Built quality JBL 705 i speakers

[jhaider]

It seems likely that the poor build quality (or consistency) actually is affecting the sound quality, because the independent measurements I've recently been shown for these speakers show severe flaws:
...

It's hard to imagine how this happened and got past the design phase from a company that prides itself in making speakers that measure well as per Harman research. So how do you think this happened?

I get why someone relatively new at trying to correlate audio measurements to sound quality would get hysterical about this notch. It is a big feature, and looks scary. However, in practice it is much ado about very little because it is so narrow and a dip rather than a peak. Most measurements will hide such resonances. You see it because these data are unsmoothed. I know from my own measurements that it completely disappears in a 1/12 octave smoothed measurement.

One need not imagine anything, because the physics resulting in that notch are easy to comprehend: large port tuned low in a very small cabinet will result in pipe resonances. That compromise is unavoidable without swapping out the port for passive radiators and upping the parts cost considerably. The larger model is larger but not tuned much lower, so it has correspondingly less pipe resonances.

Also, this notch is shown in Harman's own data, so it is obviously not a production consistency issue.

Might it be a good idea to stuff the port with foam if you're running bass management? Sure, go ahead if you want. That should kill the resonance. I can't be bothered with mine. However, I did recently pick up a BSS BLU-USB and I'm very keen to set up a comparison between 705i using the BLU-USB and BLU-50, and Neumann KH80.

 

[Blumlein 88]

Don't know if the 750 hz null is normal or not. I did just try putting a 12 db 750 hz notch in a few tracks. It is pretty audible when I can switch back and forth. Likely audible even in isolation I think on recordings you are familiar with. If you make it narrow enough, you don't hear it anymore. At 100 hz wide (which appears to match the graphs) you could hear it.

 

[echopraxia]

I get why someone relatively new at trying to correlate audio measurements to sound quality would get hysterical about this notch. It is a big feature, and looks scary. However, in practice it is much ado about very little because it is so narrow and a dip rather than a peak.

One need not imagine anything, because the physics resulting in that notch are easy to comprehend: large port tuned low in a very small cabinet will result in pipe resonances. That compromise is unavoidable without swapping out the port for passive radiators and upping the parts cost considerably. The larger model is larger but not tuned much lower, so it has correspondingly less pipe resonances.

If it’s unavoidable, then why don’t we see the same flaw in the measurements of other similarly sized and priced speakers?

Might it be a good idea to stuff the port with foam if you're running bass management? Sure, go ahead if you want. That should kill the resonance. I can't be bothered with mine. However, I did recently pick up a BSS BLU-USB and I'm very keen to set up a comparison between 705i using the BLU-USB and BLU-50, and Neumann KH80.

Why are we talking about DIY remedies for this problem? I thought you just said above that I’m being hysterical about something that’s not audible

So, which is it:

A. Is this 10db null an audible problem that should be manually fixed by stuffing bits of foam in a $2000 pair of speakers to fix a problem that arguably shouldn’t have been there to begin with (given that other similar speakers don’t have this problem)?​

​

B. Or, is it genuinely not an audible problem, and therefore no port stuffing remedies necessary, and Blumlein 88 is hallucinating things with his equalization experiment?​

P.S. To be clear, this discussion is interesting and all here is meant lightheartedly in good fun. No offense intended by anything above. In seriousness, I’m curious to see the consensus we converge to on this.

 

[jhaider]

If it’s unavoidable, then why don’t we see the same flaw in the measurements of other similarly sized and priced speakers?

Do you know of a another similarly compact speaker comparable bass extension and SPL? Those are the key factors here - small cabinet, low tuning, strong output capability.

The 705 gets away with pushing a small woofer so low because the woofer is a really solid piece of engineering. Excursion is roughly double Neumann KH120's woofer, based on the SPL limits measured in S&R. The only similarly performant 5" woofer known to me is ScanSpeak Illuminator. It costs more than $330 per woofer.

