more authoritative upper bass than most other satellites, giving you flexibility in mating it with various subwoofers. Thus, if you pair these two together in the Evolution systems approach, you have extraordinary flexibility, particularly in choosing the crossover frequency between the two. The M6 can reach down to 50 Hz, so you could select an upper frequency subwoofer rolloff as low as 50 Hz to match. Conversely, the W1 can reach up well beyond 100 Hz, so you could roll off the M6's low end response as high as the mid 100s to match.

Crossing Over the W1, Using the X1

      The X1 crossover encourages you to explore this flexibility of the W1, by giving you a continuous nominal choice of upper rolloff frequency (low pass filter) for the W1, from 50 Hz to 140 Hz. Incidentally, some vented subwoofers with a built-in power amplifier and crossover have a crossover frequency adjustment that goes as high, or even higher, but you can't actually use this upper part of the range as a crossover frequency, because you'll get horrible colorations from that spurious peak around 300 Hz from the vented subwoofer.
      If you don't invest in placing a W1 subwoofer at each M6 satellite location, then of course you should run your M6s as low as they'll go, in order to get the maximum surround ambience in that crucial upper bass and warmth region. In this case, you can set the X1 to roll off the W1 somewhere in the 50 Hz to 65 Hz range (wherever suits your room and your taste better), since the M6 running full range goes low enough to meet the W1 there. And you should probably run the M6s full range, without any intervening high pass filter (from the X1 or elsewhere).
      This tactic yields several advantages. First, you'll hear the best possible fidelity for most of the spectrum, since the signal fidelity won't be compromised by going through any high pass filter or ancillary circuitry enroute to the M6. Second, the second order transient response of the intrinsic M6 in the upper bass won't be degraded by adding any further filter poles in series at nearby frequencies. Third, the extra woofer excursion imparted by running the M6 full range will be absorbed by twin drivers, and won't contribute that much modulation distortion to the sensitive midranges, since the M6 itself crosses over away from its woofers at the remarkably low frequency of 350 Hz. Only if you play your system very loud, and hear audible distress from the M6 specifically on loud bass sounds, should you then add a high pass filter (essentially a subsonic energy protector) in series with the M6, instead of running it full range.
      If on the other hand you do invest in a W1 for each channel, placing a W1 at each M6 location, then you have a free choice among all frequencies provided by the X1, and you can simply pick the one that suits you best. If you choose a low woofer rolloff frequency around 50 Hz to 65 Hz, you can still try running the M6 full range, in order to hear the advantages noted above. If you choose a higher nominal crossover frequency, then you should use a high pass filter in series with the feed to the M6. You could use the X1's built in high pass filter (see sonic discussion below), which offers 3 fixed choices for the satellite feed: 50, 80, or 110 Hz (all at 12 dB per octave). Or, for optimum sonic fidelity, the hobbyists among you could simply insert the proper value capacitor to feed your chosen power amplifier that drives the M6s (this would be a 6 dB per octave high pass filter).
      We tried the full range of possible crossover frequencies between the W1 and M6, and the sound was great over the whole range of choices, thereby demonstrating and proving the generously overlapping capabilities of the W1 and M6 (the W1 can reach remarkably high and the M6 can reach quite low). So the flexibility of the W1 and M6, in being adaptable to different crossover frequencies, gives them the advantage of being remarkably adaptable to different rooms and different room placement of loudspeakers (and the X1's many controls provide even further advantages to this end, as discussed below).
      For example, if you have to place your satellites close to walls (not recommended for numerous reasons, including imaging degradation), you'll create a response dip in the warmth region (e.g. placement two feet from a wall creates a dip at 140 Hz), but you could use the W1's upper frequency capability to compensate by supplying extra energy in the warmth region (run the W1 up to 140 Hz, and overlap the satellite's spectral coverage), whereas you cannot run most vented subwoofers this high without running afoul of their obnoxious coloration peak around 300 Hz.
      Secure in the knowledge that the W1 can reach high in frequency without introducing spurious coloration, you can freely experiment with crossover frequency choice, for both the W1 (via the X1's front panel knob) and for your satellite (via the X1's rear panel switch, if you use the X1 for your satellite feed), as well as experimenting with the X1's other bass controls (see below), to find the crossover frequencies that work best for your loudspeakers, your taste, your room, and your loudspeaker placement. You might well find that your room modes, and the location of the various W1 and M6 (or other satellite) modules with respect to these room modes, make a particular choice of crossover frequency best for your situation.

