Purpose of flat response below 20Hz - Why does it matter? - Page 10

Originally Posted by floridapoolboy

I read something once about "the brown note", is that what you are referring to?

Originally Posted by Chu Gai

Just wondering here, Bosso. Medical science knows that various organs in the body have certain resonant frequencies (low) and sundry studies have indicated correlations with frequencies and perceived sensations (paranormal comes to mind but there are others). In your opinion, do you think the reason for the presence of low level information in movies is an attempt to tap into these perceptions?

Infrasound has been known to cause feelings of awe or fear in humans. Since it is not consciously perceived, it can make people feel vaguely that supernatural events are taking place. Truth be told, some Movie soundtracks make use of infrasound to produce unease or disorientation in the audience

Originally Posted by bossobass

A higher sensitivity driver equals a lower sensitivity <20 Hz.

Originally Posted by bossobass

When I say people are preferring that choice currently, I don't mean to say that they're doing so from an understanding of transient response vs dynamic tracking of transients.

Originally Posted by bossobass

I've said many times and posted the evidence to back it that a flat, full BW FR equals a) better accuracy (reproducing what's on the disc vs not doing so really comes under 'duh'), b) a better listening experience.

Originally Posted by bossobass

Yes, we are in agreement on many aspects of the low end of things. The difference is that I have 2-1/2 octaves more BW and lots more headroom, which equates to a more accurate system.

Originally Posted by bossobass

The subjective difference is irrelevant.

Originally Posted by bossobass

Debating the validity of non-filtered recorded source is irrelevant.

Originally Posted by bossobass

It's not that I haven't presented a shred of evidence. It's that you've decided to shred the evidence I've presented.

Originally Posted by bossobass

In the real world, all transients have content to DC. In the real world of recorded source, there is intended content to <3 Hz. If your playback system rolls off at 20 Hz, an audible form of distortion exists.

Originally Posted by DS-21

Given that modern best practices (i.e. the methods and procedures that provide the overall smoothest response in small rooms)...

Originally Posted by craig john

"Best Practices"??? What are "best practices" as they refer to subwoofer system integration.

Originally Posted by craig john

I see you use this "catch-phrase" all over the forum, and I have no idea what it means.

Originally Posted by craig john

[pointless ad hominem attack deleted]***the only "best practices" are the Geddes procedures, with random, multi-subwoofer placement with no Bass Management.

Originally Posted by craig john

In those "best practices", FR management is the only goal. System headroom, max SPL and ULF extension are not even considerations.

Originally Posted by craig john

There can be no rigidly defined "best practice" for subwoofer(s) system integration.

Originally Posted by craig john

More importantly, subwoofer optimization is dependent on the expectations and goals of the optimizer.

Originally Posted by craig john

Some might have the ability to place bass traps optimally around the room while others not so much.

Originally Posted by craig john

All these different design goals require different strategies for subwoofer optimization. There is no single "best practice", not even Geddes procedure, that encompasses all the different rooms, systems and goals

Originally Posted by DS-21

Around 0.5mH if memory serves. (The new Parallels doesn't seem to work right now, so I can't check it against my Unibox database.)

But basically, lower than anything else with the same volume displacement. An underhung (i.e. short) single-layer coil and copper sleeve on the motor will do that.

Originally Posted by DS-21

Believe it or not, Penn, you're not the only person to have owned several different models of subwoofers.

But as to your second comment, it depends on the bandwidth. If you're talking below 30Hz, sure, they're all airpumps. They'll all work fine. All that matters there is how much air each can pump. Well, that and power compression, which suggests get the largest-diameter voicecoil with the best possible cooling system.

But play 'em much higher and the story changes. You can't just EQ out the effects of that sky-high inductance, sorry.

Something with large volume displacement, linear motor and suspension, and low inductance.

Of currently-available drivers, I submit the Aura NS18 and B&C 21SW150 (I think that's the model) are tops. TC's LMS and Aura-knockoff lines are up there well, if one wants to roll the dice on TC's build quality.

Fake crashes, explosions, and the like are rather different from, say, reproducing a tympani roll recorded live. There's no real event to reference, so it's not about accuracy but believability. That is to say, does it sound like what you think a crash or explosion in that place might plausibly sound like?

