JBL 708i Monitor Review (Passive: Part 1)

[MAB]

I’m not very experienced so I didn’t really get what’s “the solution” for these break-up modes harmonics.

Add a shock absorber. Literally. The passive notch has a large impedance at the resonance, and it electromechanically damps the breakup mode similar to a damper.
PruiFi has a white paper addressing the issue:

https://purifi-audio.com/wp-content/uploads/2022/03/220211_R05-Notchfilter.pdf

The LCR notch circuit looks like this:

JBL has done such in the past, like the XPL-series passive crossovers featured passive notch.
DSP notch just reduces the driving force at the cone's breakup mode, but do nothing to damp the harmonics.
The breakup mode is above the crossover frequency so not super-audible, but the lower harmonic distortion components are in the critical band.
I'm just not 100% convinced how audible they are. And, they do show up on this speaker since it is a very low distortion device I think...

 

[sarumbear]

Woofer independent yes, but topic was fundamental / linear distortion and HD.

-> explanation:

https://purifi-audio.com/wp-content/uploads/2022/03/220211_R05-Notchfilter.pdf

I do not feel educated enough to critique a JBL driver, I'm afraid.

 

[abdo123]

Add a shock absorber. Literally. The passive notch has a large impedance at the resonance, and it electromechanically damps the breakup mode similar to a damper.
PruiFi has a white paper addressing the issue:

https://purifi-audio.com/wp-content/uploads/2022/03/220211_R05-Notchfilter.pdf

The LCR notch circuit looks like this:

JBL has done such in the past, like the XPL-series passive crossovers featured passive notch.
DSP notch just reduces the driving force at the cone's breakup mode, but do nothing to damp the harmonics.
The breakup mode is above the crossover frequency so not super-audible, but the lower harmonic distortion components are in the critical band.
I'm just not 100% convinced how audible they are. And, they do show up on this speaker since it is a very low distortion device I think...

This is … magic!

Like I understood very little of what circuit is doing but daem it’s pretty awesome!

 

[MAB]

Passive crossover is not how the speaker is marketed. The speaker is designed to be used with a DSP acting as EQ and active crossover, feeding two amplifiers, one for each driver. Why can’t you then EQ the woofer separately and reduce any anomalies?

I understand JBL don't actually come out and state this up front, but 708i is an DSP-equalized and limited, passive-crossover speaker. It has a passive two-way crossover inside that can be used either with a single amplifier plus DSP or two amplifiers plus DSP. It can be bi-amplified, but the only difference is the DSP EQ is now split between LF and HF, the amps are full-range and crossovers are all passive. That's how Amir is able to operate the speaker with a single amplifier, and he mentions the passive crossover in the review. Careful reading of JBLs literature confirms what I am writing to you.

In summary of what the speaker actually is: It's designed to use one or two amplifiers per customer's choice. It has passive crossovers only. It uses DSP for shelving and notch filtering only.

So that you don't misinterpret my words, here is a picture of the DSP filters directly from London Architect EQ files for HF and LF:

There is a separate EQ file for single-wire, it is similar.

Why can’t you then EQ the woofer separately and reduce any anomalies?

They could have added a DSP or passive notch at to the woofer at 2.6kHz. It's clear from the DSP and passive crossover that JBL have chosen to do nothing. Also, DSP notch becomes slightly interesting for single amp use, which is common.
My guess is that this harmonic distortion peak at ~900Hz is just too low in magnitude for JBL to worry about. The HD is stunningly low and free of artifacts, which maybe makes the peak at 900 Hz stick out.

I do not feel educated enough to critique a JBL driver, I'm afraid.

This is not criticism. I love this speaker, and 708p too. These breakup modes seem to be a typical of underdamped wide band woofers used now, you end up with a breakup mode at the edge of the speaker's frequency response. You just need to deal with it. Here again is Amir's graph with the resonance frequency and the passive crossover clearly marked:

The rising impedance of the passive crossover already is going help damp these breakup modes. If you measured the bare driver with no crossover the resonance peak at 2.6kHz and the harmonic distortion peaks at 1/3 and 1/5 the resonance would be much more dramatic. You could use heavily damped paper cone woofers from 50 years ago, but you wouldn't get this great result with such a high crossover point.

