JBL 4349 Review (Studio Monitor Speaker)

[levimax]

I have a difficult time believing that the opposite can also be true, ie. measure bad and sound good.

Some highly efficient horn based speakers can measure rather poorly on FR yet sound very impressive. I think "dynamics" (which I will define as the ability to play at high SPL without distortion to accommodate transients in dynamic music) is the reason.

 

[pos]

I am pretty upset reading this review. What was JBL thinking when designing this speaker?
A 1.5kHz crossover (or is it 1.7kHz like initially published?) is way too high for this woofer, and the result was to be expected.

The 1200Fe (or whatever variation this speaker is using) is a great woofer, one of the best for this size, but it was never intended to play this high, and that is clearly visible in the tolerance boundaries for the frequency test procedure found in its EDS:
http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=33374&stc=1&d=1216860198
A +/- 3dB unit-to-unit variation is allowed above 1356Hz, which is pretty normal given the random breakups that can occur with such a cone, but this is clearly no place for a crossover. This means that anomalies found with the unit tested here might not be there with another unit, or might be worse. Anything can happen exactly where everything should be under tight control.

And even if that tolerance issue was solved (might be the purpose of this particular variation, but that would imply pushing breakups modes further up: good luck doing that with a paper cone of this size), there is no way it would play nice with a compression driver up there.
I used to have a pair of 4425, employing an - arguably not as good - 12" woofer crossing to a compression driver at 1.2kHz, and the transition was very audible. They solved the problem with the 4429 going as low as 800 or 900Hz with the 1200Fe, pairing it with a small metal dome compression driver and a wide directivity horn to match directivity.
Now we are back to a 1.5kHz crossover, Eon territories with premium drivers...

I get it that the D2415K cannot play low, but then they should have used another compression driver for the task and crossover no higher than 1kHz (and preferably lower).
The 175Nd in the 4429 had no issue there, as did the poor 2416H in the 4425 at 1200Hz, and those were clearly not premium drivers at the time.
The new dual ring radiators are pretty good when it comes to high SPL (as seen in the Vertec and other JBL PA speakers) and HF, but it is notorious that they are struggling at low frequencies (eg the D2430K, which shows increased distortion under 2kHz compared to a classic 4" diaphragm compression driver), so this is no wonder that half the radiating area for the D2415K compared to the D2430K would require such a high crossover frequency. JBL does not do dome compression drivers anymore, and even had to use Radian drivers in the new Everest, but these ring radiators are also probably much cheaper to produce compared to metal dome drivers, so they should have just used the D2430K in the 4349, and call it a day.

This is shameful, how can they market this in the synthesis lineup with such compromises in the design??
I do have a pair of 1200Fe and D2415K here, but I would clearly never intend to make them play together.

These speakers probably sound good overall, they better do with the drivers at hand, but the result should be textbook perfect. Anything less than that is an insult to all the efforts and years of evolution that went into the components and technologies used here.

If only Greg Timbers was still in command there, I bet this would have turned out quite differently...

Well... sorry for the rant
I'd like to say that I feel better now, but frankly I don't...

 

[restorer-john]

@amirm Great review.

They look to be a speaker worth the money IMO. They'd have a good resale value, parts are always available and they look good, measure well, handle a ton of power and are efficient to boot.

 

[DualTriode]

Shout out to @pos,

Thanks,

At another realm you posted the values of the official JBL L-Pad and series capacitor(s) for the M2 tweeter.

With the help of an APx555 analyzer and APx1701 Transducer Interface I plotted the impedance curve of the JBL M2 Compression Driver + waveguide.

The impedance curve is much flatter (tamer) with the L-pad wired with the CD + waveguide.

Also the L-pad reduces the noise at the waveguide output by 9dB. My wife would say power to burn. I say that the Crown amplifier has plenty of watts. I like the reduced hiss and tweeter protection.

Note the saw tooth nature of the impedance curve (no processing or smoothing) that is typical Compression Driver performance, domes do not do this.

Thanks DT

 

[Asinus]

Why are most high end JBLs limited to 2 way designs?

Currently all high-end JBLs are 3-way (summit series), a single Everest costs more than a pair of Salon2 and the measurements aren't as good, but are made for live concert levels.
The monitor series are the ones that favor 2-way with compression driver + horn waveguide and I think it is in part to ride the M2 coattails and part to keep the BOM in check.

 

[Newman]

As indicated above, objective test results left a sour taste in my mouth post measurements. It was not until I listened to the speakers that I got what this speaker is all about:

OTOH Toole has shown the strong likelihood that, if exactly the same flawed sound waves came from a small cheap speaker with other less favourable non-sonic attributes, your subjective impressions would have been just as poor as the measurements.

cheers

 

[amirm]

OTOH Toole has shown the strong likelihood that, if exactly the same flawed sound waves came from a small cheap speaker with other less favourable non-sonic attributes, your subjective impressions would have been just as poor as the measurements.