Nonsense. It means wide bandwidth, low distortion, and high output. Not only one or two of those things.

And severe bandwidth-limiting compared to a good driver.

Originally Posted by DS-21


F3 is irrelevant. It could be 70 Hz and still be the foundation of a great sub! What matters is volume displacement. With enough volume displacement, you can EQ the frequency response to taste.

Originally Posted by DS-21

By contrast, the Fi Q-series just another excessively long coil driver with no shorting rings in the motor. Note that Fi doesn't even specify the inductance of the Q18. What you'll see from one is roughly what Josh got from the eD sub: plenty of output, at the expense of fidelity due to the high inductance.

Originally Posted by bossobass

Yeah, everyone regularly uses an 83dB 1W/1M driver.

Originally Posted by bossobass

Show the details so we can see what you're trying to say here. Why do you insist on talking to me like it's my first day?

Originally Posted by bossobass

Lower Fs = lower sensitivity = lower F3 in a sealed box. Higher sensitivity = higher FS = higher F3 [in a sealed box].

Originally Posted by bossobass

Lower F3 = higher sensitivity below 30 Hz. Higher F3 = lower sensitivity below 30 Hz. Sealed box 101.

Originally Posted by bossobass

The CSS SDX-15 has inductance of 1.8mh.

Originally Posted by bossobass

It's also interesting that you believe "you can't just EQ out the effects of that sky-high inductance, sorry", but at the same time, "F3 is irrelevant. You can EQ the FR to taste".

Originally Posted by bossobass

F3 is irrelevant? EQ the FR to include subwoofer BW with the B&C sub and you lose the top end sensitivity, whether you use a shelf or L/T.

Originally Posted by bossobass

Challenge: You build a system with your Aura 18 and I'll build a system with 2x18 Fi Q (price difference gives you 2 for 1 Q vs Aura).

Originally Posted by bossobass

"You're wrong", is no argument. In fact, once you've said that, what's next, "you're really, really wrong"? Prove it's artifact. Prove it has no subjective significance.

Originally Posted by DS-21

There are certainly 83dB/W/m drivers used as subwoofers. The Eclipse LMT8200 (a 12 with 30+ mm of xmax with TC's LMT coil) is an example of one.

Originally Posted by DS-21

Because you're writing as if it's your first day. I know you know better, but you seem to just want to pick a fight. Which is fine, so long as readers aren't misled by incorrect information.

Originally Posted by DS-21

Incorrect, unless you're talking about relative sensitivity.

Originally Posted by DS-21

In terms of absolute sensitivity, box size is what matters. A driver that's 99dB/W/m at 1kHz will have the same sensitivity at 20Hz as a driver that's 83dB/W/m at 1kHz. The former will just have rolled off more from its reference sensitivity than the latter. Hoffmann's Iron Law 101.

Originally Posted by DS-21

If F3 was truly that important, you wouldn't be using "undersized" boxes with a Linkwitz Transform. You'd be using Qtc=0.707 boxes, as they have the lowest F3. And you'd be massing the hell out of your subwoofer cones, to lower efficiency and thus F3. But you don't, do you? Yet, because you seem to be in attack dog mode, that is the very approach you're currently attacking.

Originally Posted by DS-21

Think for a moment about what those data show, and how it doesn't relate to my point in the slightest. Is box size held constant? If so, is volume displacement the same?

Given that I'm arguing the correct position, that box size determines low end efficiency, data where box size is not held constant just shows you didn't understand the argument to which you were replying.

Originally Posted by DS-21

And a graph of maximum output has absolutely nothing to do with sensitivity. THAT's just Reading Graph Labels 101.

Originally Posted by DS-21

Look at both drive-units, in the same sized box, with a given constant level of drive. 2.83V is the standard, but I don't see why it matters what the actual level is so long as it's not so high as to fry the coils, or so low as to be lost in the measurement's noise floor. That will tell you the sensitivity, and that they are in fact the same sensitivity down low despite the B&C driver's considerably higher broadband sensitivity.

Originally Posted by DS-21

Maximum output, obviously, is a function of volume displacement and not of sensitivity unless the system is amp-limited. Obviously, ultimately volume displacement of a subwoofer can be amp limited. But that's not particularly relevant here, because to reduce amp limiting one needs more power and/or a bigger box with either driver.