Why are we talking about DIY remedies for this problem? I thought you just said above that I’m being hysterical about something that’s not audible

You misinterpret. I only mentioned doing stuffing the port with foam to soothe one's nerves in a specific use case where the extension the port provides is not required. I'm not replying to the rest of your post because it seems calculated to provoke rather than elicit fruitful discussion.

 

[echopraxia]

Do you know of a another similarly compact speaker comparable bass extension and SPL? Those are the key factors here - small cabinet, low tuning, strong output capability.

The 705 gets away with pushing a small woofer so low because the woofer is a really solid piece of engineering. Excursion is roughly double Neumann KH120's woofer, based on the SPL limits measured in S&R. The only similarly performant 5" woofer known to me is ScanSpeak Illuminator. It costs more than $330 per woofer.

Before we go there, I think we need to establish whether or not a 10dB dip like this is audible. If it is, then I would avoid any design that has such a flaw at this price range, even if they are trading such a flaw to achieve greater bass SPL. In such a case, you can choose quantity over quality if you wish, but that certainly wouldn‘t be my preference.

But all this is moot if a 10db dip of this bandwidth at 750hz is inaudible, so we should very carefully and accurately verify that. I’m not sold on your claim yet because, again, (1) most other speakers do not have this problem, and (2) someone else just tried simulating this and claims it’s audible.

If it’s audible, then yes this 10db null counts as a severe problem with this speaker. If it’s not audible, then it’s all good.

You misinterpret. I only mentioned doing stuffing the port with foam to soothe one's nerves in a specific use case where the extension the port provides is not required.

I don’t care about soothing nerves; I care about determining which speakers are objectively flawed vs superior to others.

I also do not wish to compromise quality in exchange for greater SPL: You have every right to pursue that road if you wish, but I would definitely not call such a goal aligned with those of “audiophile” or “high fidelity” audio.

I’d rather achieve both quality and quantity, with subwoofers crossed with speakers designed to specialize in their range of frequencies without compromise.

I'm not replying to the rest of your post because it seems calculated to provoke rather than elicit fruitful discussion.

You don’t have to respond to anything you don’t want to, but I’m a little worried you still would take it that way even after I specifically added a disclaimer at the bottom clarifying that my tone is intentionally just poking fun back at you lightheartedly. Forum rules, after all: “Come here to have fun, be ready to be teased and not take online life too seriously.”

This was in response to you calling me “hysterical“. I don’t mind you poking fun at me, as long as I can poke fun back. But this is kind of weird when you can dish it out the fun-poking, but can’t take it when it comes back at you

That said, if I’ve crossed a line somewhere, please quote it specifically and I’d be glad to discuss in full “serious mode”. I don’t think anything about my previous post was rude though, so I’d love to hear what the problem is.

 

[jhaider]

Before we go there, I think we need to establish whether or not a 10dB dip like this is audible.

Go ahead, then. Find an EQ capable of rendering a notch so narrow that it is invisible in 1/12-octave smoothing, take careful acoustic measurements to confirm the acoustic results match the EQ's display, present your test protocol for review, and have at it.

To give you a running start, my guess of the EQ required to mimic the notch, based on the Sound und Recording measurements and a quick simulation in the miniDSP 2x4HD app, is Q≈30 (or ~1/20-octave wide depending on your EQ) and gain ≈-9.75. Go ahead and round off the gain to -10.

I’m not sold on your claim yet because, again, (1) most other speakers do not have this problem,

You want to know part of why I find your tone distasteful? It's this kind of crap.