Using the W1 as a Pedestal

      The tall B6 subwoofer is supposed to function as a pedestal base for the M6, placing the M6 at the correct listening height per NHT design intentions, which is with the listener's ear at tweeter height. But the B6 subwoofer does not sound nearly as good as the W1. Meanwhile, the nearly cubical W1, which provides very good bass, is short and squat (and that is in large part why its bass is superior). But the W1, being so short and squat, is not nearly tall enough to support the M6 such that the M6's tweeter is at the ear height of a seated listener.
      Thus, if you want to use the W1 with an M6, ideally at each M6 location, what are you supposed to do? Must you buy a separate stand for the M6, and park the W1 as close as you can? Seemingly, this is an unfortunate turn of events, putting you in a difficult quandary.
      Or is it? Once again with the Evolution system, misfortune serendipitously flips upside down into good fortune. It just so happens that the short, squat W1 is precisely the correct height to support the M6 such that the seated listener's ear height is exactly in line with the height of the M6's top woofer. And it just so happens that we found that we liked best the sound of the M6 when listened to along exactly the vertical axis of the top woofer height (and a horizontal axis that is 22 degrees off the midrange driver side, as discussed above). Voila! The W1 makes an absolutely perfect stand for hearing the M6 at its best (in our unorthodox suggested alignment).
      A slight further detail is worth mentioning here. The W1 is not exactly cubical, and it makes a slightly taller stand when the input terminals are at the rear. This is the perfect height for supporting the M6 if you are sitting in an upright chair or on a firm couch. On the other hand, the W1 makes a slightly shorter stand if you rotate it so that the input terminals are on the top side. This is the perfect height for supporting the M6 if you are sitting in a low slung easy chair or on a plush couch that you sink into.