Originally Posted by DS-21

What's your point? Are you saying that fits my rubic of "sky-high?" If so, you're just willfully misinterpreting what has been so clearly spelled out. Inductance on its own is meaningless. Inductance must be considered along with resistance. What's the Re of the SDX-15? According to their spec sheet, 3.6Ω. So the normalized Le is in fact about 0.5mH/Ω. That's reasonable, considering the coil is relatively short for the throw (XBL^2) and it uses shorting rings. And 0.5mH/Ω is pretty darned low, no?

Originally Posted by DS-21

There is no inconsistency there to a thoughtful person.

Originally Posted by DS-21

And, it's worth noting, I'm making the exact same argument you've made for years: EQ the F3 to your desired low-end cutoff (you specify a Linkwitz Transform, whereas I would use a combination of parametric EQ and shelf filters, guided in my approach by in-room measurements of those subwoofers in that room, but that's a detail difference) and use enough subs/amp to get the headroom you want.

Originally Posted by DS-21

You seem to have such distaste for my words that they're causing you to reject your own long-held positions and argue against them. How many times have you written that it's good practice to take advantage of the high-Q efficiency bump and then EQ the low end? Have you fundamentally changed your mind about that? If so, why?

Originally Posted by DS-21

In practice, what does that mean though? Neither the driver or the amp is working that hard. Sounds fine to me.

Originally Posted by DS-21

I have better things to do with my time than to "prove" something obvious to anyone capable of applying human thought.

Originally Posted by DS-21

Especially given that you seem to be manufacturing disagreement even where there is in fact none. What I am posting is entirely consistent with your past posts and your past and present practice, but you're just trying to make noise for some reason.

Originally Posted by DS-21

As you know, I can't currently reproduce that content. You can, yet you have no interest in asserting even its relevance, only its presence. That speaks plenty to people intellectually equipped to draw reasonable inferences.

Originally Posted by DS-21

The methods that have been shown to provide:
(a) the smoothest upper bass frequency response in a small room, and
(b) the least seat-to-seat variance in the listening area, as the listening area is defined by the user (who can, of course, choose the area over which s/he will take spatially-averaged measurements, never single-point always spatially-averaged if one cares about getting good data), with the caveat that said listening area must be reasonable. For example, trying to optimize for seating in the middle of a room and seating on the back wall is likely to result for bad overall sound, because the defined listening area is simply unreasonable.

Today, that means one of three options:
(a) Geddes if one's doing things manually in most domestic living rooms; or
(b) double-bass-array if one's room fits in a very narrow and specific geometry, one can place subwoofers appropriately, and one is willing to add some extra volume displacement to make up for the lost boundary-loading advantages; or
(c) Sound Field Management (Toole/Welti/etc.) if one wants a good automated approach and is willing to pay for the appropriate gear (JBL BassQ or Synthesis)

Tomorrow, someone may introduce a method that's easier to do and/or higher-performing and/or cheaper to implement. But we are where we are today.



I am not to be held responsible for your puzzling inability to look up "best practices" on google and/or wikipedia and learn what the term means if you are unable to derive the meaning from context.



Actually, see supra.

The fact that those three methods (Geddes, DBA, SFM) use different techniques but seem to end up with the same measurably and audibly superior result speaks well for all three methods. The fact that other methods do not compare speaks comparatively poorly about them.

(As for "bass management," It's not at all incompatible with the general setup procedure outlined by Geddes. Regardless of setup procedure, "bass management" is just a crutch for people who can't/won't use mains with closed boxes to limit excursion and sufficient volume displacement to act as viable sources in the modal region. Most systems composed of the small, inefficient mains typically marketed to home audio will require that crutch, my current one as discussed in the "modest multisub" thread included. The best systems will not. Interestingly and as an aside, with multisubs properly set up it seems that the frequency threshold for localization is higher compared to lesser setups. Whereas in a setup a single corner sub crossed over at 90Hz may at times make its presence known, with a multisub setup using that same corner sub does not have localization issues even with a 2d order electrical - and acoustical, more-or-less - crossover between mains and sub at 120Hz. Indeed, with these speakers and subs in this room, at 200Hz the localization issues with multisubs were about as bad as they were with the single corner sub crossed at 90Hz. That, I think, has interesting implications for mains design/specification.)