First, you don't know that. You're making a wildass guess based on comparative measurements taken under much less controlled conditions, or with measurement conditions ill-specified, without understanding that's what you're doing. For example, I was curious about the speakers in your signature, so I looked them up and was pleasantly surprised to find this page: http://www.ascendacoustics.com/pages/products/speakers/SRT/srtmeas.html

• The manufacturer posts some measurements - good! A Spinorama should be the floor in 2019, but this page looks considerably older. I would also like to see a horizontal polar map, though unfortunately nobody does them today except for Princeton 3D3A, Prof. Goetz at Sound und Recording, James Larson at Audioholics, and me. Regardless, they present more data than most companies do.
• The text spins the reader away from notable flaws - fair because it's marketing! The spin here is claiming a "lack of tweeter bloom" based on a very confusing figure, even though the preceding figure clearly shows a dispersion disruption where the tweeter comes in.
• The measurement conditions are inadequately specified - bad! Does the marketing team thinks the brand of automatic turntable is Very Important, but smoothing, length of the gate, or nearfield bass splicing technique (I assume splicing based on viewing the measurement) are trifles? Am I wrong? Search for "smooth," "oct," "gat," "ms," or "splic" at the above link to confirm.

Second, you seem to be willfully ignoring cabinet size, despite repeated invocations. The answer to the query "I want something like JBL 705, but without the compromises necessitated by 705's small size," is JBL 708!

I’d love to hear what the problem is.

In addition to the above, express fewer assumptions, and read more.

 

[Blumlein 88]

Go ahead, then. Find an EQ capable of rendering a notch so narrow that it is invisible in 1/12-octave smoothing, take careful acoustic measurements to confirm the acoustic results match the EQ's display, present your test protocol for review, and have at it.

To give you a running start, my guess of the EQ required to mimic the notch, based on the Sound und Recording measurements and a quick simulation in the miniDSP 2x4HD app, is Q≈30 (or ~1/20-octave wide depending on your EQ) and gain ≈-9.75. Go ahead and round off the gain to -10.



You want to know part of why I find your tone distasteful? It's this kind of crap.

First, you don't know that. You're making a wildass guess based on comparative measurements taken under much less controlled conditions, or with measurement conditions ill-specified, without understanding that's what you're doing. For example, I was curious about the speakers in your signature, so I looked them up and was pleasantly surprised to find this page: http://www.ascendacoustics.com/pages/products/speakers/SRT/srtmeas.html

• The manufacturer posts some measurements - good! A Spinorama should be the floor in 2019, but this page looks considerably older. I would also like to see a horizontal polar map, though unfortunately nobody does them today except for Princeton 3D3A, Prof. Goetz at Sound und Recording, James Larson at Audioholics, and me. Regardless, they present more data than most companies do.
• The text spins the reader away from notable flaws - fair because it's marketing! The spin here is claiming a "lack of tweeter bloom" based on a very confusing figure, even though the preceding figure clearly shows a dispersion disruption where the tweeter comes in.
• The measurement conditions are inadequately specified - bad! Does the marketing team thinks the brand of automatic turntable is Very Important, but smoothing, length of the gate, or nearfield bass splicing technique (I assume splicing based on viewing the measurement) are trifles? Am I wrong? Search for "smooth," "oct," "gat," "ms," or "splic" at the above link to confirm.

Second, you seem to be willfully ignoring cabinet size, despite repeated invocations. The answer to the query "I want something like JBL 705, but without the compromises necessitated by 705's small size," is JBL 708!



In addition to the above, express fewer assumptions, and read more.

Respectfully, I'm kind of with echopraxia here. I agree the size and volume of the cabinet are real constraints. But that isn't the purchaser's fault if it doesn't work right. The big thing we all need is for another measurement from another source to see if this is a one off quality issue or if it is endemic to the unit.

The standard to which Harman designs, and to which they helped make a standard is 1/20th octave resolution. So a 1/10th octave anomaly if normal seems like it might be a possible audible problem. 1/10th octave is a 100 hz wide null at 750 hz roughly. Could one damp the port without ruining the speaker, or duct it in another direction? Yeah, maybe. We don't hear dips as much as much smaller peaks, but dips can be heard. Especially in vocal and bass regions. 750 hz is in the female vocal region, the octave above males, and quite a number of instruments like guitars and such.

