Floor Loudspeaker Reviews

The original plan was for me to review Magnepan's MG2.5/R ($1550/pair) as part of the group review elsewhere in this issue. After some preliminary listening chez Larry Archibald, however, J. Gordon Holt persuaded me that he would be a better choice as reviewer—as Founder and Chief Tester, he does have some powerful arguments—and his review appears elsewhere in this issue. Nevertheless, I believe I cover some ground neglected by Gordon.

To keep the description short, the 2.5 is a cut-down, two-way sibling of Magnepan's successful three-way MGIIIa, and the same size as the MGIIc. It uses a rather shorter version of the III's ribbon tweeter, 42" long, with the edge clamped every 1.5" on opposite sides, crossing over below 1kHz to a "Magneplanar" driver, this consisting of wire conductors—in effect a flat coil—bonded to a 480in² Mylar diaphragm, positioned behind an array of bar magnets to give single-ended drive. A simple crossover with first-order 6dB/octave slopes is used.

The ribbon is protected by a 2.5A fuse but is more delicate than a typical dome drive-unit. (Using the broken-glass track on the Japan Audio Society Test CD, LA blew a ribbon and two fuses during his auditioning of the MG2.5/R.) It has also been pointed out by Martin Colloms that the mechanical stress endured by an aluminum-ribbon driver could result in a limited lifetime, maybe four or five years depending on the type of music played and the usual levels adopted. In any case, replacement is an easy matter, and Magnepan has an excellent reputation for customer support. (But don't expect them to be too sympathetic if you replace the fuses with lengths of copper rod in order to use 2.5s driven by 2kW amplifiers to provide the disco sound at your daughter's outdoor wedding party!)

Both drivers are mounted in/supported by a baffle, but there is no cabinet as such: the radiation pattern is that of a dipole, and the low-frequency extension is governed by the baffle size. When the wavelength of the sound is equal to the baffle dimensions, cancellation between the direct sound and the backwave will roll off the low bass in a gentle, aperiodic (nonresonant) manner.

Connection is via 4mm sockets on the speaker's rear, a second set being linked by a jumper which can be replaced by a low-value power resistor to reduce the tweeter level if necessary (see below). Unlike earlier Magneplanars, the feet offer good support and extend both in front of and behind the speaker.

The sound
When I lived in England, I had considerable experience with both Magnepan's MGIIIa and SMGa. My listening room there was some 27' long by 13' wide, and it proved impossible to get the MGIIIs to sound as good in that room as I had heard them in, for example, Martin Colloms's listening room, where they gave an astonishingly accurate and musical sound.

Placing them along my long wall, where they would have plenty of room to "breathe," proved out of the question, due to the disposition of doors, alcoves, and furniture, and to the fact that when the speakers were far enough away from the rear wall, the listening position was then up against the opposite wall, which degraded the imaging. Placing the speakers across the short wall solved both those problems, but I couldn't then get the speakers far enough away from the side walls, reflections from which both affected the soundstaging and modified the low-frequency response overmuch. In that particular room, the tiny SMGas actually gave a more balanced sound, though lacking the insightful character, transparency, dynamic range, and HF/LF extension of their more expensive siblings. A frustrating experience.

My current room (see the floor plan in Vol.10 No.1) is ostensibly a more conducive environment for big dipole speakers, as it enables me to place them along the long axis of the room and well away from both side and rear walls without compromising the need to get the listening position away from the wall behind it (although I have found that rugs hung immediately behind the seat do improve the stereo focus and sense of depth).

I started off with the speakers having the tweeters on the inside edges, placed some four feet away from the rear walls, ten feet from the side walls, and six feet apart, firing straight ahead. The upper midrange's lack of body indicated that some toe-in would be essential, but the exact degree proved quite critical. Factors to be balanced against each other were the focus of the imaging, the degree of exaggeration of the lower midrange, the amount of HF "air," and compensating for the loss of body in the upper midrange/lower treble. After an evening's experimentation, the placement solving all these interrelated matters was with the speakers eight feet apart, three feet from the rear wall (which is angled so as to fall back on the outside of the speakers), and eight feet from the side walls, with the tweeters on the outside edges. The toe-in was such that the listener was sitting just to the inside of the tweeter axis.

If you think that this indicates that I am extremely persnickety, you are right! The Maggies, being dipoles and having laterally disposed drive-units, have a complicated interaction with the room acoustics; even quite small changes in position can improve or degrade the basic sound character. It well worth making the effort, however, as you will hear.

