[josh358]

Just wanted to draw people's attention to the dual bass array, which as I'm sure you know cancels all room modes in a rectangular room below a frequency determined by the spacing of the spatial Nyquist frequency of the woofers:

http://www.avsforum.com/t/837744/double-bass-array-dba-the-modern-bass-concept

For those interested, a single bass array does the same, but rather than using a second array of woofers to cancel the reflection it adds a delayed, polarity-reversed signal to the front array. I've been able to achieve a SBA effect with line source dipole woofers, using the reflected rear wave to achieve the cancellation.

It may be worth noting that dipole woofers trigger fewer room modes than omnidirectional woofers in any configuration, line sources even more so, particularly if they're parallel to the front (or side) wall. If parallel to the front wall, the excite neither the height nor width axial modes. They do however suffer from rear wave cancellation which can lead to suckouts in the midbass.

I agree with Scotty and Jim BTW that electronic EQ, while useful and even necessary for best results, isn't a panacea. It can normalize frequency response at the listening location, and as such, is an improvement, particularly for those who can't use acoustical treatment. But with some minor exceptions, it can't fix the time domain behavior -- the ringing or smear and, at worst, pitch shift. It can only normalize the average level, integrated over time. And it can do so only for one location. Placement, multiple woofers, and bass trapping still have a role. One interesting possibility -- it seems that if you place a woofer next to the listening seat, rather than up in front, the response variations are minimum phase. This means that, in principle, it should be possible to equalize them out.

Finally, I'm not sure what to make of the stereo bass business. LP bass is essentially mono. CD's of course have the potential for stereo bass. The ear can localize bass laterally at any frequency, but in a small room, things generally become too confused. However, there are many reports of stereo woofers adding a sense of space. I'm not quite sure what to make of that. The effect may not have anything to do with stereo in the recording. Also, I have a paper somewhere on subwoofer localization. It seems that many people can localize woofers down to 80 Hz or lower. This is highly dependent on angle, and also the specific acoustics. I've seen this effect myself -- a sub that can be localized in some positions, not in others. Also, depending on slope, a sub crossing over at 80 Hz will typically have audible output at least an octave up. This may account for some reports of spatial improvement with stereo subs.

[Quiet Earth]

I have a question regarding the multiple sub concept. If your main left and right speakers already go as low as many subwoofers go, then shouldn't you count the mains as the first two subs that you already own? For example; if I add a single sub to a full range pair of speakers, I would effectively have 3 subs. If I add two subs I would effectively have 4 subs, and so on. Does that sound right?

If this is not correct, please explain why not.

[Rclark]

here is an excellent video that spells it all out

http://www.youtube.com/watch?v=ayOXz26qm3I

[jtwrace]

here is an excellent video that spells it all out

http://www.youtube.com/watch?v=ayOXz26qm3I

Meh.  It's not quite that easy.  It takes a bunch of time and patience.  I suggest you read Dr. Geddes' paper.

[*Scotty*]

Got it one, Quiet Earth. The caveat being that those front main loudspeakers have to be pretty close to the front wall or you won't satisfy the requirements necessary to create and launch a planer wavefront from the front of the room towards the rear of the room. If your speakers are too far away from the front wall you will generate a series of non-planer wavefronts that create standing waves associated with the rooms ceiling height and the distance between the side-walls which cannot be canceled out by locating subwoofers at the back wall of the room.
  The signal to the rear subs also has to be delayed by the amount of time necessary to insure that a coherent bass wavefront impacts and passes by the listening position before it is canceled out by rear subwoofers.
By canceling out the planer wavefront just before it hits the rear wall you ensure that you have "flat" bass response at virtually all points in the listening room.
Here is a link to Adrian Celestinos' doctoral thesis about optimizing the bass response in a room. His doctoral thesis fully describes the advantages of attacking the problem of standing waves in the listening room via the time domain instead of the frequency domain. http://vbn.aau.dk/files/12831869/AC-phd.pdf.
Scotty

[James Romeyn]

I have a question regarding the multiple sub concept. If your main left and right speakers already go as low as many subwoofers go, then shouldn't you count the mains as the first two subs that you already own? For example; if I add a single sub to a full range pair of speakers, I would effectively have 3 subs. If I add two subs I would effectively have 4 subs, and so on. Does that sound right?

