• 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!

JBL 4367 review by Erin

Status
Not open for further replies.

fredoamigo

Addicted to Fun and Learning
Forum Donor
Joined
Mar 11, 2018
Messages
604
Likes
998
Location
South East France
I've built enough speakers now, to know the design is relatively simple to make, using drivers and construction at least equal to the design.

And frankly, the box as is, could easily be improved by taking it active, and ditching the passive xover.

It could be bi-amped, using FIR processing to further smooth out the frequency response. A linear phase xover could be used to play with xover orders and frequency hoping for better polar optimizations. While at the same time eliminating any phase rotation or group delay above the bottom end rolloff.

For the bottom end, where the bass-reflex design is almost surely being protected from potential overecursion below port tuning frequency by a high-pass filter, I would add frequency dependent processing that doesn't put a hpf in place until there is a risk of over excursion.
This would significantly reduce bottom end group delay / phase rotation, and still protect the speaker. Make it more like sealed.


I'm sure the IKEA effect is real....and i'm sure i've no doubt suffered from it, and no doubt will occasionally continue to suffer from it, on some of my more exotic build/experiments.
But not on this guy we're discussing...it's a really simple speaker :)
There you answer me that you could do more or less a M2 ... subjectively, I did not feel any advantage to listen to this one ... but well, it was in a very different room.

On the other hand, I do not doubt the excellence of your level but it would be desirable, it seems to me, to have a little more humility towards the engineering of JBL ...
To make a copy/paste of the existing is one thing, to conceive and imagine a new thing is another.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
And frankly, the box as is, could easily be improved by taking it active, and ditching the passive xover.

It could be bi-amped, using FIR processing to further smooth out the frequency response. A linear phase xover could be used to play with xover orders and frequency hoping for better polar optimizations. While at the same time eliminating any phase rotation or group delay above the bottom end rolloff.

For the bottom end, where the bass-reflex design is almost surely being protected from potential overecursion below port tuning frequency by a high-pass filter, I would add frequency dependent processing that doesn't put a hpf in place until there is a risk of over excursion.
This would significantly reduce bottom end group delay / phase rotation, and still protect the speaker. Make it more like sealed.

While I do like this JBL model already as it is, I suspect it could be greatly improved if they they simply switched over to a more integrated “active” speaker and DSP design. We know they’re capable of this given the M2 and the 705/708 studio monitor models. But, JBL probably knows the audience market well enough where most of their audiophile buyers nevertheless will likely prefer something passive in the end.
 
  • Like
Reactions: Pio

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
While I do like this JBL model already as it is, I suspect it could be greatly improved if they they simply switched over to a more integrated “active” speaker and DSP design.
I would question that. The majority of the response fits within a +/- 1.5dB window which for production tolerance is great. The Horizontal ERDI is very smooth and pretty flat, there is zero horizontal directivity mismatch. The vertical really can't be solved any better than it is in this format. Anything that might be done actively to this speaker could also be put in a convolution filter at the input and the likelihood of those changes having a significant effect is low.

Whether anyone likes the choices and compromises JBL made here is one thing but I see little to no opportunity to make a better speaker out of these parts.

index.php
 

fredoamigo

Addicted to Fun and Learning
Forum Donor
Joined
Mar 11, 2018
Messages
604
Likes
998
Location
South East France
While I do like this JBL model already as it is, I suspect it could be greatly improved if they they simply switched over to a more integrated “active” speaker and DSP design. We know they’re capable of this given the M2 and the 705/708 studio monitor models. But, JBL probably knows the audience market well enough where most of their audiophile buyers nevertheless will likely prefer something passive in the end.
of course, under the transition zone the eq will be almost essential and there is some correction to bring towards 700hz although I do not think that it is very audible but the LW in the zone of 2KHZ highly sensitive and audible and very flat what would have perhaps not had places with a design 3. ways.
it is also for this reason that I chose these speakers.
on the other hand, when I invest a certain amount of money in speakers, I like to know that they have a certain value if I sell them one day which is not at all the case for the other cheaper alternatives.
 

Zvu

Addicted to Fun and Learning
Joined
May 1, 2020
Messages
830
Likes
1,416
Location
Serbia
Simply - not true!

Eighteen Sound NSD1424BTN:

Radian 835BP
...I don't have time to extract all measured graphs to show that you are wrong for all drivers from the links I posted, but I think above two are enough to prove my point.

You seem offended. I really don't know why. Here are measurements of compression drivers from links you've provided. I named the pics to be able to discern what compression drivers are measurements for.