I've not heard the speaker, but wouldn't automatically dismiss this null if it is present will all the 705's.

I also notice near universal acclaim for 305 and 308, and for almost all Revel's. Yet the 705 and 708 seems less universally praised. I was on the verge of purchasing some until the users opinions started stacking up. Don't know if expectations were too high, or something is a bit too close to the line on these models or what? I'd very much like more measurement info to figure it out. I know only too well how memes and ideas can either blind us or send us off into criticism when there is nothing wrong.

 

[echopraxia]

You want to know part of why I find your tone distasteful? It's this kind of crap.

First, you don't know that. You're making a wildass guess based on comparative measurements taken under much less controlled conditions, or with measurement conditions ill-specified, without understanding that's what you're doing. For example, I was curious about the speakers in your signature,

First, I have no idea why you’re bringing up the speakers in my profile when I never invoked them for comparison here. But if you are interested in my towers, check out this link. The one you linked is of the standard version without the upgraded tweeter; the upgraded version measures significantly better.

Anyways, I would not recommend using that brand as an example in this argument, because a much easier comparison would be via identical test conditions, which is only really possible if comparing Harman speakers. So let’s do that. Unless the title “Harman Audio Test System” here is referring to two wildly different and “ill-specified test conditions”, I actually think it is quite fair to say that other speakers do not have this problem:

JBL LSR705i:

Revel M105:

Now, I understand that you’re trying to argue that my comparison doesn’t count unless the external dimension, bass extension, etc. matches exactly. But that’s not what I’m talking about here: I’m simply demonstrating that this flaw seen in the 705 is not common elsewhere, even when comparing using the exact same measurement system under controlled conditions.

That you seem to be arguing that this anomaly in the 705’s spinorama is not only acceptable but common, is quite bizarre. It’s definitely not common, and many others here have posted remarking on the strangeness of this result, to say the least.

Just to pre-emptively address this:

Second, you seem to be willfully ignoring cabinet size, despite repeated invocations

This is irrelevant. If a speaker has a flaw that shows up in measurements, a flaw is a flaw independent of cabinet size. If you’re claiming it’s impossible to build a speaker of a certain cabinet size without serious flaws like this, and if that were true, then that’s simply evidence that JBL made the mistake of choosing a cabinet size that can’t be implemented without flaws like this.

Lastly: Regarding tone, I’ll remind you that I’ve used polite wording and tone throughout my discussion with you, and have gone to extensive lengths to emphasize my good intentions here — despite you having used phrases like ”hysterical” and “this crap” to describe my posts. And yet you think I’m the one with a “distasteful tone”? In that case, I’m not sure there’s much else I can do to help you not be offended here. I suspect I’ve struck a nerve with my comments on expensive JBL speakers, which I can understand may not make owners of them feel very good.

 

[Blumlein 88]

First, I have no idea why you’re bringing up the speakers in my profile when I never invoked them for comparison here. But if you are interested in my towers, check out this link. The one you linked is of the standard version without the upgraded tweeter; the upgraded version measures significantly better.

Anyways, I would not recommend using that brand as an example in this argument, because a much easier comparison would be via identical test conditions, which is only really possible if comparing Harman speakers. So let’s do that. Unless the title “Harman Audio Test System” here is referring to two wildly different and “ill-specified test conditions”, I actually think it is quite fair to say that other speakers do not have this problem:

JBL LSR705i:

View attachment 38301

Revel M105:

View attachment 38304

Now, I understand that you’re trying to argue that my comparison doesn’t count unless the external dimension, bass extension, etc. matches exactly. But that’s not what I’m talking about here: I’m simply demonstrating that this flaw seen in the 705 is not common elsewhere, even when comparing using the exact same measurement system under controlled conditions.

That you seem to be arguing that this anomaly in the 705’s spinorama is not only acceptable but common, is quite bizarre. It’s definitely not common, and many others here have posted remarking on the strangeness of this result, to say the least.

