Subwoofer: should we even use them at all?

In October 2011 Jeff Hedback (Hd Acoustics) and Nyal Mellor (Acoustic Frontiers LLC) published "Acoustical Measurement Standards For Stereo Listening Rooms." Hd Acoustics clients include Ozzy Osbourne, Lifehouse, and Trevor Horn. Page 19: “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) Page 20, paragraph 2: “ 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 original, some added note the qualifying phrase “absurdly large“)

Let us narrow the discussion to bass reproduction systems (subs or full range speakers) properly tuned with appropriate system Q. Audiophiles wrongly describe such systems as having "fast" or "slow" bass, and such descriptions are misleading or worse. Would you want a low E string on a 6-string guitar to be "faster" than the high E? No, of course not. Low E is 82 Hz, high E is 328 Hz. Obviously, the Low E must be "slower"...two octaves "slower" to be exact.

Guess what? Bass is a slower wavelength vs. mid/treble. Of course it must be slower. Double basses are larger than violins by design.

"Slow" bass more accurately describes a bass mode effect unrelated to the speaker itself. Even though modal effects often result in Frequency Response windows of 15 dB or worse, that is not their worst and most audible effect. The worst and most audible effect is that the mode wavelength is completely outside the diatonic scale and unrelated to the original bass note pitch. After the bass note ends on the program, the modal effect continues bouncing between boundaries. Imagine two bass players, one of them plays perfectly, while the other bass is tuned a quarter tone off (sometimes plus, sometimes minus, depending on the mode), and he holds bass notes about a 32nd note too long. This is the inherent effect modes cause, which results in audiophiles wrongly using terms like "slow." Yes, the note didn't end, but the speaker is not reproducing the note, the boundaries make the note. We're talking about the equivalent of over 100% THD. We're talking about bass notes not on the music program.

There are really only two ideal solutions, neither of which require any EQ nor acoustic treatment for small 3 to 6 dB FR windows, and virtually no timing errors as described above. One is a Distributed Sub Array, the other is a Dual Bass Array. The latter might perform even better than the former, but the latter also costs about twice as much and ideally requires subs built into both front and rear walls.

A properly installed Distributed Array has no integration issues up to about 80 Hz because it solves the bass mode before it occurs. All other solutions are band aids attempting to fix the mode after the fact, including EQ, which can't solve the timing error.

Initially, three or four subs seems like 2 or 3 too many. When you hear the performance level, you'll wish you found out about it earlier and never look back. Even 1cf subs can reach -3 dB @ 20 Hz with enough power for HT or music. That's another benefit: ideal bass works just as well for music as it does for HT.

I, too, dislike the idea of calling bass response "slow" and prefer the term underdamped. I also like the idea of multi-sub arrays and have experimented with variations on the theme pursuant to some great posts by Duke of Audiokinesis, so I agree with a lot of this post, but....

Spkrplus wrote:

A properly installed Distributed Array has no integration issues up to about 80 Hz because it solves the bass mode before it occurs. All other solutions are band aids attempting to fix the mode after the fact, including EQ, which can't solve the timing error.

If "no integration issues" means "no room integration issues", this is an overstatement. You won't remove room effects with four (or more) subs, but you can greatly reduce their impact on direct FR and power response with careful placement. As a side note: EQ fixes the problem before it occurs, too. Once EQ filters are set by room analysis, the subsequent electrical (music) signal is adjusted prior to reaching the driver. Tho the filters are calculated "after the fact", that does not a band-aid make. If you doubt that this approach works, simply do an RTA of the direct and power response in the room (I use white and pink noise) post EQ. The benefits in both graphs are obvious, tho usually more dramatic IME on FR than power response.

The difference in power response improvement is - to me - the primary conceptual benefit of an array vs EQ, but per my own measurements, I wouldn't personally describe that delta as dramatic. My own array experiment - admittedly not ideal, two Velodyne SPL 8s and two 12" Rythmiks optimized for power response - showed power response improvement over the dual Rythmiks, but not wildly different. In fact, the power response and FR performance of that array was clearly improved by the application of Audyssey, too.

Proviso - a different subwoofer array implementation might yield different results, but my attempt was reasonably diligent. Had I gotten more dramatic results, I'd likely own an Audiokinesis array right now. To be clear, I wouldn't bet against me buying one in the future, but I'll still use Audyssey with it.

As for long wavelengths reflecting off boundaries (2 bass players), this happens with distributed subwoofer arrays as well as single subs. As a technical matter, it happens more - reflections for each sub are inevitable (4 or more bass players). Careful placement of the subs relative to the walls can smooth the response anomalies caused by each of these reflections, but they're simply smoothed in aggregation, not removed.

It's also unclear what "timing error" EQ can't solve. If you're talking about decay characteristics of the room, I don't see how a sub array would be helpful there either.

As for in-wall woofer solutions, they do effectively address the issue of reflection off the wall behind them (this is generally the worst IME) but there are still side wall, floor and ceiling issues which are certainly not trivial.

To be clear, I'm not dissing the idea of subwoofer arrays. I actually think it's a great idea. I'm just pointing out that the particular benefits of an array are overstated in this post and the objections to EQ are also overstated. Of course, the bottom line is a listening session. IME, thirty seconds with Audyssey should be adequate to demonstrate the benefits to most people (simply disable it during play). That doesn't mean it's preferable to a well implemented array, it just means that it's a very effective solution for the bulk of the problem. The best practical solution that I know of is an array with Audyssey, but, if you really want to "fix" the issue, an anechoic chamber (or room that effectively behaves as one) is the only solution that I'm aware of.

Marty,

Thanks for the answer. Was wondering how is your signal chain configured? From server, do you go out through USB? Straight into the Onkyo? Then the Onkyo does the crossing over, the room correction, and where is the DAC performed?

I asked the miniDSP folks about suggestions involving Dirac Live and crossovers. One option they suggested was Dirac at the server and their nanoDIGI 2x8 for a crossover in the digital domain, but also a second DAC dedicated to the subwoofers. The additional DAC I'm trying to digest...I guess any timing differences between the different DACs used would be compensated by the room correction...still not easy to digest. And I also wonder about additional jitter in the chain.

What are your thoughts on this configuration?

So what is the conclusion?

The thread starter was warned against a sub due to "the lack of coherence that inheres in adding a subwoofer". I think this is system and room dependent. Bass has to be integrated anyway. There is no general rule that it is better inside a (costly) speaker than with (less costly) speakers and a sub, even if it often works out that way. So there is nothing bad that "inheres" in choosing a sub.