Celestion CDX1-1732.png CelestionCDX14-3040.png EighteenSound-ND1TPCompDriver.png EminenceN314XTeXtremeCP.png ObertonND45CompressionDriver.png Oberton-ND72CT-HBCompDriver.png LaVoce-DF10172K.png


Every one of posted goes above 3 percent and i was generous with some because they distort wildly above 10kHz. Al measured at 104dB at 1m distance.

....
But we can compare it to the measurements of many professional loudspeakers from tests in Production Partner magazine, for example Voice Acoustics CXN16, which uses B&C DE980TN compression driver - from 2 Hz to 10 kHz it measured less than 3% at the SPL=118 dB THD (not only 2nd harmonic!):

Ok, so we now know that only that B&C DE980TN measures good regarding 2nd harmonic, over 10kHz at 112-118dB / 1m.

I proved that BMS 4590 has one of the worst distortion measured from 7 kHz and up - so as far I am concerned, that case is closed. But I also wrote that it is one of the best drivers for frequencies between 600 Hz and 6 kHz...

Sorry but you've proven nothing actually. You just have claims and opinions. Proves are in the measurements which you obviously don't have.

Yes, JTR Noesis 212RT is an excellent loudspeaker with much lower price than JBL 4367 and it measured as good, if we don't count distortion from 7 kHz and up. Does it has any consequences in sound quality above 7 kHz at SPL levels usual for home listening, I don't know. Nor @Zvu.

Again you have presumptions about someone you know nothing about. Your presumptions are false yet again. I've actually listened BMS 4592Nd in normal listening room at my friend's house, here in Belgrade. Loudspeakers utilizing it are Oris Swing by BD Design. Some info on the link:


BMS 4592 indoor sounds absolutely marvelous and it easily qualifies as one of 3 best sounding systems i've ever heard. Having listened the system in more than few occasions is one of the reasons i was interested in what you have to say since i thought you also have some experience (measurement wise) with it. I think i might have made a mistake.

Edit:
Actually, I asked if there is a 15"+horn competitor to JBL 4367:

I may be overly pedantic/nitpicking, but in a view of comparing apples to apples, JTR Noesis 212RT is not 15"+horn (2-way), but 2x12"+coaxial horn (3-way). Anyway, if top two octaves are with the same sound quality as JBL 4367, we have a winner.

You're right, you're nitpicking since using 15" woofer has no advantage (maybe even small disadvantage? - depends on implementation) compared to 2x12". But i would agree, 212RT is superior loudspeaker being three way, more sensitive and last-but-not-least for far less money.
 
Last edited:

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
I would question that. The majority of the response fits within a +/- 1.5dB window which for production tolerance is great. The Horizontal ERDI is very smooth and pretty flat, there is zero horizontal directivity mismatch. The vertical really can't be solved any better than it is in this format. Anything that might be done actively to this speaker could also be put in a convolution filter at the input and the likelihood of those changes having a significant effect is low.

Whether anyone likes the choices and compromises JBL made here is one thing but I see little to no opportunity to make a better speaker out of these parts.

index.php

I do agree that there is very little if any improvements to be made on the frequency response. Though I also mean other benefits of active DSP designs like ability to provide some kind of tailored limiter protection (not that I expect many to push these in a home environment) and internal programmable EQ filter presets. But, then again, I’ve already guessed that the audience for this sort of design likely have other priorities.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
it is also for this reason that I chose these speakers.
on the other hand, when I invest a certain amount of money in speakers, I like to know that they have a certain value if I sell them one day which is not at all the case for the other cheaper alternatives.

Well, I fully respect that… and it’s only my own personal preference for something that’s with an active DSP design from the start.
 

Absolute

Major Contributor
Forum Donor
Joined
Feb 5, 2017
Messages
1,084
Likes
2,125
Measurements can tell a lot, when enough information is presented. Here, the top speaker is overall better, because it has better pattern control and better directivity, bottom speaker will likely sound bright.

The difference is caused by wider baffle for the top speaker, at lower freq, then the horn has better directivity in mid and up.

About the "matching-woofer-to-horn" at crossover: This can not be done when you have a horn with very different pattern from the direct radiator woofer. And the woofer does not suddenly start to beam at some frequency, the radiation pattern changes gradually. Fortunately, it does not need to be perfect to be able to present reasonable sound quality.

