Even since Harman International published Subwoofers: Optimal Number and Locations, home theater enthusiasts and speaker manufacturers have accepted the idea that using four subwoofers--either in the corners of the room or centered on each wall--provides the flattest possible frequency response across a large seating area. It's also generally accepted that two subwoofers put out six decibels more bass than a single sub, and four subwoofers put out an additional six dB on top of that. I had a chance to test this assumption when I recently reviewed the OEM Systems ICBM (Integrated Custom Bass Management) subwoofer system, a package of four subs with a purpose-built subwoofer amp. What I found contradicts what many audio enthusiasts and manufacturers believe about subwoofers.

Before we get into the testing I did, there are two key things to understand. First, if you put two subs side by side in an anechoic chamber, you get an increase in output of almost exactly six dB. I've confirmed this in my outdoor CEA-2010 output measurements, where placing two subs side by side usually gives me an extra six dB of output (plus or minus about one dB). If the two subs delivered perfectly matched performance and could occupy the same physical space, the result would be exactly six dB. This is what leads to the general assumption of a six-dB increase in output when you double your subwoofing capability.

Second, the reason four subs deliver flatter response is that some of the subs cancel out the other subs' response peaks and dips caused by room acoustics. Place a single subwoofer in a room, and it'll have huge peaks at some frequencies and huge dips at others. The frequency and magnitude of the peaks and dips vary tremendously depending on where you're sitting. Add another sub somewhere else in the room, and because of its different position, it will interact differently with your room's acoustics. It'll have peaks and dips at different frequencies, and thus will partially cancel the first sub's response errors. Add two more subs, and the response should get even smoother.

However, if you think about it, you might get the idea that some of the energy from those extra subs is spent cancelling out the first sub's errors instead of increasing the total output of the system. Several manufacturers I know have speculated about this, but to my knowledge none have actually tested it. With the ICBM system's four SE-80SWf eight-inch subs, mounted in their ENC-816LP enclosures, I finally had a chance to see exactly how much using four subwoofers in the corners of a room adds to bass output.

I powered all the subs with my ATI AT2007 multichannel amp instead of the P-500XB amplifier that OEM Systems provides. My measurements showed that the P-500XB has plenty of power to drive the subs to their limits within their operating band, but the amp relies on bridging and various parallel/series connections to drive four subs; so, just to make absolutely sure each sub would get plenty of power under all conditions, I switched to the ATI, which weighs more than 100 pounds and has so much energy storage in its power supply that it will run for about 30 seconds even after being disconnected from AC power.

First, I wanted to see the effects that the four subs had on frequency response. So I measured the in-room response of one sub in my room's left front corner; then, one sub in each of my room's front corners; and finally, one sub in each of my room's four corners. One set of measurements (the first chart) was taken from a position next to my usual listening seat. Another set (the second chart) was taken a few feet back and to the side, in a spot where bass response tends to be uneven. To calibrate the levels, I ran a pink-noise test signal and adjusted the level so that each system gave me an average sound pressure level of 75 dB.

As you can see, the response in seating position one is flatter with two subs and even flatter with four. In position two, the response is clearly flatter going from one sub to two, but going from two subs to four doesn't produce a clear improvement. Still, I was definitely getting flatter response overall than I would with the single sub.

Next, I measured the maximum output of the subwoofer system using CEA-2010 technique, but I do so in-room so that the effects of room acoustics would be included in the measurement. I measured the same way I did the frequency response: with one sub, two subs and four subs, and in the same two positions in the room. I've placed the results on a couple of spreadsheet charts. Each one also includes lines that show what the output would be if the added subs achieved the theoretical increases: six dB for two subs, 12 dB for four subs.

These charts show that, while I did get an overall increase in output by adding subs, the results varied a lot from frequency to frequency and position to position.

For me, the most surprising part is that at 20 Hz--the toughest frequency for the subs and the amp to handle--the output increased almost exactly six dB in both measurement positions when I went from one sub to two, and another six dB when I went from two subs to four. Yet at 63 and 80 Hz, increasing the number of subs had little or no effect on output. In fact, at 63 Hz in the second measurement position, using four subs gave me the least output.

At other frequencies, it's a mixed bag. Occasionally the increase in output was slightly greater than predicted. Sometimes there was barely any increase. Here's how the results averaged:

Going from one sub to two would, in an anechoic environment, would give me six dB more output. However, in my room, I got 4.3 dB more output in measurement position one and 1.9 dB more in position two.

Going from two subs to four would, in an anechoic environment, give me 12 dB more output. However, in my room, I got 9.7 dB more in measurement position one and 6.5 dB more in position two.

Of course, these results are just from one room, and only two measurement positions in that room. Your results in your room will be different because your room's acoustics aren't the same as mine. It's also worth stressing that these results are true only for systems with one sub in each corner. If you want more output from four subs, you could put, say, two subs each in the front corners of your room, stacked or side by side, instead of putting one sub in each corner.

But my experiment did allow me to come to one general conclusion: Adding more subs in different locations is likely to increase the average bass output of a system significantly, but not by the theoretical six dB with two subs and 12 dB with four subs. And depending on where you're sitting and what you're listening to, you might not hear any improvement at all.

Additional Resources
• How DSP Can Take Audio to New Heights in 2016 at HomeTheaterReview.com.
• How to Choose a Subwoofer for Surround Sound or Stereo at HomeTheaterReview.com.
• Why Do Audiophiles Fear ABX Testing? at HomeTheaterReview.com.