If this is not correct, please explain why not.

That is a great question.  Potential weakness of the above philosophy: Down to about 200 Hz, the main L/R speakers generally sound best sited in lateral symmetrical fashion, equidistant from sidewalls near the speakers.  As I understand it, above 150-200 Hz, the ear would adversely notice disparity in different delay times resulting from non-symmetrical side wall spacing.   

Conversely, it is ideal to avoid any and all symmetrical, repetitive spacing for drivers reproducing the bass range.  Consistent with the above, makers of single column full range speakers, do encourage disparte L/R wall spacing for this very reason. 

The above dichotomy lead Jeff Hedback (Hd Acoustics) and Nyal Mellor (Acoustic Frontiers LLC) to write:

To obtain the best possible LF response…boundary interference issues can be tougher to address. Varying the fixed distances from ‘speaker to boundary’ and ‘listener to boundary’ will reduce strong cancellations. It is a balancing act as one location that may offer a smoother LF response may not provide the optimal midrange and treble response…” (emphasis added) and “…everyone desires a ‘flat’ LF response and no modal ringing. Simply, this is a tough achievement. The absurdly large collection of interrelated variables between two fullrange speakers and the room (speaker design, speaker/listener location, room size/construction and acoustical control within) makes this so. It is up to the individual to determine what their limits are as regards placement and acoustical treatments…

(some emphasis in original, some added…note the term “absurdly large”) http://www.hedbackdesignedacoustics.com/files/QuickSiteImages/AMS_for_Stereo_List._Rms.pdf

Completely independent siting of mains vs. subs is a defining feature and benefit of multiple distributed subs.  Those like me steeped in this philosophy consider any full-range single-column speaker an archaic, terminally flawed audio philosophy.  This is because domestic rooms interact with sound waves differently below vs. above about 200 Hz.  (Ancillary benefits of distributed subs: ability to employ stand mount monitors with improved spatial effects and smaller panels for less noise, less likelihood of internal resonance in the mid bass pass band, and my Ambiance/Mode Cancelling Array if you can afford a second matched set of monitors.) 

Multiple subs naturally result in physical distance between the subs and mains.  Theoretically 1/4 the crossover wavelength is the maximum distance.  My subs cross 4th order @ 70 Hz with a maximum distance between sub and main speaker of 18'.  Speed of sound 1100 fps/70 Hz = 15.7' x 1/4 octave = 4' maximum theoretical spacing.  Yet, in practice in my system, the subs cross absolutely seamless, and are impossible to locate, even standing in front any sub.  The entire room energizes with bass energy in transparent fashion, yet it generally sounds like the bass instruments are located at the front of the stage.     

Multiple subs effectually make the boundaries disappear in the bass range, while the boundaries may be properly employed to add appropriate ambiance in the range above the bass.  Also, too much absorptive makes bass sound overly dry.  Effectively damping bass modes requires absolutely huge quantities of OC703 (Ethan Weiner of Real Traps recommends buyers purcahase as much as they can afford), enough to make bass sound strangely unnatural.  I know this from experience.  I built an acoustic soffit in my last room (equivalent Acoustic Science Foundation value about $7k), properly treated even for high-frequency dispersion.  It still sounded strangely overly dry, and did not fully cure bass modes.       

My 1cf sub are easy to site against and firing toward walls.  The first sub sites in the room where it produces the smoothest bass (by itself) at the sweet spot.  The distance from Sub1 to its nearest corner x 1.62 (Golden Ratio) = Sub2 spacing to its nearest corner.  The same relationship exists between Sub2/Sub3, and Sub3/Sub4. 