This is the classic old-school. Big woofer, horn on top. Usually comes with great dynamic capacity and much better directivity compared to the typical small and narrow speakers. Add a decent bass-system for below 100hz or so, mount a high-quality horn tweeter on top, if you still can hear the highs, use dsp to put it together.
Ignoring the stuff below the waveguide area where the M2 looks to have a series of resonances that ruins the dispersion pattern, I think the waveguide behavior of M2 looks better overall. Both have off-axis rise from about 1 khz to around 3 khz, but the M2 less so and avoids the high-Q peaks of the 4367. The latter also falls off far too rapidly and in general looks very weird on this graph compared to M2;


JBL%204367_360_Horizontal_Polar.png


JBL%20M2_360_Horizontal_Polar.png


With the same-ish drivers on both woofer and tweeter I find this to be a very interesting case-study on the effects of cabinet and waveguide construction. I wish we had normalized off-axis graphs for these speakers because that would make it easier to compare in a like-for-like situation.

Some people claim that limited dispersion vertically may be preferable over wider dispersion due to our high vertical reflection sensitivity. Thoughts on that? In that same line, may we actually prefer less dispersion up top vs wide?

Imagine if we could listen to 4367 and M2 back-to-back with EQ'D on-axis to be the same, that would be interesting indeed.
 

Vladimir Filevski

Addicted to Fun and Learning
Joined
Mar 22, 2020
Messages
550
Likes
713
Every one of posted goes above 3 percent and i was generous with some because they distort wildly above 10kHz. Al measured at 104dB at 1m distance.
There is no data above 10 kHz (if higher, 2nd harmonic will be higher than 20kHz, so it is not relevant).
For frequencies below 10 kHz, you have to look again and compare distortion, for example at 7 kHz, where BMS 4590 has very high 6.3% 2nd harmonic distortion at SPL=112dB/1m.
True, all compression drivers from the Voice Coil tests were measured at 104 dB. But:
Sorry but you've proven nothing actually. You just have claims and opinions. Proves are in the measurements which you obviously don't have.
I posted measurements from Production Partner magazine for one particular B&C compression driver - that is enough. If you want to see how many other compression drivers are better than BMS 4590 (hint: many!) in the two top octaves, you have to subscribe to Production Partner magazine.

BMS 4592 indoor sounds absolutely marvelous and it easily qualifies as one of 3 best sounding systems i've ever heard. Having listened the system in more than few occasions is one of the reasons i was interested in what you have to say since i thought you also have some experience (measurement wise) with it. I think i might have made a mistake.
I didn't take measurements of 4590 at low SPL levels (as in home), nor I have listened to it in those conditions. But at regular sound reinforcement levels (read: high SPL), the BMS 4590 has unmistakable sound signature at the top two octaves. Maybe that is one of the reasons why BMS is offering the same driver without his high-frequency ring-driver, as midrange-only drivers 4591 and 4592ND-mid.
As I wrote before, BMS 4590 is one of the best drivers in the world in the frequency range 600 Hz - 6 kHz.
 
Last edited:

Music1969

Major Contributor
Joined
Feb 19, 2018
Messages
4,641
Likes
2,809
Jeeze, I thought this was the affordable model that some were talking about on forums.

Then I looked at the price :oops: and realise I confused this for the 4306 :D
 

Tom C

Major Contributor
Joined
Jun 16, 2019
Messages
1,501
Likes
1,370
Location
Wisconsin, USA
I’m struggling with the concept that the 4367 could or should be available in an active version with DSP. It in fact is, marketed as the M2. True, M2 is not in a beauty cabinet. But the cabinet is larger (which can be both good and bad, but bigger adds cost over smaller), and the price is lower. A pair of M2’s without the matched Crown DSP/amp is $12,000 at full retail. Yeah, you don’t get the analogue crossover either, but that’s the whole point of an active system. So you don’t pay for what you don’t want, and you don’t have to rip anything out and rewire.
I chose 4349 because of price, and I think larger for my medium sized room might be overkill. For myself, at these prices, I’d go full M2 over 4367. Of course, tastes differ, so JBL offers both M2 and 4367. They could offer a powered 4367, but increasing the number of available options increases production costs further, so their choices seem reasonable to me.
 