That is correct. System headroom, SPL, and ULF extension should not be considerations in the set up phase, which is what the Geddes method, DBA, SFM are about. If one must futz with setup to obtain the desired SPL or extension, then the system was badly designed/specified from the start.

Headroom/SPL and LF cutoff are, rather, design phase considerations. One can only reach the desired SPL and LF cutoff one seeks by selecting subwoofers of appropriate volume displacement and bandwidth* and obtaining the required amplifier power and signal processing to meet those goals.

*Low inductance is required to meet the upper-bass bandwidth. If one is using subs that unload below tuning (vented/PR, >4th order bandpass), the subs' native tuning also matters, because the subs won't pressurize the room below tuning. With sealed and 4th-order bandpass subs, native LF cutoff is irrelevant because any well-designed system will include processing to tailor the low end response to taste, be it the one-trick pony "Bassis" that Bosso likes or a more flexible box with parametric EQ and/or shelf filters.



That depends, of course, on one's definition of "rigidly defined."

If you think that term fits the following:
specifying subwoofers adequate volume displacement, power, and signal processing to meet one's goals (the system design phase), and then fitting them in the room where you realistically can, and setting them to optimize upper-bass frequency response (the system setup set-up phase),
then I simply disagree with your definition of "rigidly defined."

It seems, to my mind, rather a flexible process.

And, as you know, I've done it several times, with a wide variety of gear but one constant: superior results. And that's not because I have any great competency or much better equipment.

Most goals are met (or fail to be met) in the design phase, assuming some intellectual rigor was used in setting realistic goals (per one's budget, aesthetic tolerances, etc.). One can roughly estimate the volume displacement required for a given output and LF cutoff in a room of a given volume. (With in-room measurements of previous systems, knowing their theoretical abilities, one can have less of a fudge factor in those estimations.)

The iterative set-up process then optimizes the system along the only major variable that cannot be accounted for in the design phase: the response of that system in the modal region of that room.



Unless you're arguing that the optimizer may have low expectations as to sub-mains integration and flatness of response in the modal region, how is that even the very slightest bit inconsistent with anything I've written?

Again, LF cutoff and SPL are design phase considerations, not setup phase considerations. If one must kludge the setup to make up for lack of rigor or faulty assumptions in the design phase, that's not the fault of the setup method but rather of the system designer.



That raises the question of what value "bass traps" have. I don't think they have much. As you've seen, I routinely achieve excellent performance without even considering such eyesores.



Based on what you've written, you seem to hold that position primarily because you don't understand my arguments. Perhaps that is my fault, because I have written with insufficient clarity. Hopefully now you understand them better.

Originally Posted by mojomike

Your method for bringing about a smooth frequency response without using eq or room treatments works well at doing just that, but it introduces artifacts that are worse and are not really repairable.

Originally Posted by mojomike

Having asymmetrically placed subs around the room will cause a smearing in the time domain.

Originally Posted by Dr. Floyd Toole

However, because we know that low-frequency room resonances generally behave in a minimum-phase manner, we know that if there are no prominent peaks protruding above the average spectrum level, there will not be prominent ringing in the time domain. It is this indirect, inferential knowledge that permits us to confidently use frequency responses as a primary source of information about room behavior at low frequencies.

Originally Posted by mojomike

What makes it much worse in your case is that you are smearing things well up into the upper bass (200hz?). That includes the fundamentals of voices and almost everything.

Originally Posted by mojomike

Your technique is a perfect example of how you can have a flat frequency response, and still have unrealistic sound because of time related smearing and it's well up into the range that many of us consider lower midrange.

Originally Posted by mojomike

You are focusing on a flat response but ignoring other factors that contribute to realism and are likely to be even more significant.

Originally Posted by MKtheater

There are many ways to get a flat response. Here is a response that uses no EQ, subs all in the front behind the screen, crossed over at 80hz into 2 speakers(mains).

Originally Posted by bossobass

Great. Now, post the 99dB 12" driver specs and we'll do the comparison.

Originally Posted by bossobass

There is only relative sensitivity.

Originally Posted by bossobass

Box size doesn't change a drivers parameters.