Sited in this way, each subsequent sub spacing is least likely to repeat/replicate the modal effects resulting from the prior sub placement.  The equation maximizes random chaos in the bass range for the least bass mode effects.  (My current practice is to invert one sub's polarity for the same reason.  I plan to experiment to see if the randomizing benefit is worth the loss of 20-25 Hz energy.)  With one exception, this equation will work for any four subs, not just my own.  I'd like to hear feedback if anyone else tries.  Actually, you can try it even with four full range speakers, as long as they are not high-Q in the bass.  Use your own two full range speakers and a couple subs, or borrow two full range speakers from a friend.  Make sure your friend hears this experiment.  Actively cross the full range speakers around 70 Hz and site them as described above (obviously you'll have no sound above 70 Hz, but don't worrry about that.)  After you set it up, disconnect three "subs" to hear only one "sub", advance gain 4.5 dB for equal loudness, and listen.  I virtually guarantee immediate conversion.  You can't go back.  Trade the large full range single column speakers for stand mount monitors (then try my ambiance/mode cancelling array for two monitors).               

As Scotty and others properly stated, audibly horrific FR abberations (+/- 12 dB is common) resulting from bass modes is the least audible problem.  Timing distortion, bass notes ringing long after the bass player or players already hit the next note, is far more detrimental to overall performance.  This is a universe worse than harmonic distortion in the bass range, exactly replicating the audible effect of two bass musicians, or two sections, where only one is on the recording.  Remember the human ear/brain mechanism is extremely tolerant of THD in the bass range (possibly as high as 30%!), but absolutely nothing masks two dis-harmonious bass notes where only one exists on the recording.  One musician or one section plays completely different musical notation vs. the other.  Once you become used to the absence of this awful effect, it is easily discerned in virtually any system employing two full range single columns speaker regardless of cost/complexity.  Some systems can mask it pretty well, but never completely.  It constricts the size of the venue where it would otherwise sound immense and without boundaries.   

Thanks for the post about Dual Bass Arrays, which I found extremely interesting.

[Rclark]

Meh.  It's not quite that easy.  It takes a bunch of time and patience.  I suggest you read Dr. Geddes' paper.

I will. I've been reading this sort of material lately. But this guy's advice seems like a solid place to start: 4 subs placed in the middle of each 4 walls, mono, in phase.

[James Romeyn]

I will. I've been reading this sort of material lately. But this guy's advice seems like a solid place to start: 4 subs placed in the middle of each 4 walls, mono, in phase.

I tried the above, four subs/one per corner, and four subs in various random spacings from corners.  IIRC, Todd Welti (he and Toole both at Harman Labs now) covers these placement options in one of his AES papers.  All the above sitings work better than two full range single column speakers.  The method I describe above works better IMO.  Distributed multiple subs can not be matched high-Q, so beware of that.  High Q are reflex speakers, generally not sealed.  It's pretty easy to stagger high-Q by adding various port stuffing to four matched high-Q subs (maybe seal the port on one of the speakers).   

[*Scotty*]

James, have you had a chance to read the information contained in the PDF doc http://vbn.aau.dk/files/12831869/AC-phd.pdf
This approach allows you deal with the room's resonances below the Schroeder frequency as though the room has only significant resonant mode, that associated with the rooms longest dimension. The room's height and width are taken out of the equation which greatly simplifies the task of achieving flat bass response in your room.
Scotty

[rodge827]