But so many woofers these days are of the same design and sound great. PuriFi are one example. I built a pair of DSP speakers with Seas W18 magnesium woofers which have a massive breakup mode, and tried both passive and DSP notches:

Active DSP DIY speakers, for testing a few ideas

FFLOTSAM (Fabricated From LeftOver Things Studio Active Monitor) Active DSP speakers, for testing a few ideas I had some stuff lying around; a few random Baltic Birch-ply panels, some countertop material, a few cans of partially used brown spray paint, and some not-too clumpy spar varnish (that...

www.audiosciencereview.com

And Zaph has a nice set of measurements that show this cone breakup and harmonic distortion phenomena across multiple speakers:

Zaph|Audio

JBL's solution to put the crossover far enough below the resonance is fine I think. I am eager to see part 2 of the review.

 

[sarumbear]

It has passive crossovers only. It uses DSP for shelving and notch filtering only.

How do you know that?

On my very early post, JBL clearly showed that the speakers were bi-amped. Why would JBL bi-amp a speaker if the DSP is only going to be used for EQ?

Not to mention it clearly says on the speaker that you can bi-amp them.

Or why would they have a shortening connector on the bi-amp inputs in order to use the single input?

When operating the speaker using the Single Wire connection, the supplied plug must remain inserted in the Bi-Amplified connector. Removing this plug will prevent single wire operation.

https://jblpro.com/en-US/site_elements/7-series-set-up-guide

 

[MAB]

This is … magic!

Like I understood very little of what circuit is doing but daem it’s pretty awesome!

I counting on it being electronics!
I certainly didn't invent anything new.
I think I first heard about this from Jon Marsh, he was addressing the wave of magnesium, aluminum, ceramic, etc. drivers that had become available.
Linkwitz discussed active notches. He used some really hard to tame drivers in his famous designs.
I don't think these resonances are unusual or hard to deal with either.

 

[MAB]

How do you know that?

On my very early post, JBL clearly showed that the speakers were bi-amped. Why would JBL bi-amp a speaker if the DSP is only going to be used for EQ?

Not to mention it clearly says on the speaker that you can bi-amp them.

View attachment 254327

Or why would they have a shortening connector on the bi-amp inputs in order to use the single input?

https://jblpro.com/en-US/site_elements/7-series-set-up-guide

Honestly, I wondered the same thing.
I loaded the Crown DCI filter files:

https://jblpro.com/en-US/site_elements/708i-crown-dci-settings

And looked at the raw files to make sure I didn't miss anything. And cross-checked with friends with the London Architect install to make sure.
I even included a screenshot of the actual filters for the Bi-Wire config to pre-answer your question.
That EQ would be hard on the ears and hard on the tweeters!

 

[sarumbear]

Honestly, I wondered the same thing.
I loaded the Crown DCI filter files:

https://jblpro.com/en-US/site_elements/708i-crown-dci-settings

And looked at the raw files to make sure I didn't miss anything. And cross-checked with friends with the London Architect install to make sure.
I even included a screenshot of the actual filters for the Bi-Wire config to pre-answer your question.
That EQ would be hard on the ears and hard on the tweeters!

Could it be that the LF & HF filtering is done on the BSS unit shown on their diagram? It is madness to use two amps and not utilise an active crossover.

 

[MAB]

Could it be that the LF & HF filtering is done on the BSS unit shown on their diagram? It is madness to use two amps and not utilise an active crossover.

It's a valid question, I wonder too. And also wonder exactly what is inside the 708p.
I will ask. And double-check to make sure I have not missed anything.
I say this because the same questions came up a while back, with different Architect versions loading different filter sets.

 

[Bartl007]

I know this thread is about the JBL 708i, but someone mentioned the JBL SCL-4 and I have a question. Will the SCL-4 be audibly better with anechoic data than without it? Will it be audible at all? Isn't it rather pointless to bother about it?