Originally Posted by bossobass

Wow. Box size won't change the point being made. Your .5mh Le 99.7dB driver has less sensitivity below 30 Hz than a pair of the 1.8mh [per driver] SDX-15 [3.6mh in series, so what does "sky-high" mean?

Originally Posted by bossobass

Box size is determined by Cms/Vas. It's not arbitrary.

Originally Posted by bossobass

Holding box size constant for different drivers is a ridiculous with no point.

Originally Posted by bossobass

Of course it does. Read the graphs again, please. There's a nearly 1 to 1 correlation.

Originally Posted by bossobass

You said F3 is irrelevant because you can EQ to taste. I said most every sub available is amp limited. The 2 are directly related.

Originally Posted by bossobass

It's inductance and 1W/1M specs that are irrelevant here, which is the point I've been trying to make.

Originally Posted by bossobass

I've never said or agreed with no shorting rings = low fidelity.

Originally Posted by bossobass

I've never said ULF content in recordings is unintended artifact.

Originally Posted by bossobass

I've never said ULF is subjectively irrelevant.

Originally Posted by bossobass

I've never disparaged Fi products as inferior and not worthy of attempting to use in a subwoofer build.

Originally Posted by bossobass

I've never favorably compared the Dayton Ref driver to a Fi Q as a superior product, nor would I with a straight face.

Originally Posted by bossobass

I've never said the Fi Q has unacceptably high Le, then in the same sentence admit I don't even know what its Le spec is.

Originally Posted by bossobass

To the first and main question, you've gone on and on about the lack of relevance, that it's artifact from sloppy production practices, that it only appears in source you consider garbage, etc. You've tried to defend this position by burdening others to prove it to you personally while posting nothing whatever to back your assertions.

Originally Posted by bossobass

To the second point, although Dolby, THX and everyone else in the MC audio world has standardized the LR4 LPF at 80 Hz, you assert that a subwoofer must be crossed over at 150-200 Hz to exhibit fidelity. Again, zero data to back your position,

Originally Posted by bossobass

It's relevance is that a) it's on the disc,

Originally Posted by bossobass

b) my system accurately reproduces it and

Originally Posted by bossobass

Ic) if your system does not, there is an audible distortion in the presentation.

Originally Posted by DS-21

No, it doesn't.



In theory and practice, that is simply incorrect.

As for the theory, let's dispense with that first with Dr. Toole's words from Sound Reproduction. at 245.


As for practice, frankly, your argument is one only available to a person who simply has no relevant practice. That is to say, s/he has simply not experienced a properly set-up multisub system. Subjectively, I've found that a well set-up multisub system will sound firmly "locked" in time, far more so even than a stereo pair of full-range speakers, let alone a system using a single subwoofer crossed over in the modal region. Transients have more impact, the sonic image is more palpable, and the overall sound of the upper bass is just...natural. In a way that no stereo pair of monopole speakers (or speakers sitting atop subs, or single subs with a pair of monopoles, etc.) can be. (Dipoles can get close - the best 2-channel bass I've heard, including the Orions at AXPONA this year and a friend in Vienna's Audio Artistry Dvoraks, were from a pair of Gradient Revolutions) but have placement restrictions that make them untenable, at least for me. I suspect the same applies to cardoids.)

My experience is consistent with that of others who have tried it. (As well as high school buddies now playing in major symphony orchestras who have heard my system. One asked me to design a similar system for her - yes, her, the ULF stuff doesn't seem to appeal to anyone but a small subset of men, but more natural sound in the upper bass interestingly enough seems to appeal to everyone who loves music - after she heard it.) Look on the forums for subjective reports from people who have heard Dr. Geddes' home system. All comment on how natural the bass sounds. In fact, the only negative report of a multisub system I've heard has some from Penn, who simply did not (and still seems not to) understand the concept well enough to apply it with the slightest modicum of competence.

Not only that, but also experience listening to such a system will sensitize one to the wretched state of upper bass reproduction in most audio systems. IOW, the nasty bass response measurements one sees even from stellar "full range" stereo mains well set up by knowledgeable people are audible in practice, even if people without experience in high fidelity bass reproduction do not seem to notice.

And the fact of the matter is, in-room upper bass response for stereo pairs of loudspeakers in domestic living rooms is just tragically bad. (Measurements for systems using a single subwoofer are generally worse, because there is one one pressure source driving the room, as opposed to two.)

