Some comments have been made about the "speed" of subwoofers, and that's an interesting subject.
When people think of "speed", they think of good transient response, rapid rise time, quick decay, low group delay, and so forth. That seems logical, but it overlooks the way the ear processes sound at low frequencies - which in this case is somewhat counter-intuitive.
You see, the ear has very poor resolution in the time domain at low frequencies. In fact, the ear is unable to even detect the presence of less than one wavelength of a bass tone, and cannot determine the pitch of a bass tone from less than several wavelengths. The implication is that what's happening in the time domain - transient speed, group delay, etc. - is not readily detectable. And yet as most of us have experienced, subwoofers that have better transient response (such as low-Q sealed boxes) are clearly perceived as being "faster". So, is the science wrong?
No, but there is more to the story. Unlike in the time domain, the ear has very good resolution in the frequency response domain and can hear differences of less than 1 dB in the bass region. The types of enclosures that have good transient response (low group delay) also tend to have frequency response curves that slope gently downward with decreasing frequency. This gently downward-sloping characteristic is beneficial because it tends to offset typical room gain. On the other hand those enclosures that have relatively poor transient response tend to be "flat" anechoic down to a much lower frequency, such that once room gain (gain from boundary reinforcement) is factored in, the low-bass region is over-emphasized... and the result is perceived as "slow" bass. So the observation that speakers with good transient response sound "faster" is generally correct, but not for the reason we think!
Not long ago a paper was published in the Journal of the Audio Engineering Society that examined the audibility of group delay. They looked at group delay along the order of magnitude one might find in a typical vented loudspeaker system, and the tested its audibility in isolation (that is, the frequency response was equalized to be the same for the different trials so that the difference was limited to group delay). The data indicated that group delay was marginally detectable on test tones and undetectable on music. This is very interesting, because it confirms that transient speed, in and of itself, is not what determines the subjective "speed" of a subwoofer system.
So we are left with the in-room frequency response as that which best predicts the subjective perception of "speed" in the bass region. I am among those who really like "fast" bass, but I believe that it is most effectively pursued in the frequency response domain, rather than in the time domain.
The implication for the person embarking on a subwoofer quest is this: Analyze individual subwoofers with the expectation that room gain will gently boost the bottom end, and that if you have to err, err on the side of a sub that could use a bit more room boost rather than on the side of a sub that will have too much deep bass after room boost. Typical room gain is estimated at about 3 dB per octave from 100 Hz on down, so fudge that estimate based on your assessment of whether your room is "lean" or "bass-heavy".
I also think it's beneficial to take steps to smoothe the in-room bass and to that end advocate multiple small subs distributed around the room instead of a single ubersub (see Drew Eckert's first post above), but that's another subject.
Duke
dealer/manufacturer