I hear you but cost is not a factor. One minute I am testing a $100 speaker, another a $7,500 one. I have given higher marks to JBL Stage 130 for example despite it being very cheap. I am also not sold on the look of the 4349 speaker. I know it has fans but I am not one.

Really, we have been around this watering hole a million times. As I keep explaining, I provide my opinion but you don't have to read it. Your doubts also don't amount to facts.

 

[ROOSKIE]

In past loudspeaker measurements, Amir was criticized (rightly, in my opinion) for not following the manufacturer's specifications when choosing the reference axis.
In this case, Amir did it right and followed the manufacturer's instructions.

If after the measurements it turns out that another vertical reference axis could provide a better result, then the manufacturer is to blame if the manual is faulty or sloppy.
It is not the reviewer's job to present a speaker as favorably as possible, the manufacturer should pursue that in their own interest.

It's really a bit much to expect Amir to think before starting the measurement, "reference axis where?...manual says tweeter axis...surely a marketing guy wrote the manual...it's a speaker with a big waveguide...the speaker designer must have tuned the speaker to a reference axis between horn and woofer...yep, okay that's it! Screw the manual!"

Yes, but we want to know if there is a better way to use the product.
If discovering a better axis happens then why on earth would we not want to know that?
Again as I clearly stated I understand the logistics and dont expect any retest or Amir to constantly adjust things but in the end refusing exploring the options is never science and we ought to at least discuss the validity of the potential finding in the comments area.
And really if you think about it.
What if a manual had a misprint and said face the speaker toward the wall?
Anyway. Just saying that it makes no sense to refuse to improve the "manual".

 

[richard12511]

I actually said "-3 dB @ 70 Hz so subwoofer support required," not no subwoofer support required. Sacrificing bass extension and recommending only for use with sub(s) is the trade-off for increased efficiency and dynamics at a reasonable price point.

Many are curious about how PSA's speakers would measure. If @amirm or @hardisj wanted to test one PSA has a 60 day trial period so they could buy one, return it to PSA after testing for a refund and only have to pay for return shipping.

PSA and Chane are two that I'd really like to see. Chane has a couple new lines out.

 

[Valentin R]

Yes, but we want to know if there is a better way to use the product.
If discovering a better axis happens then why on earth would we not want to know that?
Again as I clearly stated I understand the logistics and dont expect any retest or Amir to constantly adjust things but in the end refusing exploring the options is never science and we ought to at least discuss the validity of the potential finding in the comments area.
And really if you think about it.
What if a manual had a misprint and said face the speaker toward the wall?
Anyway. Just saying that it makes no sense to refuse to improve the "manual".

Yes I second this thought

And the hole spin will be more representative of what Amir heard

My 2hz

 

[amirm]

Quick note on changing the reference axis. I can't tilt the angle like you all can visually by using the spin data. I only have controls in Z which I can lower to be something else. In other words, the on-axis is at 90 degrees to the speaker and I can only move it up and down. Is this what you all want? If so, where should that point be?

 

[MZKM]

Quick note on changing the reference axis. I can't tilt the angle like you all can visually by using the spin data. I only have controls in Z which I can lower to be something else. In other words, the on-axis is at 90 degrees to the speaker and I can only move it up and down. Is this what you all want? If so, where should that point be?

The quick calculations I did in an earlier comment show that to get the 10° vertical offset, there is some good amount of distance. Like even 12” lower may not be enough. But if the cabinet is 29" tall, then a foot lower may be around the center of the woofer, so why not.

 

[richard12511]

I am pretty upset reading this review. What was JBL thinking when designing this speaker?
A 1.5kHz crossover (or is it 1.7kHz like initially published?) is way too high for this woofer, and the result was to be expected.

The 1200Fe (or whatever variation this speaker is using) is a great woofer, one of the best for this size, but it was never intended to play this high, and that is clearly visible in the tolerance boundaries for the frequency test procedure found in its EDS:
http://www.audioheritage.org/vbulletin/attachment.php?attachmentid=33374&stc=1&d=1216860198
A +/- 3dB unit-to-unit variation is allowed above 1356Hz, which is pretty normal given the random breakups that can occur with such a cone, but this is clearly no place for a crossover. This means that anomalies found with the unit tested here might not be there with another unit, or might be worse. Anything can happen exactly where everything should be under tight control.

And even if that tolerance issue was solved (might be the purpose of this particular variation, but that would imply pushing breakups modes further up: good luck doing that with a paper cone of this size), there is no way it would play nice with a compression driver up there.
I used to have a pair of 4425, employing an - arguably not as good - 12" woofer crossing to a compression driver at 1.2kHz, and the transition was very audible. They solved the problem with the 4429 going as low as 800 or 900Hz with the 1200Fe, pairing it with a small metal dome compression driver and a wide directivity horn to match directivity.
Now we are back to a 1.5kHz crossover, Eon territories with premium drivers...