Once I was satisfied that the 2.5s were on song, I set down to some serious listening. On went the HFN/RR Test CD, track 11, Anna-Maria Stanczyk playing Chopin's Waltz in C-sharp minor. Well, the sound was definitely that of a Magneplanar: there was something about the lower midrange, call it "Maggie slam," or whatever, but there was a slight thickening of the sound in the octave centered on middle-C. This aspect of the sound seemed most affected by the changes in position and toe-in. Apart from that, the speakers were invisible, the midrange being almost totally uncolored. Instrumental tone colors were superbly delineated across the band, whether it was rock recordings where bass instruments such as the kick drum and bass guitar had similar midrange overtone structures, or the contrast between similarly voiced instruments, oboe playing in its low register vs English horn playing high, for example. Voice, however, was where the 2.5s scored the highest points, soloists hanging in the air between the speakers with naturally sized images and a total lack of nasality, chestiness, or sibilant emphasis. They just sounded real!

Above the midrange, it proved easy to get an insufficiency of energy in the region where the Magneplanar driver crosses over to the ribbon. This aspect is affected by the toe-in and care should be taken in this regard. Listening height is also critical: Your chair should place your ear at least 36" off the floor, else you will be off the vertical axis of the 2.5's tweeter, resulting in a loss of highs. I could not totally integrate the two drivers, the tweeter overall sounding a little shelved down in level. On the scale passages in the Chopin Waltz mentioned earlier, the highest notes were always a little quieter than the others. The upper two strings of the violin also sounded a little recessed compared with the lower two, and some listeners may be put off by this laidback character. However, apart from that idiosyncrasy and a slight mid-treble emphasis (audible on pink noise) directly on the tweeter axis, the high frequencies just soared evenly and smoothly upward. The MG2.5/R is a true high-end speaker in this region.

On the other side of the midrange, bass frequencies were clean though not that extended, covering the fundamental region of the double bass, which had good weight, but lacking power on kick drum and the lowest organ-pedal notes. Dynamics were also good. After playing a number of rock recordings, I felt that this was the ideal speaker for those who like rock and classical, due to the lack of strain to the sound. Drums came over well, particularly at high replay levels, the kit on John Hiatt's "Tip of My Tongue" (from Bring the Family, A&M SP5158) sounding well-defined both in space and in tone color. And Jerry Marotta's thunderclap tom-tom interjections on Peter Gabriel's hymn to prisoners of conscience, "Wallflower" from his 1982 album, knocked me backwards in my seat.

In fact, the 2.5 seemed to have almost unlimited subjective dynamics—until the Mark Levinson No.20s ran out of headroom and clipped, resulting in "squashed blat" sounds coming from the woofer/midrange panel and the tweeter shaking rather more than I would have liked. I mused whether a second high-pass filter should be used a couple of octaves down from the crossover to protect the tweeter from low-frequency garbage, but apparently Magnepan's Jim Winey prefers the transparency created by leaving the tweeter ribbon relatively unguarded (apart from the protection offered by the series fuse). I didn't actually damage anything by clipping the amp, so I am probably being overcautious here.

I wasn't happy with the 2.5s' soundstaging as originally set up in my room, the image lacking center-fill; ie, being localized too much in the speaker positions. But once the tweeters were moved to the outside edges, it was excellent, both regarding lateral definition and the ability to present an image with finely layered depth. The fact that the tom-toms are paradoxically farther away than the snare drum in the John Hiatt track was laid clear. The piano image on my Chopin recording was about as accurate as I have heard, both in the way the piano occupied the stage from the center to the right-hand speaker, and in that the piano was set back the correct distance behind the plane of the speakers. One odd factor, which JGH noted also, is that anti-phase signals, such as vertically modulated LP surface noise, were quite stably located outside the loudspeaker positions. This is presumably due to the time-coherent nature of the speaker—flat drivers in the same plane and what are effectively first-order crossover slopes.

Conclusion
From reading my listening impressions, it may be thought by some readers that the MG2.5/R is almost without flaw. Bearing in mind its price, this is pretty near to the truth. When optimally set up, it offers a musical performance head and shoulders above the other three loudspeakers in this report, and gives its purchasers much of what constitutes true high-end performance. However, it does have a signature—in particular, the slight lower-midrange thickening, the shelved-down treble, the relative lack of low-frequency extension, and the lack of energy in the crossover region to the tweeter—which may or may not be to the purchaser's taste. Coupled with its specific room needs, I would advise careful auditioning in your own room and system. Strongly recommended.—John Atkinson