The scl-4 shares the same compression driver used in the 708i so this is as on topic as it gets in this thread

I have not heard the scl-4 in either form, but just looking at the spins, i would say the differences would be audible. Otherwise, why would JBL even bother making the EQ files available?

 

[Tangband]

I understand JBL don't actually come out and state this up front, but 708i is an DSP-equalized and limited, passive-crossover speaker. It has a passive two-way crossover inside that can be used either with a single amplifier plus DSP or two amplifiers plus DSP. It can be bi-amplified, but the only difference is the DSP EQ is now split between LF and HF, the amps are full-range and crossovers are all passive. That's how Amir is able to operate the speaker with a single amplifier, and he mentions the passive crossover in the review. Careful reading of JBLs literature confirms what I am writing to you.

In summary of what the speaker actually is: It's designed to use one or two amplifiers per customer's choice. It has passive crossovers only. It uses DSP for shelving and notch filtering only.

So that you don't misinterpret my words, here is a picture of the DSP filters directly from London Architect EQ files for HF and LF:
View attachment 254307
There is a separate EQ file for single-wire, it is similar.


They could have added a DSP or passive notch at to the woofer at 2.6kHz. It's clear from the DSP and passive crossover that JBL have chosen to do nothing. Also, DSP notch becomes slightly interesting for single amp use, which is common.
My guess is that this harmonic distortion peak at ~900Hz is just too low in magnitude for JBL to worry about. The HD is stunningly low and free of artifacts, which maybe makes the peak at 900 Hz stick out.

This is not criticism. I love this speaker, and 708p too. These breakup modes seem to be a typical of underdamped wide band woofers used now, you end up with a breakup mode at the edge of the speaker's frequency response. You just need to deal with it. Here again is Amir's graph with the resonance frequency and the passive crossover clearly marked:
View attachment 254321
The rising impedance of the passive crossover already is going help damp these breakup modes. If you measured the bare driver with no crossover the resonance peak at 2.6kHz and the harmonic distortion peaks at 1/3 and 1/5 the resonance would be much more dramatic. You could use heavily damped paper cone woofers from 50 years ago, but you wouldn't get this great result with such a high crossover point.

But so many woofers these days are of the same design and sound great. PuriFi are one example. I built a pair of DSP speakers with Seas W18 magnesium woofers which have a massive breakup mode, and tried both passive and DSP notches:

Active DSP DIY speakers, for testing a few ideas

FFLOTSAM (Fabricated From LeftOver Things Studio Active Monitor) Active DSP speakers, for testing a few ideas I had some stuff lying around; a few random Baltic Birch-ply panels, some countertop material, a few cans of partially used brown spray paint, and some not-too clumpy spar varnish (that...

www.audiosciencereview.com

And Zaph has a nice set of measurements that show this cone breakup and harmonic distortion phenomena across multiple speakers:

Zaph|Audio

JBL's solution to put the crossover far enough below the resonance is fine I think. I am eager to see part 2 of the review.

If this is true , that 708i is a passive crossover loudspeaker with dsp corrections then this is a slightly flawed design .

They should have used a passive notch filter , but they didnt .

They could have used a dsp crossover with more than 36dB / oct slopes to minimize the breakupmodes , but they didnt do that either .

Im not impressed by the passive crossover in my jbl 530 and Im not impressed here either . Jbl makes some great drivers though , but its sad that they dont use as much knowhow with their crossovers in their speakers .

Its clear that this seems to be a passive loudspeaker with a nonperfect executed passive crossover that could have been done in a better way , with a notch filter implemented.

If combining passive crossovers and active dsp techniques, both must be done in a correct way , because if its not, you will easily end upp with a less sum of its two parts, ie : -1 -1 = -3.

One can correct the frequency response afterwards with a dsp and make it look good on sine-sweeps , but the sound will be worse than doing it right in the first place .

 

[Adi777]

The scl-4 shares the same compression driver used in the 708i so this is as on topic as it gets in this thread

I have not heard the scl-4 in either form, but just looking at the spins, i would say the differences would be audible. Otherwise, why would JBL even bother making the EQ files available?

