel the skin off your face without a sub, yet maintains its full fidelity regardless of level. A few second demo would make you want to get it as it has for anyone that we demo it for. I have not had access to 4367 so can't comment on that.

I've only seen DSP EQ'd measurements for the M2. The 4367 shown here:

Any thoughts on what might be going on there above 400Hz?
I think I recall you mentioning hearing resonances when you toured rooms at an audio show, so I sort of presumed you have done the Harman training.
Did you notice anything with the M2s 2216ND woofer?

cheers

AJ

 

[FrantzM]

This thread is about the M2 and perhaps the 4367.. SO let me fire away

The more I am looking at the M2 the more I am interested: The darn thing cost about $20,000 in the US, complete with amps.. No chump change but ... We are all audiophiles, here and we know that such sum would buy just the cables in some systems or be a down payment on a speaker cable in some others so let us keep things in perspective.
I am more than convinced that the best bass in a small room is obtained with multisub. EQ in itself cannot solve some problems. EQ plus Multlisubs is the best of all worlds . I do not see myself going forward without multi-subs .. The difference is that : staggering. No "I-think-I-heard-a-difference" .. No! Truly Night and Day , World-shattering kind of differences and as opposed to mere "differences". And with a smooth response in the bass. And one other thing with multi-subs really cheap subs that is in audiophile parlance, <$500 will bring you single digits with low distortion ( <5% easily) at 100 dB SPL in a 6 X 8 x 3 m room (at least that what I got).. with Paradigm and/or Sunfire subs ... You go to with 3 or 4 <= $1000 subs and you surpass most full range passive speakers whatever their prices ( I am an audiophile so I should be allowed to make exaggerated claims ) ...

How would the M2 work with multiple subs and the Arcos?
Does the M2 requires EQ to have this Kansas-flat FR? I suppose "Yes"
If I were to forgo its recommended amplification would it work as well without EQ and keeping in mind that I intend to do bass with multi-subs?

Would the 4367 be a better fit for multi-subs and choice of amps? I guess the answer is "Yes" too.

 

[amirm]

I've only seen DSP EQ'd measurements for the M2.

If you mean DSP for the speaker itself, yes, that is all you will ever see since it is an active speaker and DSP is mandatory there.

The 4367 shown here:

Any thoughts on what might be going on there above 400Hz?

What *is* going on AJ?

I think I recall you mentioning hearing resonances when you toured rooms at an audio show, so I sort of presumed you have done the Harman training.
Did you notice anything with the M2s 2216ND woofer?

I spent a bit of time with Harman training software which improved by acuity substantially above average person. But I am still way away from the minimum required to be in their trained listener level.

As to hearing resonances at the show, I heard it in Martin Logan but in the others, I observed them. Those were cabinet resonances which I hope you are not confusing with transducer ones.

 

[AJ Soundfield]

EQ in itself cannot solve some problems.

Hi Frantz, that is correct. EQ can reduce peaks...and associated modal decay. EQ cannot fill nulls, where energy is being stored at that point.

I am more than convinced that the best bass in a small room is obtained with multisub.

I'm not. As leat not above 40-50hz.
Yes, it is true that multiple sources will spatially average out the holes and peaks....and thus result in smoother amplitude. Yes, it is also true that smoother amplitude is perceptually preferred over amplitude with peaks...and nulls though to a lesser extent. Is smooth amplitude alone the final arbiter of bass quality?
But...it is also true that we can localize above 80-90hz and lateralization down to around 40-50hz.

EQ plus Multlisubs is the best of all worlds .

If this is true, how do you spatially average out nulls > 90Hz, when sources become localizable? What do you do to 500Hz, when EQ can't fix holes? How do you maintain spatial effects from interaural phase differences in the recording if your bass below 90hz is mono? Do you listen to classical/jazz etc, or is this for pop/rock/movies?

Have you read any of my links (Griesinger, JJ, etc) ?