Gents,

I have a dual sub and monitor configuration (2.2), see system on my profile.
I was running the subs to the left and right of the monitors. The sound was good but not great.
The bass was soft and somewhat muffled in tone.
I was lent a Dspeaker 8033s from the distributor and took the couple of minutes to put in the system, ran the algorithms and WOW!
All bass issues were gone, the midrange opened up and the music had meaning again. Bass was tight and deeper than ever and imaging was off the charts.
The bass was so good and got better when I removed all the big and bulky room treatments.
I have since sold the 8033s and purchased the Dspeaker Antimode Dual Core 2.0 $1100.00.
The D/C is an incredible piece of gear, frequency adjustment from 500hz on down , phase adjustments, graphic EQ, multiple inputs, custom house tools, and all with out a computer. The whole set up time took about 10mins and I was in musical heaven.
Check out the D/C from the link and take some time to read all about the other Dspeaker products. Some may be what you may need.
 
http://www.dspeaker.com/en/products/20-dual-core.shtml

Oh yeah, Robert E Green (REG) Golden Ear Award on TAS!

Enjoy,

Chris
   

[Habs Fan]

I found this cofiguration to work well in my room which just happens have near identical dimensions as the example room.http://musicanddesign.com/DP_woofer_room.html

[James Romeyn]

In TAS Issue 204, page 41, TAS Senior Editor/UCLA Math professor Robert E. Greene wrote in his first DSpeaker review:

…in principal the multiple subwoofer concept, as realized in the Audio Kinesis (Duke LeJeune) Swarm system, should work better than the single sub that I was using-I hope to try that system later...

Since that time Duke achieved great music industry success with his Thunderchild and Thunderchild A/F (acoustic friendly) cabinets for bass and all acoustic instruments (any instrument except such requiring distortion, such as solid body electric guitar). 

I'll frankly admit when Duke first showed me his early SWARM circa 2007, with four smallish 8" subs, I thought it seriously needed about three less subs.  Conversely, now I consider such systems to define state of the art bass reproduction in a domestic space, and don't take very seriously any audio system not employing the technique to flatten bass mode effects.   

[James Romeyn]

Swept Sine Waves, 0 dB = 60 dB with mic @ sweet spot.     

First column is test tone frequency. 

Second column is one sub ideally sited for smoothest bass (one of DEBRA's four subs employs this location) no EQ, no treatments.   

Third column is DEBRA (Distributed-EQ Bass Reflex Array), four 1cf subs, one sub phase-inverted, no EQ, no treatments.  I loaded about 30cf OC703 into my last room, with dispersive exterior film, including a professional 3-wall acoustic soffit, with only mixed results. 

                    One Sub     Four Subs (One Inverted)
100    Hz    -11.5  dB       -1 dB
  80    Hz      -9     dB        -3 dB
  63    Hz       0     dB       0 dB
  50    Hz      -4      dB     -2 dB
  40    Hz     +0.5   dB     +3 dB
  31.5 Hz    -12.5  dB          0 dB
  25     Hz    -10     dB        +3 dB
  20     Hz     -6      dB      -1.5 dB

Frequency Response Aberrations From Flat

One sub: +.5 dB/-12.5 dB, much larger peaks and nulls throughout the room, especially huge corner modes
DEBRA +/- 3 dB throughout entire room including all four corners

Again, the numbers above hide DEBRA's best audible advantage, being complete absence of timing smear caused by bass notes audibly ringing after the original note stopped.   

Below is my latest, more reliable, and more efficient setup instructions.  I encourage readers to try these instructions at home with any combination of four subs and/or main speakers active low-pass crossed at 70-80 Hz.  IMO it's worthwhile to hear the results as long as you don't employ subs with matching high-Q resonance.  Likely less linear than DEBRA because of its RGC (room gain complementary tuning) and other differences.

The point is to compare response aberrations, with meter and audibly, one sub at the ideal location near a wall vs. four subs setup properly in distributed array.  One sub requires +4.5 dB gain for same SPL (would be 6 dB if all subs were in-phase). 