Okay, so will be expensive

 

[changer]

The breakup mode does not show up in 708p spinorama, so it is attenuated good enough.

 

[Tangband]

The breakup mode does not show up in 708p spinorama, so it is attenuated good enough.

The breakup mode should have been corrected with a notch filter or a steeper crossover in the passive crossover , not corrected with a dsp afterwards.

Another alternative would be to ditch the passive crossover entirely and use a dsp crossover with 48 dB/oct slopes , but this was not done.

 

[changer]

The breakup mode should have been corrected with a notch filter or a steeper crossover in the passive crossover, not corrected with a dsp afterwards. It is reasonably controlled physically and attenuated by the crossover (same slopes as 708p).

Another alternative would be to ditch the passive crossover entirely and use a dsp crossover with 48 dB/oct slopes , but this was not done.

There is no correction in the DSP settings that is specifically directed at the break-up mode at all.

LR8 filter might not be a good idea, to steep of a filter inserts issues with phase.

 

[Tangband]

There is no correction in the DSP settings that is specifically directed at the break-up mode at all.

LR8 filter might not be a good idea, to steep of a filter inserts issues with phase.

The correction of the break up mode could have been done in the passive crossover - this passive crossover is there and they should have gone further and done a notch filter . If you do it all digitaly , theres no problem at all with a good dsp with a steeper crossover filter . In my opinion, you cant hear the phase issues caused by a LR8 DSP crossover.
The breakup of the woofer is shown with the blue arrow. Its about 10 dB below the tweeter response .
Maybe Im nitpicking but this seems othervise to be a very good loudspeaker.

 

[Dj7675]

One thing to note about the use of EQ regarding the 708i/705i/M2 if you choose to not use supported Crown Amps and BSS Processors… it may not always be as straight forward as it seems I have learned. It is my understanding that both the BSS Processors and Crown amps use constant q bell filters. Most other devices do not and use constant q bell filters. StormAudio uses constant q bell filters, but mini dsp, roon, and the ones posted by @pierre for example do not (@pierre can correct me if I’m wrong) In the attached document, Stormaudio talks about the differences of constant and not constant bell curves. The resulting EQ can be different if you are applying EQ for example that has been pulled from a BSS processor that is constant Q but using a device to EQ the speaker that does not use constant q bell curves.
Depending on the source of your EQ and the device you are using to EQ the speaker, may have to do some conversion. I found an excel file that someone created that converts them which I found helpful that I believe is converting them correctly. LINK
I never knew that some systems were constant q and some weren’t and thought it was interesting.

 

[Robbo99999]

One thing to note about the use of EQ regarding the 708i/705i/M2 if you choose to not use supported Crown Amps and BSS Processors… it may not always be as straight forward as it seems I have learned. It is my understanding that both the BSS Processors and Crown amps use constant q bell filters. Most other devices do not and use constant q bell filters. StormAudio uses constant q bell filters, but mini dsp, roon, and the ones posted by @pierre for example do not (@pierre can correct me if I’m wrong) In the attached document, Stormaudio talks about the differences of constant and not constant bell curves. The resulting EQ can be different if you are applying EQ for example that has been pulled from a BSS processor that is constant Q but using a device to EQ the speaker that does not use constant q bell curves.
Depending on the source of your EQ and the device you are using to EQ the speaker, may have to do some conversion. I found an excel file that someone created that converts them which I found helpful that I believe is converting them correctly. LINK
I never knew that some systems were constant q and some weren’t and thought it was interesting.

