Floor Loudspeaker Reviews

Incidentally, the 1-As are very sensitive, which is a mixed blessing. While this does allow them to produce prodigious sound levels without taxing the amplifier or endangering the speakers, it also increases the likelihood that any hum originating between the preamp volume control and the power amp outputs will be raised to audibility.

Sound Quality
So, how well does the Soundfield 1A do what it's supposed to do?

Extraordinarily well. I found, in fact, that I could stand outside the lateral location of either speaker and still hear a balanced soundstage. In short, this is the least antisocial speaker system I have heard. (The current Ohm speakers, which I hadn't yet heard at time of writing, are increasingly directional at frequencies above 1kHz, and direct their "beams" forward and inward by 45° to achieve somewhat the same result as the 1-As. I doubt that this would provide as wide a listening area as does the Soundfield system, but will soon know; we have a pair of the Ohm 5s on hand. See Dick Olsher's review.)

Unfortunately, the 1-A can't work miracles. It is still subject to the rule that the wider the listening area, the vaguer the imaging. Although the system's imaging from across the entire listening area is much more specific and stable than is usually obtained from toed-in directional systems, there is no listening position from which the imaging is as specific as can be obtained from highly directional systems which produce a "sweet spot" just an inch or so wide. And a lot of perfectionists will think twice before compromising pinpoint imaging from that narrow sweet spot in order to get less specific imaging over a wider listening area. The fact that pinpoint imaging rarely exists in real life (footnote 3) will not dissuade those who demand it from their audio systems.

It is almost pointless to try to describe the 1-A's imaging and soundstaging characteristics, because these qualities are so very susceptible to speaker placement, and the 1-As allow such flexibility of same. You can get almost any soundstage presentation you desire, just by judicious placement. But the best I could do in terms of center bunching (image tightness from mono sources) was only very good. It's not that the system made instruments sound larger than life; just that their normal-width images were not as specific from most recordings as I have heard from some other speakers. (On the other hand, a test tape we had previously made, to illustrate the directional accuracy of various stereo mike techniques, showed that the 1-A does have superlative locational capabilities, although, in JA's opinion, not to the standard shown by the Sound Lab A-3s.)

Off-center stability of the 1-A is positively amazing! From anywhere across the listening area, everything remains in exactly the same spot. Even when listening from the left side of the left-hand speaker, centered instruments are midway between the speakers, (although no longer between the speakers and the rear wall.) And it almost goes without saying that the Soundfield 1-A can image way beyond the speakers with appropriate program material. It did it with some of the material on our test tape, as well as with a number of commercial recordings containing the necessary phase relationships.

An obvious application for these speakers is for large-screen front-projection video systems, where it is important that film dialog remain centered on-screen for listeners who are seated to the sides of the projection axis—most film dialog is mixed as A+B mono—and that stereo sounds "track" the visible on-screen positions of their sources. The 1-As work admirably in this capacity, eliminating the need for a front-center dialog speaker under the screen. As can be inferred from the above description, center imaging with the 1-A was not as specific as it can be with a dedicated center speaker for dialog, but this is not a liability for film sound because one can see where the source of the dialog is, without having to rely on aural cues alone. In fact, the slightly vague stereo image can be an asset here, because if the actor is seen speaking at one side of the screen, it is easier to believe his voice is also coming from that side when the dialog isn't pinpointed dead-center. The 1-As' high sensitivity easily allowed for movie-theater-style 105dB signal peaks with a relatively modest 100 watts of amplifier power. Their only drawback for video use is that they lack the midrange projection and punch necessary for proper reproduction of film soundtracks.

Surprisingly—considering the multiplicity of drivers—there is virtually no VVB effect when one moves one's head from side to side. It appears that dbx has achieved exactly what they set out to achieve: a speaker system which favors no part of the listening area.

The reproduction of depth by these speakers must be heard to be believed! Instruments are clearly ranged in rows from front to back, with a definite wall behind them, and have an almost palpable roundness to their sound instead of the flat, cardboard-cutout "depth" one often hears when perspectives are rendered this well. It's not that depth is exaggerated; just that it is so conspicuously audible. (Reproduced through these speakers, recordings which I made myself place the performers in almost exactly the same relationships to one another and to the stage boundaries as they were originally.) This kind of depth does not "come with the territory," though. The speakers must be well in front of the rear wall for this to happen; distances of less than four feet cause increasing compression of soundstage depth.