Again, note more the advantage of a complete lack of time smear with the array.  With standard bass systems (two full range speakers, stereo subs, etc) bass notes "ring" after the bass player (or section) hits the next note, resulting in two disharmonious notes playing simultaneously, the dreaded bass note smear.  Audiophiles almost always wrongly attribute this phenomena to speaker tuning, whereas it is almost always a function of speaker location, boundary location, and listener location. 

As Scotty correctly pointed out too, it is virtually impossible to overstate how much spatial performance is impaired, no matter how good you think it is now.  (For reference, stereo is an inferior benchmark for ultimate spatial quality, and I'm not talking about standard multi-channel recordings...see Ray Kimber's proprietary Iso-Mic 4.0 DSD recordings played back on eight huge Sound Lab stats, two per corner...these speakers will not be displayed again because of wear and tear on the stats.)

Once you hear this "cured" I don't think you can go back.     

Maybe try this with another audiophile, combining subs and/or main speakers, cutting the work load, both having the opportunity to hear the results.

If you have no active sub crossover with 70-80 Hz pole, consider using the sub output of an appropriate DD processor, as long as it allows low enough crossover pole.   

Copyright instructions, James Romeyn:

Siting Four Subs

The following mimics or “mirrors” the boundary, subwoofer, and listening seat relationships during normal listening.

Measure and note your ear height in your favorite listening seat.  Set the sub amp crossover pole to 70-80 Hz, no EQ, and if phase control present set to 90 degrees.   

Clear all space including furnishings about 2-3′ from all walls or from as much of the walls as possible.  Clear the space normally occupied by the main listening seat.  Under the primary listening site center one "sweet spot sub".   

Through the "sweet spot sub" play well recorded music program with electric or upright bass, preferably with brief note values and lasting at least one-two minutes.  Kneel or bend at waist to locate your ears about 2-4″ from the wall, equal to ear height in the listening seat.  Listen at all points 2-4″ in front of all wall surfaces, for the smoothest, most linear bass, with the least ringing and lowest overall gain.  Attach a “Post-It” note to walls at one to three sites maximizing the qualities described above.  Among the “Post-It” notes select one site with the best performance.  Take your time with this process.  Double-check results. 

Site “sub 1″ near the wall at the final location with the best performance.  Now play music through the “sweet spot sub” and “sub 1″.  Repeat the above procedure, selecting two to three preferred locations, then narrowing it down to one ideal location.  Locate “sub 2″ at this point. Now play music through the “sweet spot sub,” “sub 1,” and “sub 2″.  Repeat the above procedure, selecting two to three preferred locations, then narrowing it down to one ideal location.  Locate “sub 3″ at this point.  Now play music through the “sweet spot sub,” “sub 1,” “sub 2,” and “sub 3″.  Select two to three preferred locations, then narrow it down to one ideal location.  Move the “sweet spot sub” to this final ideal location.   

The four subs are now best sited to act as four new random modes, unequal to the room’s fixed modes, thereby maximizing potential for the smoothest and most linear bass throughout the entire space of the listening room.

Sub Amp Phase Control

Set this rotary control knob at 12:00 (90-degree phase) for maximum expansion of the sound room’s boundaries, thus minimizing modal effects.  The final stage, inverting the polarity of one sub, further minimizes modal effects.

Invert Polarity of One Sub-Wear Noise Protection! 

A tripod-mounted SPL meter at ear level at the sweet spot, and swept sine waves of 100 Hz, 80 Hz, 63 Hz, 50 Hz, 40 Hz, 31.5 Hz, 25 Hz, 20 Hz greatly aids this process.  Lacking such, we recommend, as before, employing well recorded jazz upright bass playing a long section of fast notes of equal time value.

Inverting the correct sub produces the smoothest, most linear bass, with the least ringing and lowest overall gain.  Play all swept sine waves above four times, noting SPL each time, each time inverting the polarity of a different sub.  You will easily note smoother performance with one particular sub inverted vs. the others.
 
 

[DaveC113]

Swept Sine Waves, 0 dB = 60 dB with mic @ sweet spot.     

etc...
 