Just to pre-emptively address this:



This is irrelevant. If a speaker has a flaw that shows up in measurements, a flaw is a flaw independent of cabinet size. If you’re claiming it’s impossible to build a speaker of a certain cabinet size without serious flaws like this, and if that were true, then that’s simply evidence that JBL made the mistake of choosing a cabinet size that can’t be implemented without flaws like this.

Regarding tone, I’ll remind you that I’ve used polite wording and tone throughout my discussion with you, and have gone to extensive lengths to emphasize my good intentions here — despite you having used phrases like ”hysterical” and “this crap” to describe my posts, you say I’m the one with a “distasteful tone”. In that case, I’m not sure there’s much else I can do to help you. I suspect I’ve struck a nerve with my comments on expensive JBL speakers, which I can understand may not make owners of them feel very good.

Good example. I'd read those charts and expect the M105 to be the better speaker. From what I can tell of Harman's method of judging so would they. Toole emphasizes smoothness in response, and those DI charts especially those first reflection DI charts. M105 is better here.

And that is an interesting question. I don't know I've seen any detail on how Harman decides which is better. They have a nearly perfect correlation with bookshelf speakers with no response below 100 hz and a very high correlation between full range speakers. But exactly how is that determined? Would their algorithm tell us the M105 ranks higher than the 705? We need the predicted in room response for both.

 

[Olli]

Another great bang for the buck at 800 USD/EUR seems to be the KEF R3.

In my use case (wall mounting) the 705 should still do better - it’s front ported, whereas Revel and KEF are rear ported.

 

[maty]

Another great bang for the buck at 800 USD/EUR seems to be the KEF R3.

View attachment 38308

In my use case (wall mounting) the 705 should still do better - it’s front ported, whereas Revel and KEF are rear ported.

800 USD/EUR... each -> pair: 1,600 USD/EUR

In EU/NA KEF R3 is a great option.
In NA, Ascend Sierra 2-EX RAAL is the other. http://www.ascendacoustics.com/pages/products/speakers/SRM2EX/srm2exmeas.html

 

[echopraxia]

800 USD/EUR... each -> pair: 1,600 USD/EUR

In EU/NA KEF R3 is a great option.
In NA, Ascend Sierra 2-EX RAAL is the other. http://www.ascendacoustics.com/pages/products/speakers/SRM2EX/srm2exmeas.html

Yes, and while we’re on that topic, the Sierra 2EX wins over the KEF R3 in a blind test here.

 

[gr-e]

First, I have no idea why you’re bringing up the speakers in my profile when I never invoked them for comparison here. But if you are interested in my towers, check out this link. The one you linked is of the standard version without the upgraded tweeter; the upgraded version measures significantly better.

Anyways, I would not recommend using that brand as an example in this argument, because a much easier comparison would be via identical test conditions, which is only really possible if comparing Harman speakers. So let’s do that. Unless the title “Harman Audio Test System” here is referring to two wildly different and “ill-specified test conditions”, I actually think it is quite fair to say that other speakers do not have this problem:

JBL LSR705i:

View attachment 38301

Revel M105:

View attachment 38304

Now, I understand that you’re trying to argue that my comparison doesn’t count unless the external dimension, bass extension, etc. matches exactly. But that’s not what I’m talking about here: I’m simply demonstrating that this flaw seen in the 705 is not common elsewhere, even when comparing using the exact same measurement system under controlled conditions.

That you seem to be arguing that this anomaly in the 705’s spinorama is not only acceptable but common, is quite bizarre. It’s definitely not common, and many others here have posted remarking on the strangeness of this result, to say the least.

Just to pre-emptively address this:



This is irrelevant. If a speaker has a flaw that shows up in measurements, a flaw is a flaw independent of cabinet size. If you’re claiming it’s impossible to build a speaker of a certain cabinet size without serious flaws like this, and if that were true, then that’s simply evidence that JBL made the mistake of choosing a cabinet size that can’t be implemented without flaws like this.

