Top notch speakers with their own sub

They don't exist apart from speakers with dipole bass where you won't get enough output to cover the last octave for things like home theater and organ music (symphonic works at subjectively realistic levels or 90dBC rock/techno are fine with Linkwitz Orions in reasonable sized rooms like 2500 cubic feet or 13x19x8 but open on most of one wall to the entire 600 square foot floor).

With monopole woofers placement for mid+high frequencies is never going to be good for low (below the Schroeder frequency where the resonances no longer overlap thus imposing significant peaks and nulls on the bass response) frequencies because good low frequency placements are not symmetric.

With dipoles there's little (-20dB from the on-axis sound in practical examples) output towards the ceiling and along the dipole nulls 90 degrees off-axis horizontally. The speakers put audibly and measurably less energy into the room's height and width modes. You also get multiple bass sources stimulating the room length-wise with different phase although integrated dipole bass isn't going to be in an ideal location for this.

Linkwitz Orions, some of the Emerald Physics speakers, and the Audio Artistry line which are still made to order have dipole bass.

The better solution is probably designed to integrate with separate bass systems that are placed separately. Audio Kinesis, Lyngdorf, and Gedlee do that. NHT's Xd system did (there may be 10 pairs left).

Earl Geddes is applying custom transfer functions to each of the multiple sub-woofers which should work better than sending them the same signal. Duke could comment on what he's doing.

Martykl writes:

>The Deqx idea I mentioned in my previous post is simply a "brute force" EQ solution. It allows the peaks and nulls to develop and beats them into submission with EQ. Audyssey and Velodyne (among others) also make EQ products for this purpose.

Nulls can't be fixed with equalization. With a 10dB NULL you need 10X the power or 3X the displacement you have at other frequencies. With 20dB it's 100X or 9X.

If you need 100W for sufficient bass head-room and aren't excursion limited, you'll need at least 2000W to overcome a 10dB null due to thermal compression. If you are excursion limited and are stacking woofers in the same location you'll need to triple their count.

Listener and/or speaker movement are the only reasonable solutions for nulls. I had to move my chair six inches to get decent bass in my current living room.

Jax2 writes:
>Hey Duke (or anyone else who cares to comment) - Is a flat response always necessarily an ideal target?

The goal is flat on-axis response with a gradual directivity increase (or decrease in total power response). Your ears take a few cycles to pick up low frequencies so total power response comes into play more there although there seems to be some time domain component with steady-state measurements being an incomplete approximation.

Floyd Toole and Sean Olive at the Harmann Group have done studies on this with blind listening and their computer controlled speaker mover. The preferences hold regardless of listeners preferred musical genre, country of origin, and experience/training in critical listening.

>What I've objected to in some other approaches is that you become very aware of the low end to where it becomes distracting. I don't know whether this is due to overemphasis, room nodes, or some other imbalance.

It's the room and speaker+listener placement. Peaks really over-whelm the music. Placement too close to boundaries increases the whole bass spectrum. In-phase bass signals in the music add +3dB to total power response at high frequencies but +6dB at low enough frequencies. The room has up to 12dB/octave of gain below its fundamental resonance.

Reducing modal problems and room/boundary gain does a lot for natural bass which is like music as opposed to some fast, slow, tubby, or thin approximation that's noticeable and distracting.

Spatine writes:
>Discussion on sub equalization and placement going on today is precisely the reason I hesitate straying from the mainstream speaker establishment. Now I have more plausible theory as to why major speaker manufacturers don't want to package non-integrated subs with their main speakers for music listening just yet. The technology is not sufficiently developed.

The technology is _fine_. Existing implementations are audibly, measurably, and theoretically superior in real rooms. "audiophile" exposure to cheap one-note sub-woofers and poor integration have created marketing prejudice against separate woofer enclosures. Spousal acceptance of additional boxes is an issue especially when placement constraints are taken into account. Some products call for more technical setup procedures and measurements for the maker to tailor transfer functions.

Mainstream speakers sound like speakers not live music due to inherently flawed physics which are addressed in alternative designs. You'll end up with a much more natural sound, better decor match (at the fringes it's small companies with made to order products. Figured woods/veneers, inlays, and finishes can be mixed and matched), and spend less money (you're mostly paying for parts and a furniture maker's time).

Speakers like the B&W Nautilus Prestige (acoustically small drivers+baffles with damped transmission line enclosures) and RAAL Requisite Eternity (uniform horizontal polar response, and I'd guess that the vertical response matches up well around the cross-over frequencies) take some steps to get there, although tighter uniform dispersion interacts less with the room. Bass needs separate enclosures and/or dipoles. Room specific bass filter functions are a good idea.

> Secondly, the idea of having 4 sub is quite intrusive, one way to get into major fight with your wife.

All but one location can use smaller woofers (they don't have to extend below the room's fundamental resonance, so they're not really sub-woofers) in smaller enclosures. AFAIK Earl Geddes is using 6th order band-pass boxes which can be 6dB more efficient for a given cabinet size and low frequency cut-off than a sealed design. In theory they ring more than a sealed or 4th order ported design, but the room dominates time domain behavior. In theory they have more group delay, but you need cycles of audio to pick up bass. Generalizing from small "sub-woofers" sold to the mass-market with a strong resonance to substitute for real bass doesn't apply to good ported designs and shouldn't be applicable to 6th order ones.