Here is the best in room measurement of a stereo pair of "full range" monopole-in-the-upper-bass loudspeakers I've ever seen.



The loudspeakers are Revel's Studio2, and they were measured for Stereophile by JA in Slate foreign affairs columnist Fred Kaplan's room. There is also an in-room measurement for a lesser speaker - Verity's nasty sounding $40k Parsifal - in the same room on that graph; The Revel is in red.

Note that from 30-120Hz there is about a 12-13dB swing. (The lesser speakers have about an 18dB swing over the same bandwidth; even that is better than most in-room measurements) Compare that with my various multisub setups' in-room measurements. Most people, because of both the relative inferiority of their speakers* and setup, will have much worse results than the Revels achieve in Dr. Kaplan's room!

For reference, here is the Revel's native frequency response, in the bass and elsewhere:



*Performance-wise, the Revels are basically unimpeachable. Also, I suspect the big Revel's combination of two bass drivers relatively high off the ground and a bottom-firing port with an EBS tuning frequency offers smoother in-room upper bass than would a single woofer of equivalent volume displacement placed near the floor.)



Not really. I wouldn't use a 200Hz crossover to subs. Well, maybe Parham-style "flanking subs," were I to use them (and there are measurable benefits to doing so).



Again, such things could only be written in the absence of practical experience with such a setup.



You have the weight of basically the entire serious audio community against that assertion.]

Originally Posted by mojomike

Dr. Floyd Toole is simple dismissing time domain

Originally Posted by mojomike

Stereo sound images better when all parts of the sound sources originate from the same proximity.

Originally Posted by DS-21

Actually, had you read the quote and understood its content you would not have the mistaken impression that Dr. Toole dismissed anything. He simply pointed out that one can infer the time domain performance from the FR. If the FR is flat, the time domain performance will be good.

Subjectively, I've certainly heard improvements in "timing" as upper-bass FR has improved. I suspect anyone else who has set up or even heard a high-fidelity bass system would report likewise, because the improvements as one goes from a stereo pair to a stereo pair overlapping with a sub, to a stereo pair overlapping with multisubs is decidedly unsubtle.

Originally Posted by DS-21

Again, a reasonable person can only assume you're talking without a lick of relevant experience.

Also, given your bleating about proximity, do your mains use coincident drivers or some other form of point-source (e.g. Quad ESL-63)?

I've not had a stationary audio system in the last 8 years that hasn't been anchored by mains with KEF Uni-Q's or Tannoy Dual Concentrics. So I know a thing or two about the advantages of getting as much of the spectrum as possible from the same point in time. And because I have functioning ears, I know that the only thing that happens if the bass comes from the same place as the rest of the spectrum is lumpy, amusical upper bass.

Originally Posted by DS-21

I don't know why you seem to be so adamantly opposed to the obvious physical reality that the volume of the cabinet sets the system's low end efficiency. Not the driver, the box.

Again, that's just Hoffman's Iron Law.



Simply incorrect on fact.

There is absolute sensitivity, i.e. x dB at y Hz at when driven with z V, and there is relative sensitivity, i.e. down m dB at n Hz re: p Hz.



We agree on something, it seems.

Box size changes system parameters, however. Such as, for example, low end efficiency.



Obviously, drivers in different sized cabinets will have differing low end efficiency. Nothing new to anyone here, including you or me. So why are you arguing, again?

Also, you're simply not so daft as to not understand the notion of normalized Le, which is Le/Re. You just seem to be playing that way on this thread. I get the feeling that no matter what I write, you're going to say I'm wrong. I could write that speaker wires don't matter as long as they're a sufficient gauge, and you'll counter with "nonsense, you need High End cables that cost at least $500/ft to hear anything."



Wait a minute. Are you claiming what I think you are, i.e. that there is a platonic ideal closed box?

I'd like you to elaborate on that point, because I've taken something quite different from your previous posts on the subject. Perhaps I've misinterpreted.



Hardly. In the real world people have space constraints, and may wish to see if a given driver will give them advantages in a given box volume. A driver that offers increased sensitivity in the top end of the subwoofer's bandwidth (120Hz, maybe as high as 150Hz) will be advantageous, if it still has sufficient volume displacement to reach the desired low end cutoff at a given SPL. The amp will have to work less hard in the upper bass that way, given that the system is more efficient up top and equally efficient (because ULF efficiency is set by the box size) down low.