Lastly: Regarding tone, I’ll remind you that I’ve used polite wording and tone throughout my discussion with you, and have gone to extensive lengths to emphasize my good intentions here — despite you having used phrases like ”hysterical” and “this crap” to describe my posts. And yet you think I’m the one with a “distasteful tone”? In that case, I’m not sure there’s much else I can do to help you not be offended here. I suspect I’ve struck a nerve with my comments on expensive JBL speakers, which I can understand may not make owners of them feel very good.

Revel M105 is a much bigger speaker, and what's most improtant, it's rear ported. You don't need any measurements to know it doesn't have this issue.
Here's a measurement of another small front-ported speaker with a very low port tuning, iLoud Micro Monitor:

by noaudiophile

 

[617]

The issue we are twisting out knickers over is a direct result of trying to make the unit very small. The smallness may be a very desirable factor to some people, combined with the evidently considerable bass output the woofer provides.

I do not believe all the 70** products have this issue but people should be aware of it.

Using dsp to fix this issue would be difficult in a production setting I think. Small differences in cabinet volume and port length could shift this resonance around a little bit, which would be a big problem due to the narrow affected bandwidth. Then what happens if the user does stuff the port? You have a 10db peak.

As a humble diyer the speaker design starts with the woofer, this is the stuff you need to get right. I am open to the idea that JBL decided this wasn't an issue for the use case they saw for this device but it's not for me.

 

[Blumlein 88]

Maybe stuff the port with straws. Barely resistive damping, but might decrease the notch without changing much else.

 

[jhaider]

Respectfully, I'm kind of with echopraxia here. I agree the size and volume of the cabinet are real constraints. But that isn't the purchaser's fault if it doesn't work right.

While I agree with your last sentence as a general priniciple, here it is misapplied. This speaker does work right. It works exactly as the physics dictates it must. If that particular set of compromises does not fit a particular use case, and one is really offended by the graph, there is a larger model of similar design (708).

The big thing we all need is for another measurement from another source to see if this is a one off quality issue or if it is endemic to the unit.

I do not understand this. There is no dispute regarding the existence of this notch in this speaker It shows up in factory data - to Harman's credit. It shows up in very high resolution third party measurements.

The standard to which Harman designs, and to which they helped make a standard is 1/20th octave resolution. So a 1/10th octave anomaly if normal seems like it might be a possible audible problem. 1/10th octave is a 100 hz wide null at 750 hz roughly.

Are you sure your math is right? At least using the miniDSP 2x4HD's graphics as a guide, Q=15 (a little bit steeper than 1/10oct) is quite a bit wider than this notch. Q=30 looks like a better fit. I'm not in the mood to make screenshots right now but if you need to see them I can do that.

I've not heard the speaker, but wouldn't automatically dismiss this null if it is present will all the 705's.

To avoid doubt, my dismissal is based on my own listening evaluations.

There's also some anecdotal backing for my position. It seems JBL marketing folks did an event with a party trick where they sat 705 atop M2. Here's an ASR post with a link to Dr. Toole's narration of his experience at the event: https://www.audiosciencereview.com/forum/index.php?threads/jbl-705p-708p.1303/page-7#post-46886
Here's Dr. Toole's bottom line: when the sound was switched around the models the only obvious difference was "horsepower"

I also notice near universal acclaim for 305 and 308, and for almost all Revel's. Yet the 705 and 708 seems less universally praised.

I guess we've seen different things. I've read complaints about amplifier reliability (P-model), fit and finish issues, stupid mistakes such as shipping the wrong power cord for a given market (P-model), etc. Those are very good reasons to withhold praise.

The only substantive sound-related caveats I recall are Thomas's here, and his comparisons are to expensive Vivid's with wider dispersion and cabinets sculpted to attack diffraction, There was also a review from someone who preferred the midrange from a measurably more colored speaker (KEF LS50). Reading between the lines, my hunch is the reviewer saw the measurements and was "listening" to that knowledge.