I get it that the D2415K cannot play low, but then they should have used another compression driver for the task and crossover no higher than 1kHz (and preferably lower).
The 175Nd in the 4429 had no issue there, as did the poor 2416H in the 4425 at 1200Hz, and those were clearly not premium drivers at the time.
The new dual ring radiators are pretty good when it comes to high SPL (as seen in the Vertec and other JBL PA speakers) and HF, but it is notorious that they are struggling at low frequencies (eg the D2430K, which shows increased distortion under 2kHz compared to a classic 4" diaphragm compression driver), so this is no wonder that half the radiating area for the D2415K compared to the D2430K would require such a high crossover frequency. JBL does not do dome compression drivers anymore, and even had to use Radian drivers in the new Everest, but these ring radiators are also probably much cheaper to produce compared to metal dome drivers, so they should have just used the D2430K in the 4349, and call it a day.

This is shameful, how can they market this in the synthesis lineup with such compromises in the design??
I do have a pair of 1200Fe and D2430K here, but I would clearly never intend to make them play together.

These speakers probably sound good overall, they better do with the drivers at hand, but the result should be textbook perfect. Anything less than that is an insult to all the efforts and years of evolution that went into the components and technologies used here.

If only Greg Timbers was still in command there, I bet this would have turned out quite differently...

Well... sorry for the rant
I'd like to say that I feel better now, but frankly I don't...

Most of these questions have me wondering, why not just go 3 way? It seems that JBL really tries to avoid anything more than 2 way. This is in contrast to Revel/Infinity, and I wonder why.

 

[More Dynamics Please]

PSA and Chane are two that I'd really like to see. Chane has a couple new lines out.

The Chane 752 is an interesting big bookshelf for $650. Two 8" Scanspeak woofers with Italian Eighteen Sound titanium compression driver/60x40 aluminum horn in a sealed cabinet rated at 92 dB sensitivity and -3 dB @ 50 Hz. A little more bass extension and a little less efficiency than PSA likely due to the woofers not being pro drivers.

 

[MZKM]

@amirm, quick question for you.
Are your "hand drawn" lines intended to be a trendline or are they intended to relay a slope from bass to highs? When you first started reviewing I thought the purpose was to draw a -1dB/octave slope from LF to HF and I hadn't paid much attention to it deviating from that until I saw the above graphic. I assume from this graphic, that you're drawing a trendline instead of what I thought originally. The reason I ask, though, is because up until now it seemed your trend lines started with the bass and trended down to the HF. Meaning, in this case of this speaker, I'd have expected your line to start ~ 90dB @ 100Hz rather than 86dB. But that may just be because the other ones I'm referencing below have had less of a bass bump and a more linear trend from LF to HF. Again, I'm just asking to make sure I understand this is the intent. Not starting or picking fights.

He hand draws them to where they fit best.

The formula does a log trend line from 100Hz-10kHz:

 

[jhaider]

The Chane 752 is an interesting big bookshelf for $650. Two 8" Scanspeak woofers with Italian Eighteen Sound titanium compression driver/60x40 aluminum horn in a sealed cabinet rated at 92 dB sensitivity and -3 dB @ 50 Hz.

Did ScanSpeak revive the 1980s (or earlier) Vifa drivers? That flange doesn’t look like part of their current line.

 

[milosz]

One can buy better speakers for $7,500. Loads of profit margin in these I'd guess.

 

[More Dynamics Please]

Did ScanSpeak revive the 1980s (or earlier) Vifa drivers? That flange doesn’t look like part of their current line.

ScanSpeak has quite a few 8" drivers listed on its website. Not sure which one the Chane 752 uses. The images on the Chane website could be of a prototype.

 

[jhaider]

ScanSpeak has quite a few 8" drivers listed on its website. Not sure which one the Chane 752 uses. The images on the Chane website could be of a prototype.

Not that many actually, and I didn’t see the 1980s era Vifa P21 on their website. However and interestingly, Madisound does seem to carry a Scanspeak branded 1980s Vifa woofer.

https://www.madisoundspeakerstore.com/approx-8-woofers/scanspeak-classic-p21wo20-8-woofer-poly-cone/

They must have bought a few pallets of them at good pricing, because there’s basically nothing better about that one compared to current ScanSpeak Discovery. (@hardisj measured that one for his old site; the ones he measured are in my basement right now). Then again Chane seems to be on a retro/closeout kick right now. They have David Smith’s 1990s Snell subwoofer for sale as well.

 

[Valentin R]

Quick note on changing the reference axis. I can't tilt the angle like you all can visually by using the spin data. I only have controls in Z which I can lower to be something else. In other words, the on-axis is at 90 degrees to the speaker and I can only move it up and down. Is this what you all want? If so, where should that point be?

It would be my choice center between unit centers

Here is an image with scale : 29.9cm from the top