Kvalsvoll

Addicted to Fun and Learning
Audio Company
Joined
Apr 25, 2019
Messages
878
Likes
1,643
Location
Norway
Ignoring the stuff below the waveguide area where the M2 looks to have a series of resonances that ruins the dispersion pattern, I think the waveguide behavior of M2 looks better overall. Both have off-axis rise from about 1 khz to around 3 khz, but the M2 less so and avoids the high-Q peaks of the 4367. The latter also falls off far too rapidly and in general looks very weird on this graph compared to M2;


JBL%204367_360_Horizontal_Polar.png


JBL%20M2_360_Horizontal_Polar.png


With the same-ish drivers on both woofer and tweeter I find this to be a very interesting case-study on the effects of cabinet and waveguide construction. I wish we had normalized off-axis graphs for these speakers because that would make it easier to compare in a like-for-like situation.

Some people claim that limited dispersion vertically may be preferable over wider dispersion due to our high vertical reflection sensitivity. Thoughts on that? In that same line, may we actually prefer less dispersion up top vs wide?

Imagine if we could listen to 4367 and M2 back-to-back with EQ'D on-axis to be the same, that would be interesting indeed.
Those speakers are different enough that they will sound different, even when eq-ed to exactly same response. Below 1K it is the baffle that makes the difference, a wider, larger baffle makes for significant improvements in radiation pattern. Then the horn is different, which one is better may not be obvious, they seem to have different trade-offs.

Vertical pattern control below 1K is limited, due to the single direct radiation configuration, with no horn or acoustic radiation control. Vertical is very important for a speaker, and the best pattern is different in vertical direction compared to horizontal.
 

Purité Audio

Master Contributor
Industry Insider
Barrowmaster
Forum Donor
Joined
Feb 29, 2016
Messages
9,051
Likes
12,150
Location
London

Bjorn

Major Contributor
Audio Company
Forum Donor
Joined
Feb 22, 2017
Messages
1,286
Likes
2,562
Location
Norway
Last edited:

NYfan2

Active Member
Forum Donor
Joined
Jan 10, 2021
Messages
209
Likes
446
Location
Netherlands
Reading the review on stereophile:
Yet, after about 200 hours of play, the 4367s had improved dramatically, and after 300 they began to sound great. I later learned that the pair I received had only a few hours on them. Mystery solved. :facepalm:

So I skipped that part and looked at the measurements.
Conclusion of JA: —it offers textbook measured performance.—
 

charleski

Major Contributor
Joined
Dec 15, 2019
Messages
1,098
Likes
2,239
Location
Manchester UK
As much as I appreciate the measurements from Stereophile, the reality is that, especially the off-axis measurements and how they are portrayed are completely outdated. He could do much better with a simple software like Dayton Omnimic.
It’s certainly true that the measurements look quite dated these days. But I think the one major advantage of not changing his technique is that it allows direct comparison across the large corpus that he has built up over the years. This is a significant factor, as trying to extrapolate between different measurement techniques always ends up with assumptions that might confound things.
 

Bjorn

Major Contributor
Audio Company
Forum Donor
Joined
Feb 22, 2017
Messages
1,286
Likes
2,562
Location
Norway
It’s certainly true that the measurements look quite dated these days. But I think the one major advantage of not changing his technique is that it allows direct comparison across the large corpus that he has built up over the years. This is a significant factor, as trying to extrapolate between different measurement techniques always ends up with assumptions that might confound things.
It's more important to show measurements in a good way rather then comparing with graphs that are hiding too much. But one could of course do both. Add at least a proper view of the polars.
 

Purité Audio

Master Contributor
Industry Insider
Barrowmaster
Forum Donor
Joined
Feb 29, 2016
Messages
9,051
Likes
12,150
Location
London
As much as I appreciate the measurements from Stereophile, the reality is that, especially the off-axis measurements and how they are portrayed are completely outdated. He could do much better with a simple software like Dayton Omnimic.
Totally agree I guess that reflects the magazines attitude towards measurement , only under duress.
Keith
 

charleski

Major Contributor
Joined
Dec 15, 2019
Messages
1,098
Likes
2,239
Location
Manchester UK
It's more important to show measurements in a good way rather then comparing with graphs that are hiding too much. But one could of course do both. Add at least a proper view of the polars.
Adding a more modern form of data visualisation as an addition to his standard set would be nice, but that's all. Given the large body of measurements he's built up over more than 30 years it's far more important to maintain consistency. This is especially true given the numerous ways in which such measurements can be confounded, and I'd have thought this point would be rather obvious.

As others develop similar bodies of results using more modern techniques there will, hopefully, be sufficient overlap in the data sets to allow meaningful comparisons. But this critically depends on each body of measurements being conducted in a consistent manner rather than jumping around.
 
Status
Not open for further replies.
Top Bottom