How about overall sound quality? The Soundfield 1-A is very good, ranking about on the borderline between our Class B and Class C recommended systems. Its only real problem is what sounds like an underdamped resonance (read "peak") at around 7kHz, which imparts a somewhat sizzly quality to cymbals and vocal sibilants and a consistent "tick" coloration to disc surface noise. (Subsequent measurements, with both steady and warbled sinewaves, indicated moderate, narrow peaks at 7 and 12kHz. The 7kHz one is by far the more noticeable.)

Low-end extension is awesome. I measured it as being essentially flat (–3dB) to 22Hz at the listening location, and that was with the LF EQ set to "flat." With (additional) LF boost, these can blow your socks off! This is one of the few systems of its size that really reproduce the shudder of a large pipe organ and the almost-subsonic ambience of a large performing space with nothing going on in it but outside noises from traffic, subways, and earth tremors. Actually, the extreme low-end range of these is not all that obvious, because they do not have the LF detail to call attention to the fact that they are reproducing astonishingly deep material. Bass detail and control are only fair, however, and though LF pitch definition is quite well reproduced, the speakers lack the "throb" that characterizes live bass (and that of a very few loudspeakers).

Except for that sizzle riding along at the top, the 1-A reproduced instrumental timbres quite well, although it is difficult to define its "sound" because the tilt adjustment on the controller allows the 1-A to produce a wide range of spectral balances, from dark, fat and dull, to lean and strident. I actually found that I used the tilt control quite often, because large-scale musical forces, such as orchestras and pipe organ, sounded more realistic with a richer and more laid-back sound than did small-group recordings, which did better with a leaner, more upfront sound. With the tilt control optimized for a specific recording, I found that the system did better with woodwinds and violas than with brasses and violins, the latter tending to be little wimpish. Making allowances for spectral balance, I would judge the 1-A to have rather good focus, with somewhat less veiling than is typical of dynamic drivers, but also a bit less transparency than I have heard from some dynamics (such as Thiels and Spicas), and markedly less than from full-range electrostatics. At all usable tilt settings, though, there was a consistent vague lack of body and punch to the lower midrange—here we go again!—which tended to make sounds a little lighter than they should be. (A cornet, for instance, sounded more like a trumpet, an alto voice more like a soprano.) Tubas, trombones, and piano bass strings tried valiantly to sound stentorian, but never quite made it.

Finally, a few placement caveats. While the 1-As can give a solid center image when placed much farther apart than other speakers, it is not suggested that you sit closer to them than 1.5 times the distance between them. Close listening does enhance detail in most rooms (you get more direct sound in relation to the room reflections), but too-close listening will cause too great an amplitude increase from the closer speaker, when you move to one side, for the polar-pattern shaping to compensate for; you'll get an image shift toward the closer speaker.

Conclusions
How, then, would I sum up this system? In absolute terms—of smoothness and detail—I would give it fairly high marks, but I cannot characterize the 1-A as an audio perfectionist-type system. Its level of coloration is not as low as it should be at this price level—there will be few audiophiles who will put up with that somewhat sizzly high end—and though it does some audiophile-type things, such as low-end range and soundstaging, extremely well, it does not provide the kind of pinpoint imaging demanded by perfectionists. The sizzle can be pretty much tamed by rotating both speakers so that their inward-firing tweeter beams are aimed a few degrees toward the rear wall, but this also tends to impair their ability to offset changes in lateral listening position, requires that they be used somewhat closer together in order to maintain adequate center fill, and kills some of the system's apparent detail. In short, the cure is arguably worse than the malady.

The dbx Soundfield 1-A neatly solves some of stereo's knottiest problems, but does so at the cost of less-than-impressive performance in a couple of areas where other comparably priced speaker systems are far superior. At the price, this system is an agglomeration of very mixed blessings.

---

Footnote 3: I am afraid that I will have to disagree with you about this, JGH. As I wrote in a 1981 article, human hearing is generally amazingly directionally precise in the lateral plane, depending on the frequency range of the source and the ratio between the direct sound and the reverberant field. The only areas of ambiguity concern vertical localization, which is to be expected from a system using two laterally spaced ears, and (to a lesser degree, because of the response shaping of the pinnae) front/back discrimination.—John Atkinson