Maybe stuff the port with straws. Barely resistive damping, but might decrease the notch without changing much else.

That won't work, because the straws won't conform to the flared S-shape of the port.

 

[jhaider]

First, I have no idea why you’re bringing up the speakers in my profile when I never invoked them for comparison here.

Again, here is the context. Reread to understand why I went there.
"First, you don't know that. You're making a wildass guess based on comparative measurements taken under much less controlled conditions, or with measurement conditions ill-specified, without understanding that's what you're doing."

My underlying assumptions were (a) you inferred the speaker you bought measured well based on the measurements the company's marketing team published, and (b) their marketing team's omissions regarding measurement techniques/processing such as smoothing, gating, or nearfield bass splicing technique, were not noticed.

But if you are interested in my towers, check out this link. The one you linked is of the standard version without the upgraded tweeter; the upgraded version measures significantly better.

Now there is something interesting! Unfortunately basic measurement conditions and processing are still omitted. But that's not the main point.

Recall, I characterized their spin on their old tower's off-axis frequency response as "fair, because it's marketing!" However, in your link they straddle the line between spin and bad faith. What's different? I was able to quickly identify the first tower's dispersion disruption in this figure:

Yet for some reason that figure is absent from the new tower's marketing measurements. The only figure purporting to show the upgraded tower's dispersion characteristics is an incomprehensible line jumble in the format used to support misleading statements regarding "tweeter bloom" for the old tower.

The crossover region dispersion disruption in the new tower could be less acute than in the old tower. However, that cannot be determined from the published data. My first rule of unspooling marketing spin is "what's not there is more important than what is," so I have my doubts.

None of that is to say it's a bad speaker, you shouldn't like it, etc. My only point their marketing strategy includes spinning figures that look like comprehensive measurements rather than publishing comprehensive measurements. While this is much better than most speaker companies offer, such figures are not equivalent to JBL 705's fully specified measurements.

Bottom line: you're raising hell about a "flaw" shown in honestly defined measurements without understanding what it actually is and how easily it could easily be hidden by minimal changes in the measurement conditions or processing. It helps to consider and understand such factors when interpreting loudspeaker measurements.

I actually think it is quite fair to say that other speakers do not have this problem: [705 and M105 graphs deleted]

A speaker with a much larger cabinet tuned higher and with considerably lower bass SPL limits* will have less apparent port resonance than a tiny low-tuned high output speaker? Now you're blowing my mind!

*Based on knowledge of both woofers.

Now, I understand that you’re trying to argue that my comparison doesn’t count unless the external dimension, bass extension, etc. matches exactly. But that’s not what I’m talking about here:

That is exactly what you are talking about, even though you don't realize it.

That you seem to be arguing that this anomaly in the 705’s spinorama is not only acceptable but common,

I really wrote the opposite. I've already stated that 705's tiny + high output + extended LF configuration - the sole cause of this inevitable-because-physics measured phenomenon you consider an "anomaly" - is uncommon. I previously remarked that I only know two 5" woofers with have the requisite linear xmax and motor strength to make such a speaker possible: JBL 705's proprietary drive unit and the $330+ ScanSpeak Illuminator.

Let me add a little bit here. Neither woofer could offer such bass extension in a passive alignment of this cabinet volume. In both cases, an assisted alignment - i.e. lower than standard port tuning with some electronic boost to bring up the hole between closed box rolloff and port output, is required.

Also uncommon is such broad and even dispersion in the top octave. Note the DI stays very flat from 10-20kHz. The compression driver's unique phase plug and waveguide are novel and impressive engineering.

As for "acceptable," I don't personally give two shits about this narrow little notch. People who can't bear to see any apparent flaw, even an extremely narrow bandwidth dip, in a loudspeaker's measurement, may consider it unacceptable. I think the intense focus on the scary looking irrelevancy is funny, because the graphs show a potential problem area in the speaker's performance that has not yet been mentioned!