Thanks!

[Folsom]

Wow... I am impressed that this got broken down to such a science. But I got to say real world experience of playing with 2 subwoofers in stereo for years... I only believe in one setup.

I prefer subwoofers in the corners of the left and right, pointed to the left and right walls (not diagonal, not to the corner, not to the rear wall). After trying every direction (including tilting) this is the only position I will ever recommend again for stereo subwoofers. It was by far the most musical, and accurate. I even had a closet on one side and it still gave the best experience. Pointed at you loses too much due to wave length, and pointed at the opposite way does the same thing. Pointed in, well just doesn't make sense (obviously). I think pointed out left and right simply makes the lower frequencies readily be absorbed into the room like the frequencies from bass instruments would be. Even if they are in low less audible ranges, and sometimes non-directional, the loading of the room is a perceivable thing.

Also to note I don't believe in mono subwoofers except under something like 40hz. In simple theory it is mostly all non-directional but to re-iterate that isn't true of how we perceive it unless we are directly hearing the sound-wave (hard to do with such long waves). I tried on multiple occasions to use just one and get a good sound. It always failed, always left the music boring by comparison. Small sized speakers can't do bass right (something about how it sounds because of dispersion I guess, it doesn't roll off small speakers right), and neither can single subwoofers. I prefer subwoofers with a little talent, and reaching higher up into the 130-200hz range though, too.

Funny part? I have Horn Shoppe Horns only, because of apartment space constraints. They are fantastic but the richness of full bodied powerful sound I had with dual subwoofers... well I miss something closer to a live experience and some day plan to regain that. But heck who else has Ed's Horns in a studio apartment, or anything near so nice? I'm lucky.

[James Romeyn]

Note: The single sub (87 dB) siting (the smoothest location in the room near the walls) has a 12.5 dB trough @ 31.5 Hz vs. the DSA which was flat.  Presume all four subs sited at this single location, all in-phase (93 dB), or +6 dB extra gain.  Still, the -6.5 dB trough requires 4.16 times the amplifier power.  My amp is rated 1kW.  4.16kWrms exceeds the subs power handling by 1616W.  I know of no sub with 4.16kW power handling @ 93 dB sensitivity.  If one exists, it's huge and extremely costly and heavy!  +6.5 dB EQ @ 31.5 Hz worsens FR at other locations away from the sweet spot.  Have you ever seen anyone promoting EQ post before/after results other than the sweet spot?  Especially between the speakers?  Do you think the modes existing between the listener and speakers have no effect on spatial performance, nor phase relationships above the Schroeder frequency?

Considering how low is the cost for computing power (receivers <$500 msrp include mics and automated EQ), why, after this many years, is EQ not considered absolutely mandatory for every single reference-quality reproduction system?  Meridian offers this since about 12-15 years ago.  How often is it mentioned as a world-class reference?  I must admit to minimal experience with Meridian's digital EQ systems.   

I plan to measure FR around the room, even in the corners, both with one sub and the DSA.  I'm sure the results will be interesting.         

Stereo vs. Mono < 70 Hz

Speed of sound 1100fps/70 Hz = 15.7' wave x 2 = 31.4' for one complete (audible) cycle.  For signals 70 Hz and lower (longer) human perception requires some ratio of signal bounce between walls spaced less than 31.4'. 

I'm anxious to find the error above, really.  Possibly transform functions contradict this.  If it's correct, the math seems to support the claim that no microphone nor human ear can perceive stereo (disparate) audio signal below 70 Hz. 

If the math is incorrect, and we could perceive stereo bass below 70 Hz, the DSA would fail crossed @ 70 Hz.  In this room bass is completely linear relative to location, even in the corners.  Also, standing directly in front of any sub one can perceive no sound from it, neither with music nor swept sine wave test tone below 70 Hz.  It's barely detectable with about 6" between your head and one sub.  At all times the stage is front-loaded, even with bass, yet the entire room is energized with music power in a way no other bass reproduction system can match, at least that I know of.     