Have you every heard gradient bass in a small room? (Not the brand, though that's what the name is about)

cheers

AJ

 

[amirm]

If this is true, how do you spatially average out nulls > 90Hz, when sources become localizable?

With subs and mains both firing during those frequencies you have new set of tools. Namely how you optimize both to get smoother response during crossover. Here is an actual, measured performance: http://www.audiosciencereview.com/f...s/computer-optimization-of-room-acoustics.12/

The dip is nicely fixed not only with respect to one speaker, but group of speakers simultaneously. And further matching the overall timber that the target curve (in dashed lines) required. It is not something you can do manually but with a computer and DSP you can get a lot done -- certainly a lot more than just EQ.

What do you do to 500Hz, when EQ can't fix holes?

By 500 Hz you are above transition frequencies. Your modal density becomes high enough as to not worry about individual modes. And further, the frequency discrimination (resolution) of your hearing starts to reduce so what you measure is not what you see. You need to apply smoothing to the measurements to get them psychoacoustically correct.

In addition, in that domain and higher the speaker performance starts to dominate. Get a well designed speaker with good off-axis response, don't screw up its response with wrong acoustic products and life will be good. Really good!

 

[AJ Soundfield]

If you mean DSP for the speaker itself, yes, that is all you will ever see since it is an active speaker and DSP is mandatory there.

Yes its remarkably linearized on axis:

Some serious DSP going on there. Its the "bobble" around 00hz in the off axes/DI that had me puzzled.

What *is* going on AJ?

Paper cone resonance.
The benefits of that high sensitivity/huge dynamics from pro drivers you speak of, is because the cone is low mass...and 99.9% of the time, paper based. As such, it will always have resonances within the passband, somewhat the opposite of what Toole had Infinity do (I'm sure you remember the Toole/Olive papers on detection of resonances) with developing CMMD, a very rigid material that pushed resonances well out of the passband.

I spent a bit of time with Harman training software which improved by acuity substantially above average person. But I am still way away from the minimum required to be in their trained listener level. As to hearing resonances at the show, I heard it in Martin Logan but in the others, I observed them. Those were cabinet resonances which I hope you are not confusing with transducer ones.

The Harman training should be mandatory for any serious listener and/or designer, as well as the Klippel tests. I'm sure audiophiles will avoid them like the plague, as these are real differences, not the imaginary ones they are so trained for.
Gotcha, yes, I was going by some vague memory and was more interested in your photos. Good on you hearing cabinet resonances through the show room/speaker interface!

cheers

AJ

 

[AJ Soundfield]

With subs and mains both firing during those frequencies you have new set of tools. Namely how you optimize both to get smoother response during crossover. Here is an actual, measured performance: http://www.audiosciencereview.com/f...s/computer-optimization-of-room-acoustics.12/

View attachment 735​

The dip is nicely fixed not only with respect to one speaker, but group of speakers simultaneously. And further matching the overall timber that the target curve (in dashed lines) required. It is not something you can do manually but with a computer and DSP you can get a lot done -- certainly a lot more than just EQ.


By 500 Hz you are above transition frequencies.

Ok Amir, let me ask again, this way: What do you do for 100Hz-500Hz nulls with multi-subs/EQ?
Thanks.
Oh btw, are those subs in mono?

 

[amirm]

Ok Amir, let me ask again, this way: What do you do for 100Hz-500Hz nulls with multi-subs/EQ?

You want me to retype that post??? I specifically showed problems solved at 100 Hz. And provided explanation for higher frequencies.

Oh btw, are those subs in mono?

The subs are driven differently but the input to them is summed mono from the source.

 

[AJ Soundfield]

You want me to retype that post???

No I want you to tell me how multisub EQ fixes nulls at 150Hz, 200Hz, 250Hz, 300Hz, 350Hz, 400Hz, etc etc?

Ok the answer: It CANNOT

Oh and if your example above is mono bass, that's great for movies and Bieber. It won't cut it for Classical, Jazz or other acoustic music, sorry. YMMV.
So maybe mono multisub and EQ works for some, but not others.

cheers,

AJ