If you’re claiming it’s impossible to build a speaker of a certain cabinet size without serious flaws like this, and if that were true, then that’s simply evidence that JBL made the mistake of choosing a cabinet size that can’t be implemented without flaws like this.

You're going to need to calm down if you want to continue a discussion. Your hysteric hyperbole serves no purpose.

First, it should be obvious that a public company generally makes something for one reason: they think they can sell it. Here, one can surmise JBL saw a niche in for a compact monitor with linear frequency response, high output, and good extension for "next generation multi-channel monitoring in post rooms, broadcast facilities and trucks." The resulting speaker happens to sound really great as a home hifi speaker is nice, but I don't expect home hifi specific concerns had a single line item on the design brief.

Second, again you pin a position on me that's opposite of what I previously articulated. I wrote that was possible for JBL to design the narrow little notch that so offends you out of the speaker: passive radiators instead a port. Read my Amphion Argon 3S review in SECRETS for a primer on ports vs. PRs. However, in addition to adding greatly to the cost and complexity of production, PRs are a bad choice for their intended market niche, which includes trucks. PRs can be damaged in transport, and may not react well to being constantly jostled between venues.

 

[echopraxia]

If you’re claiming it’s impossible to build a speaker of a certain cabinet size without serious flaws like this, and if that were true, then that’s simply evidence that JBL made the mistake of choosing a cabinet size that can’t be implemented without flaws like this.

You're going to need to calm down if you want to continue a discussion. Your hysteric hyperbole serves no purpose.

I invite any impartial party reading this thread to look at what he’s quoting me for above and calling “hysterical” and “hyperbolic”. I seriously doubt any normal person not emotionally evolved here would find my quote to be either hysterical or hyperbolic. And you don’t have to agree with my stance on this technical discussion, in order to see that.

That jhaider keeps attacking me like this is gaslighting and borderline harassment. None of my posts were directed at him personally in any way, and I’ve made that extraordinarily clear.

Independent of where these personal attacks are coming from, this needs to stop. Let’s keep the discussion focused on technical discussions about speakers — not attacks or commentary about the people making arguments you don’t like.

 

[amirm]

Let's dial down the emotions and personal remarks.

 

[echopraxia]

Ok perfect, now back to the discussion about speakers:

I understand everything that’s being said about why the flaw exists. But my point is that reasons do not excuse the existence of a problem; they only explain how it came to be.

So it might be more productive to first reach consensus on whether or not this is actually a flaw/problem. It is clear that we have not reached consensus here. It’s not just me who is concerned about this measurement (as you can see in this thread), but here is my rationale:

While I understand that all products make design tradeoffs, I claim that a 10db loss at any frequency is never an acceptable tradeoff for a $2000 “master reference monitor”!

First, imagine you’re consuming some electronic music where as a part of the song there is a pure sine wave of constant amplitude sweeping across 750hz, or sine wave notes of constant amplitude being played that land on 750hz one or more times. An out-of-place nulled response to this particular frequency will definitely disturb the accuracy and the enjoyability of the song.

But that’s not even the worst part. Imagine you’re using these as they’re advertised — as master reference monitors — as the ultimate reference for how your track really sounds, as you (the artist) mixes and validates that your track is as perfect as possible. This null would damage the mastering quality of the above song, if the artist trusted these monitors as their “master reference”.

I suppose you could argue “buyer beware” — that any recording artist should take it upon themself to verify that their speakers measured frequency response doesn’t have any major flaws. And maybe many people do. But I don’t think blaming the customer would be the right thing to do here: if JBL markets these as master reference monitors (as they do), I don’t think it’s right to tolerate this kind of flaw independent of where or why it ended up in the product.

And of course as always, to readers, feel free to disagree! There’s no reason we shouldn’t be able to disagree widely on this topic while also being civil and friendly to one another