The math is my understanding behind Geddes' suggesting that solving the issue of modes is the first priority for ideal bass reproduction.  Compared to the issue of modal effect, everything else associated with ideal bass reproduction is practically irrelevant, simply because the variables are well known and their solutions well established for many years. 

I do think the FR described above, especially -1.5 dB with four 1cf boxes, supports Duke's claim of the importance of his RGC (Room Gain Complementary Tuning), which high-pass filters (mechanically via special tuning) @ the rate of about -3 dB per octave, the inverse of the average trend of room EQ.   

I very much enjoyed all the reading in this thread. 

[*Scotty*]

Speed of sound 1100fps/70 Hz = 15.7' wave

This is correct.

' wave x 2 = 31.4' for one complete (audible) cycle. 

This is incorrect.
 Here is an graphic of a sine wave illustrating how a wave length is measured.
The wavelength of a sound wave is the distance over which the wave starts at a point on one curve, goes through an entire cycle, and  returns to that point on the next curve. The diagram below offers a visual definition of wavelength. The space between those two points are considered one cycle of the wave.
             
              15.7 feet

Scotty

[JohnR]

Have you ever seen anyone promoting EQ post before/after results other than the sweet spot?

It's not hard to do, but seems a lot of work for no reason... Where are your results posted?

Do you think the modes existing between the listener and speakers have no effect on spatial performance, nor phase relationships above the Schroeder frequency?

Do you have any links or references on this? Thanks

[DaveC113]

Do you have any links or references on this? Thanks

That's what Scotty explained to me earlier in the thread too. If the air between your speakers and ears contains +12 dB bass modes I would think that would be somewhat akin to an obstruction, at least partially. I have found good bass is really important for the best imaging/soundstage. But I do agree it would be interesting if any research was done in this area...

[James Romeyn]

Thanks Scotty.

I don't use the single-channel analog parametric EQ in my sub amp.  I have a 2003 Pioneer receiver with automated digital EQ, but never used the EQ.  The DAC is pretty awful full range.  Maybe it's OK for bass.  If I try it, only for the subs, I'll test FR changes at different locations.  Such EQ has limited range, almost positively less than 12.5 dB needed to fill in my 31.5 Hz null.  I'd need only 6.5 dB gain if I was willing to stack four subs in that one location, but I'm not.  Even if I was willing, I'm a bit short amp power and sub power handling.       

I'll test and post FR with one sub at the smoothest site vs. the DSA, at:
1. A corner with the worst (or among the worst) modal juju when one sub is employed
2. In the location where would otherwise be a seat just L, R, or centered/behind the sweet spot (whichever site has the worst modal effects when one sub is employed)
3. Some location with modal effects when one sub is employed, in the space between the speakers and the sweet spot (possibly L of center and R of center, toward the question of modal effects on spatial performance...imagine a 6-8 dB null on the L side and a 6-8 dB peak on the R at the same frequency...how can there be separate "stereo" bass signals below 70-80 Hz when FR and phase are so dependent on almost infinitely erratic and variable relationships between listener(s), two speakers, and boundaries?) 

This room has impossibly bad modal effects.  Results will be interesting.

Did anyone see Kalman Rubinson's article in Nov. Stereophile?  He tested the latest highly-regarded DSpeaker stereo EQ, about $1k IIRC.  Frankly, I was pretty shocked how bad was the bass FR shown in the graph.  Prior to EQ it was horrible, with troughs no known sub could fill.  Post EQ a bit better, but still pretty bad.  TAS' Dr. Robert E. Greene loved the first DSpeaker sub EQ a few years in ago.  He mentioned a desire to test Duke LeJeune's DSA, adding it should perform better than his sub with the DSpeaker EQ, which combo he raved about.