Paul Barton: It Doesn't Get Much Better Than That Page 3

Barton: Because the first-arrival sound is predominant. Sound power compiles a lot of measurements to give you a picture of what the speaker's total energy output is. Picture the speaker in the middle of a sphere and microphone measurements being made all over the surface of that sphere. In a room, in varying degrees depending on how well the room is damped, some of the energy from all of the directions around the loudspeaker will eventually reach the listener at some point in time. The intensity of that will be very much determined by the surface of the room itself and the directivity of the speaker in that room. So it's not enough to just look at "frequency response." You have to tie it into sound power.

Atkinson: Floyd Toole's work showed that listeners preferred speakers that have a flat frequency response and a well-controlled off-axis dispersion. While some listeners might like loudspeakers that depart radically from either of those two paradigms, most people don't.

Barton: When we say "flat frequency response," it's not enough. We also should define the bandwidth of the system. Certainly, the wider the bandwidth, the more transparent the speaker becomes. It's my preference to keep the on-axis response flat without compromising the power response, which may not necessarily be flat—in fact, it has a tilt—but has a very smooth transition.

Atkinson: Many two-way speakers have an off-axis valley at the top of the woofer range, but an off-axis mountain at the bottom of the tweeter range.

Barton: Right. That's the challenge, to get the tweeter—whose directivity at its lowest frequencies is very good—low enough so that doesn't happen. And you don't want the midrange driver or woofer to be pushed [high enough in frequency] that it becomes very directional.

Atkinson: This is why we laughed when you said your first PSB design was a 12" two-way. You just really can't take a woofer that big much higher than 600 or 700Hz.

Barton: I would have had to cross that 1" tweeter over at about 500Hz to match up the woofer's and tweeter's dispersions. And then I would have had a power-handling problem! [laughs] The speaker would probably play with 1 watt for maybe an hour. And 5 watts for maybe a minute!

Directivity and frequency response are so important. Other things have to fall into line—like power handling, which relates to distortion, and power compression—all those characteristics. But one thing I've learned over the years is how to make good tradeoffs when I don't have the freedom to just throw anything that's needed at the speaker because price is of no consequence, or less of a consequence. As you get into smaller and smaller budgets on a loudspeaker, that's where the challenge is, in making the right tradeoffs.

I'll give you an example. If you're sitting in a typical room, listening to a pair of loudspeakers, the first sound you hear comes directly from the loudspeaker. What is the second sound you hear? In a typical room?

Atkinson: The reflection from the floor?

Barton: Typically the floor has some treatment, so it's not as good a reflector, and because the tweeter and woofer aren't coincident in traditional two-way systems, you tend to get a lot of interference when you're that far off-axis. The path lengths are so different, the reflection is not as aggressive. The most aggressive second sound you hear, in my experience, is off the side wall, both because it's normally fairly reflective and because, with a vertically oriented speaker, the horizontal dispersion is usually pretty good over a wide frequency range.

If a loudspeaker's response 60 to 75 degrees off-axis looks good, but 30 degrees is terrible, I'm not as concerned as I would be if 30 degrees was nice and 60 to 75 degrees was terrible. Because the coloration of the off-axis sound comes soon enough after the first sound that you tend to integrate the two. And it's the sum of the first and second arrivals that gives you the sonic signature. Not just the first sound. That's something I pay an awful lot of attention to.

Atkinson: I'm reminded of a loudspeaker some years ago that measured flat on-axis. In a blind listening test I was involved in, the listeners all marked the speaker down as being horribly bright. If you looked at the off-axis response, the speaker featured big "horns" around 2 or 3kHz. Exactly as you say, the reflections of the sound from the side walls were as bright as all get-out. Even though the first-arrival sound was flat, what we were hearing was the integration of the first neutral sound and the second, much brighter reflected sound. And that's what we all marked down on our score sheets.