Interesting pdf, I didn't know about differences between constant-Q and non constant-Q. I use parametric filters a lot for EQ'ing headphones & speakers using REW software to model the effects of the parametric filters on the measured frequency responses of the headphones & speakers, and that uses constant-Q from my understanding of your pdf. In fact, every piece of EQ software I've ever used stipulates Q in this same way - as in the filters are universal & interchangeable (give the same results) between all the different EQ software I've used. I'll list the different EQ software and "systems" I've used: REW / EqualiserAPO (& Peace) / Neutron Player / AutoEQ (github) / miniDSP - it's also the same definition of filters used by everyone in the headphone EQ arena - Oratory1990 / Crinacle / Resolve / Innerfidelity / jaakkopasanen(@AutoEQ). I've never come across any implementation that uses non-constant Q. (I think you made a mistake or typo in your first or second sentence, because you're referring to constant Q in both so there's a logical mismatch happening there regardless.) And I think you must be wrong in saying miniDSP / Roon / Pierre don't use constant Q filters.

EDIT: unless there's a 3rd type of Q that isn't described as constant-Q or non-constant-Q. Either way all the parametric filters I've ever used behave like the following pic in your pdf you linked (note within each graph below the Q value is the same between each of the different graphed lines (within a graph), and the only difference is the amount of Gain, which is where the misunderstanding could be happening with those who have glossed over the pdf):

and they've never behaved like the following which is non-constant Q (note that the pdf is saying the Q value of the following two graphed filters are the same, only the Gain is different):

EDIT #2: I noticed @pierre put a "Like" on your post, I think he didn't read the pdf in-depth as he's an experienced guy using EQ, so I think he just made an assumption that "constant-Q" meant a Q of 1.41 rather than a person being able to stipulate their Q-value, but that's not what the pdf is referring to in terms of what is meant by "constant-Q".

EDIT#3: There might be a third type called "proportional Q" (after I googled the topic), which might be what most EQ software uses, but that is certainly not the same as the "Figure 2" graph above, and instead looks more like the "Figure 3" graph above (when I model it in REW) which is what Storm Audio is saying is Constant-Q. I'm not sure that Storm Audio have covered off all the different types of Q. @oratory1990 , could you clear this up for us, any misunderstandings that are happening in my post & the one I replied to? (Meanwhile & also @GaryH, I know you like understanding & using EQ, have you come across any of these definitions? I was surprised I didn't know about them.)

 

[3ll3d00d]

And I think you must be wrong in saying miniDSP / Roon / Pierre don't use constant Q filters.

Storm are talking about the issue described in this question, ie counteracting the warping that occurs as the filter centre frequency approaches Nyquist. A minidsp that operates at 48kHz will have this problem, jriver also does, I don't know about roon. I thought the variable q filters were all old school analogue filters tbh.

Practically speaking it's still a constant proportional (edit: mixed up constant and proportional) q filter, just one with an implementation flaw. Solutions include not attempting to do anything more than broad (low q) shaping at high frequencies and running your DSP at 96kHz

Audio EQ Cookbook without frequency warping

The famous https://www.w3.org/TR/audio-eq-cookbook/ offers a set of [real] biquad filter calculation formulas that generally work fine. However, when filter's frequency approaches Nyquist frequency...

dsp.stackexchange.com

Fwiw another type is a raised cosine filter as implemented on lake processors

Lake Processing – Raised Cosine Equalization im Detail

www.felusch.de

I don't recall seeing these on anything else ,

 

[Robbo99999]

Storm are talking about the issue described in this question, ie counteracting the warping that occurs as the filter centre frequency approaches Nyquist. A minidsp that operates at 48kHz will have this problem, jriver also does, I don't know about roon. I thought the variable q filters were all old school analogue filters tbh.

Practically speaking it's still a constant q filter, just one with an implementation flaw. Solutions include not attempting to do anything more than broad (low q) shaping at high frequencies and running your DSP at 96kHz

Audio EQ Cookbook without frequency warping

The famous https://www.w3.org/TR/audio-eq-cookbook/ offers a set of [real] biquad filter calculation formulas that generally work fine. However, when filter's frequency approaches Nyquist frequency...

dsp.stackexchange.com

Thanks, but they're talking more than just about that issue. They're talking about the following happening when using "non-constant Q", note the Q value is the same (beware the potential confusing juxtoposition of terminology) in the following graph, just the Gain (dB) of the filters is different:

I'm basically saying that I've never seen a piece of software use this particular definition of Q.