I've read the graphs. They simply don't support your point.



Perhaps because I assumed that reasonable people would fill in the rest of my sentence. Since my assumption may have been unfounded, here's the rest of it.

"F3 is irrelevant because you can EQ to taste, within the limits of the volume displacement of the subs, which is limited by the drivers' excursion and/or the available power."

Obviously, for any LF cutoff at a given SPL, one needs first to have the volume displacement to get there, and the power to push the woofers to that volume displacement. Among reasonable enthusiasts, those points should be givens.



We agree that inductance is irrelevant to ULF performance, and also presumably also agree that it very much matters for upper-bass performance. Also, I presume we can agree that upper bass appears constantly in musical program material, so it makes sense to prioritize its reproduction over the reproduction of signal that appears sporadically at best. (And is of debatable perceptual relevance in the first place.)

We also now agree that 1W/1M specs are irrelevant in the low bass, apparently, though you claimed above that less efficiency was ipso facto better.

Hopefully, we also agree that the reason why is because the low-end efficiency of a closed-box subwoofer system is set by the trapped volume and not the driver.



Yet, of course, in actual practice you prefer drivers with shorting rings and low inductance. Just as I do.



Though you must concede that it can be, right?



Obviously not. You're asserting the opposite, but aren't able to offer any evidence for that assertion other than the occasional mere presence of ULF. And even then, you've only given one example, outside of the program material we both agree contain ULF that in the current day is probably intentionally-placed and that some people may prefer to reproduce.



Nor have I, only that they are not worthy of use except in very limited circumstances, i.e. as ULF air pumps in systems with multisubs to cover the spectrum above their very limited bandwidth. The primary disadvantage of that approach is that one needs more subwoofers, because a ULF air pump cannot also act as a pressure source in the modal region without compromising the whole system. An advantage is that despite that requirement it may be cheaper, because all of the drive units can be less expensive if they are well-chosen to play specific bandwidth-limited roles.



For most uses, the Dayton Ref is heads and shoulders superior. It can be used as both a ULF air pump and as an upper-bass pressure source at the same time, because of its motor design.



A reasonable person can infer from the design (long, heavy coil, no attempt to control inductance or inductance modulation over stroke) and the lack of a published spec that it has unacceptably high Le to be anything but a ULF air pump with extremely limited top-end bandwidth.



Again, I'm not the one making the rather extraordinary claim that a Sting album released in 1999 (before basically anyone had a system capable of perceptually-relevant levels of ULF) has perceptually-meaningful ULF content.



First, you're misstating. I wrote 120-150Hz.

Second, I have data: upper-bass measurements of actual systems in real rooms.



Obviously not dispositive to a question of perceptual relevance, even to a simpleton.



Totally irrelevant and out of the scope of the discussion.



Way too many assumptions in there that may not hold.

Originally Posted by bossobass

Yeah, got it.

BTW, purpose of flat response to <20 Hz... why does it matter?

Originally Posted by mojomike

The whole point is that a flat frequency response does not necessarily infer time domain performance.

Originally Posted by mojomike

If that were the case, then any system which shows a flat frequency response whether by eq or otherwise would automatically have no time domain problems.

Originally Posted by mojomike

You are simply trading away time alignment in the bass range for a flat frequency response.

Originally Posted by mojomike

If that sounds good to you, that's great.

Originally Posted by mojomike

there are different ways to achieve satisfying audio results.

Originally Posted by mojomike

Unlike you, I'm not claiming that the way I do it is the only way or the best way. It is a means to an end that works very well for me in my room.

Originally Posted by mojomike

No, my speakers are not time aligned in the midrange or in the highs to the degree of a single point source.

Originally Posted by mojomike

Fortunately, there is nothing lumpy or amusical about my upper bass. Perhaps in your case, you were getting lumpy bass because your room had poor acoustics and you didn't want to improve that with acoustic treatment or eq of any kind.

Originally Posted by DS-21

probably "it doesn't, much."

Originally Posted by DS-21

a reasonable reader can safely dismiss the rest of your post as silly hand-waving about a total non-issue.