It's predictable for someone like Steve to say that 3-15Hz is theoretical, meaningless, audiophile mumbo jumbo, or whatever Steve's latest explanation is...because he has no output in that region.
If by looking up facts Steve means visiting a forum that publishes the latest GP test results to 10Hz or 15Hz, then my case is already made for his biased opinions.
Let's see what real science reveals about my audiophile mumbo jumbo.
The frequency spectrum has a periodic structure with zero force at n/T intervals where n is an integer and T is the duration of the transient signal. The frequency range where the transient signal can be analyzed accurately is from 0 Hz to a frequency f after which the signal magnitude decays by 10 to 20 dB.
A most important feature of the magnitude graph (relative to our earlier periodic-function spectra) is that our transient signal has frequency content at every frequency, not just at discrete frequencies. Our example also shows that the frequency content gets smaller as we go higher in frequency. This trend will be observed in all spectra of transients; thus we can use such spectra for estimating the highest frequency for which a system model must be accurate, just as we did with Fourier series for periodic inputs. Note also that much of the frequency content is located below the frequency 1/T, so 1/T is useful in roughly gauging where the spectrum begins to drop off. Thus if T=1.0 second, we have strong content to 1Hz, whereas with T=0.001 second, we have equally strong content to 1,000 Hz. That is, the shorter the duration of a transient, the more its spectrum extends to higher frequencies. This feature is applicable to general transients, not just our simple example.
So, a transient, any transient, foot steps, a car door slamming, a kick drum, a snare drum, a cough, a mortar round, a gunshot, etc., etc., has content to zero.
It's interesting that there exists an industry that is paid to design the sound of a closing car door. "Improving the Sound Quality of the Closing Sound of a Car Door" is one of many studies, all of which analyze the component parts to the nth degree to result in a transient event that will be associated with quality.
Every one of those tests includes analysis to and emphasis on single digits.
I also take a 'snobbish' approach to those who say that single digits in movie soundtracks are BS, artifacts, useless, mistakes, etc. because the state of the art in soundtrack production involves extremely complex equipment that isolates and synthesizes real events, like a lightning strike, and combines parts of synthesized versions of real events to arrive at the best possible effects for a given scene.
These aren't some knob jockeys with a Casio keyboard. I have friends who do this for a living, and I wouldn't insult their intelligence with those comments. I certainly will never agree that Spielberg would preside over incompetent effects production that places single digits in a soundtrack by mistake or even though they thought it irrelevant.
Visit the master list sticky and see the number of movies with <10Hz content Here
as well as a partial list of sub 10Hz soundtracks that include prime time network TV shows and Superbowl commercials here
from the TRW (the 'fan sub' Ear refers to) site.
The scientific fact remains that all transients have content to zero. This can easily be verified by anyone who wants to do so. Is there a 125dB requirement to properly reproduce the sound of a car door closing? Of course not. Will it sound the same with a 15Hz high pass filter? Apparently not according to those whose job it is to analyze and utilize such an event, but that's up to the individual to decide (which is hard to do if you lack the capability to compare.Here
is a blurb from a company that's involved with car door closing sound. The graphic is poor but you can see the reference graph (in blue) has less higher frequency content and more lower frequency content. The interesting thing to note here is that even the cheapest car door closing spectrum is to single digits.
I'm certainly not saying that everyone should attempt to achieve single digit playback capability. I couldn't care less who has what or whether or not they're happy with it. I'm saying that it's the next frontier. I'm also just trying to state the obvious. Having a full spectrum playback is better than having a truncated one.
Again, short of the price and logistical hurdles the TRW presents, a sealed system is the only way to reproduce the source material. DIY's biggest advantage over commercial is, for me, in this area. Otherwise, I would bought a commercial sub. Of course, over the years, you'd have had to buy another, and another and another, etc., as the need to upgrade manifests itself in the commercial world.
Pointing to Ilkka's tests as the reference, or "the great equalizer", is like saying CD is the reference for all digital disc formats.
For example, the JL Gotham would top his lists, and the many excellent on board features would barely get a mention much less be tested, but the price would be talked about by guys like Steve as the reason it shouldn't be considered at the top of the list, which, BTW, only charts comparisons down to 12.5Hz, and only considers comparisons to 20Hz.
I first saw the g213 at CES in 2004. It compares to a subwoofer I built a year before that with similar signal shaping features, although my on board signal shaping is analog, similar power plant, although mine is outboard and similar enough driver compliment. Just take Ilkka's f113 numbers and add 7-8dB, more down low due to better extension capability.
Now, add 12dB to those numbers, add anechoic F3 capability to 9Hz and modify the HP in your signal chain and you have a similar system to mine. I hardly think 4 of Steve's subwoofer would be practical, and even if it were somehow possible to fit them into my HT and tolerate the loss of space, I'd still only have large passive ported subwoofers with cheap pro amps and no signal shaping capabilities with no output in the single digits and a steep roll off.
Can anyone tell me why on earth I would ever consider doing something like that? Ilkka's test results? Some hidden 'facts and measurements'? Cost savings? More output at 16Hz and less everywhere else? The need to lower my X-over point? The carpeted tubes would double as sound absorption? They would 'mimic' a sealed sub?