Optimum Orientation for W1

      As you know, subwoofers radiate omnidirectionally, even when there is just one driver, so the W1 with its two oppositely facing drivers is surely ominidirectional, over its entire operating range. Thus, conventional wisdom would say that it doesn't sonically matter which way you point the W1, and it can't matter whether you rotate the W1 so that the woofer drivers are pointing straight ahead (and straight back) into the room, or are pointing straight at the side walls.
      Furthermore, bass wavelengths are very long at these low subwoofer frequencies (a 100 Hz wavelength is roughly 10 feet long, and a 25 Hz wavelength is about 40 feet long). Thus, conventional wisdom would say that temporal alignment on the order of inches or 1 foot doesn't make any sonic difference with such long wavelengths, and it can't matter whether you rotate the W1 so the woofers are facing you (and away from you), or are facing off to the side.
      But conventional wisdom is wrong. Very wrong.
      One's natural inclination might be to orient the W1 so that you are listening to one of its woofers pretty much on axis. After all, this intuitively seems as though it would give you the best bass impact, since one of the large drivers is then pumping air straight at you. It also makes cosmetic or aesthetic sense, since you want to see all the drivers you paid for in action, and you listen to all other drivers in loudspeakers pretty much on axis. And, since bass this low is omnidirectional, it shouldn't matter anyway which way the W1 is pointing, so you might as well have a woofer pointing at you.
      Don't do this.
      The owner's manual does not provide specific instructions on W1 orientation, so we experimented with all possible orientations of the W1. We found huge sonic differences. The intuitively obvious orientation, with one woofer pointing straight ahead or straight at you, is by far the worst sounding. This orientation cruelly degrades all aspects of the W1's very good bass, making it sound much like the B6: woolly, flabby, with mediocre impact and mediocre definition.
      What's the best sounding orientation for the W1? Have the woofers pointing sideways, so that one of the blank end panels is facing you, and toe in each W1 so that this blank end panel points straight at you the listener. Note that the M6 atop each W1 will not match the W1's horizontal orientation, since you will be directly on axis of the W1 cabinet, but you should be 20 degrees (to the center of your head) off axis for the M6, inboard of the axis of the M6.
      If you are aesthetically bothered by the subwoofer cabinet pedestal and M6 cabinet being oriented in different directions, it is permissible to rotate the W1 outward until its orientation matches the M6, so that you will be the same 20 degrees off axis from the W1 cabinet's front blank panel as you are from the M6 cabinet. This degrades the W1's bass quality only slightly from the ideal. And it does provide the sonic advantage of imposing less diffraction on the M6 output, since now the front bottom edge of the M6 cabinet can be perfectly lined up and flush with the front top edge of the supporting W1 cabinet.
      We also found that the bass of the W1 improved slightly when the sideways firing drivers were placed as close to the floor as possible. This lessens interference from the floor bounce, which acts as a second bass source. You can put the woofer 2.75 inches closer to the floor by flipping the W1 cabinet so that its input terminals are on the top side (and of course toward the rear, to make room for supporting the M6 toward the front of the W1's top). This flipping of the W1 also lowers its height as a stand for the M6, putting the M6 at the correct height for listening from a low slung easy chair or plush couch. In other words, if you want to hear the very best bass from the W1, you should discard your upright listening chair or firm couch, and get a low slung easy chair or plush couch instead, so you ears will be lower, at the correct height for hearing the M6 atop the W1 in its lower height orientation.
      Incidentally, you can also opt to place the M6s atop conventional loudspeaker stands, which match the height of the W1 cabinet, and then place your W1 subwoofers elsewhere in your room. This tactic would obey the general maxim that you should locate your main loudspeakers or satellites to provide the best spatial imaging, and then separately locate your subwoofers to provide the best bass for your particular room dimensions. Whole treatises can be (and have been) written about optimum placement of subwoofers, to best deal with various problems of listening room acoustics, and this is not the place for another such treatise. Suffice it to say that the experts disagree with one another, so this topic is alive with controversy.
      Now, what could be the explanation for the W1's orientation being so critical, even by inches, and making such a huge sonic difference to bass quality? As noted, the bass wavelengths are so long that a few inches one way or the other should not make any sonic difference. Furthermore, the human ear/brain is notoriously supposed to be very insensitive to (and therefore tolerant of) phase and timing differences in the low bass. The explanation seems to be twofold.
      First, the ear/brain might be relatively insensitive to absolute phase of a single bass signal, but it seems to be very sensitive to differences or mismatches in relative phase when two (or more) sources of bass are involved. Second, the ear/brain functions as a very sensitive time domain waveform analyzer, with time slicing capability. Thus, if we change the relative position of two bass drivers (or a subwoofer and a satellite, or even a single bass driver and its floor reflection), then we are actually changing the bass time domain waveform slightly, and the ear/brain's sensitivity as a time domain waveform analyzer easily detects this difference, interpreting this waveform difference as representing better or worse quality bass (in various ways).
      In our research we had previously conducted experiments wherein we played with the relative alignment of two bass sources, to discover just how sensitive the ear/brain was to hearing time or phase misalignment, and also to find out by measurement what if anything was changing in the time domain bass waveform that was causing the audible change in bass quality. In this research experiment, the two bass sources were a subwoofer and a satellite, both contributing to the same bass note, since they were overlapping at their crossover. We simply moved the subwoofer forward and backward, relative to the satellite, to discover how much physical misalignment might be required to produce an audible difference in bass quality. To our utter astonishment, we found that the bass quality, from these two joint bass sources, was audibly degraded if the subwoofer was moved away from the optimum sounding alignment by merely 1/2 inch. And the ear/brain had this high sensitivity, to alignment within 1/2 inch, for a wavelength that was about 10 feet long!
      How did the bass sound change, when the subwoofer was merely 1/2 inch off its best sounding alignment? The bass became less tight and dry, with less impact, and there was worse booming overhang (on the verge of being a lingering one-note boom, but not quite).
      Would measurements show any change in the combined bass output of the subwoofer and satellite, to prove that the change we heard was real? Yes, indeed. The bass step transient response measurement (a time domain measurement) showed that the system's undesirable negative overshoot, occurring after the correct initial positive spike, became considerably worse in amplitude (in the negative direction), when the subwoofer was moved to be just 1/2 inch away from what we had heard as the sonically optimum alignment with the satellite.
      This measurement not only proved that what we had heard was real, but the nature of the measured change also specifically corroborated the nature of the sonic change we had heard in bass quality. The perceived loss of impact correlated with the measured fact that the increase in negative polarity energy, occurring so soon after the correct positive spike that provides the sense of initial bass impact, effectively cancelled some of that perceived impact, since negative polarity energy effectively cancels out positive polarity (when the two are in immediate temporal proximity, as these were). The worsening increase of undesirable boom and overhang we heard correlated with the measured fact that the lingering, delayed energy (the overhanging boom) was now greater in magnitude.
      This correlation shows how the ear/brain works as a sensitive time domain waveform analyzer, hearing that there is more overhanging energy at a later time than the initial bass impact transient, when the subwoofer is misaligned even slightly. Our perception of an incipient one-note boom correlated with the measured fact that the positive spike, followed by the increased amplitude negative spike, effectively created a waveform much like a sine wave cycle with a specific periodicity, and the ear/brain naturally interprets this periodicity as having a pitch, i.e. as being a lingering boom at one note, one frequency, the frequency corresponding to the periodicity determined by the timing of the negative spike following the positive spike.
      That, incidentally, is precisely why vented bass systems have especially bad sonic problems with one-note booms. Their frequency response plot might look flat, but they sound as if they are putting out a huge amount of extra (and lingering) energy at just one frequency (sounding as though they did have a huge peak in frequency response at that one note). And they put out just this one bass note, regardless of how musical bass transients (e.g. a plucked bass) might be playing up and down the scale.
      The reason they sound this way is that their actual output, in response to any bass transient, overshoots and continues to ring with a certain fixed periodicity that's repeated for every bass transient, a periodicity that is pre-determined by the parameters of the driver and port design (including the sharpness of the system's corner frequency and the ringing from the port's high Q peak). You hear this fixed periodic ringing, and your ear/brain analyzes the time domain waveform, hearing the periodicity of the waveform's ongoing ringing pattern as though it were a single strong sine wave note, with excessive and lingering energy at the frequency corresponding to the periodicity of that ringing.
      Thus, you hear excessive energy at that one frequency, that one note, and you hear that one note of apparently excess energy repeating over and over on every bass transient, regardless of the actual changing musical notes that the music is playing. Sealed bass systems like the W1 can have less of this problem, since they can have less overshoot and ringing (especially if their corner is gentle, and their rolloff slope is gentle, which is why we advocate that the X1 have an option for defeating its subsonic filter, thereby providing a gentler corner at 27 Hz and a gentler rolloff below 27 Hz for the W1).
      How does this research experiment relate to the optimum orientation of the W1? The experiment showed that, with two bass sources, alignment between them was critical to within 1/2 inch for sonically optimum bass, and that misalignment beyond 1/2 inch produced worse boomy overhang and worse (reduced) bass impact. These are precisely the sonic degradations we heard when we oriented the W1 in the intuitively obvious position, so that one bass driver faced the

(Continued on page 114)