The Fifth Element

Spendor's US importer Mike Grubb let me borrow the pair of Spendor's next speaker up, the S8e, that Sam Tellig had just finished writing up for the June 2005 "Sam's Space" (Vol.28 No.5). Like the S6e, the S8e is a floorstanding two-way, but its cabinet is an inch or two larger in all dimensions than that of the S6e. The S8e has an 8" woofer instead of the S6e's 6.5-incher. The S8e's suggested retail price is $2999/pair.

Both speakers are sturdily built and have high-quality biwire terminals, cleverly arranged so that it's less likely that the jumper plates will be mis-installed so as to short out your amplifier. One welcome feature is that all of the wood finishes (black, cherry, oak, and rosenut) cost the same.

I unpacked the S8es, then first listened to some brief tracks on the S6es: John Atkinson's solo electric-bass channel-ID tracks and relative phase test, found on Stereophile's Test CD 2 and Editor's Choice. (I run those tests every time I change speaker or amp connections, as cheap insurance against making stupid mistakes.) Next, I made the changeover, played the same tracks on the S8es, and marveled at the difference. Then I took up pad and pencil.

Peter Walker once claimed that all of audio engineering was just Ohm's Law and common sense. I'm not sure I'd go that far, but I do think that a homeopathic dose of mathematics cures a lot of audio confusion. What follows is admittedly a simplification, in that I calculated the "frontal area" of the drivers; ie, the area of the plane described by the woofer cone's surround. One could also calculate the surface area of the inside of the cone or the volume of the cone's interior, but I'm not sure that those calculations would tell us any more than comparing the frontal areas.

Using our old friend Area = ϖ r², first we plug in 3.14 for ϖ. A 6.5" woofer has a radius of 3.25". Squaring that yields 10.56in². Times ϖ equals 33.18in². So far, so good.

An 8" woofer has a radius of 4"; that squared is 16. Times ϖ equals 50.24in². Now it gets interesting! 50.24in² happens to be 151% of 33.18in². In other words, the 8" woofer, despite being only 1.5" wider than the 6.5" woofer, is actually half again as "big": 50% larger.

Now, if all other things were equal, one would expect the S8e to have 50% more bass extension than the S6e. Due to the logarithmic nature of sound, 50% more bass extension would be half an octave; ie, four notes. However, all other things are not equal. Chief among these inequalities is that the S8e's cabinet is not scaled up strictly proportionally to its larger woofer; its cabinet volume is about 32% larger than the S6e's, not 50% larger.

Spendor's own specifications show the S6e with a –6dB point of 36Hz and the S8e with a –6dB point of 32Hz—that's more like one note than four (from D down to C). But it could be that the difference between the two bass-response curves constitutes the "tipping point" between a speaker that is ever-so-slightly frustrating for its lack of low bass and one that has "enough" low bass.

I think that for the $600 price difference between Spendor's S6e and S8e, the S8e actually represents better value for money. Your mileage may vary. Spendor also makes the S9e, a three-way with a –6dB point of –28Hz and a list price of $4999/pair, which just goes to show that those last few Hz of bass extension get increasingly more expensive.

Now, if Spendor's S8e is a good enough speaker for our hypothetical music teacher, why would anyone ever spend more? Good question. An 8" two-way speaker has some things going in its favor and some things stacked against it.

On the plus side, the crossover can be simpler and more transparent. Integrating two drivers in that particular (and in many others) is an easier job than integrating three drivers. The minus side is that an 8" driver is slightly larger than optimum for handling the midrange and upper midrange. Even if the cone's construction is sufficiently sophisticated to avoid breakup modes, as the frequencies being reproduced rise they will become increasingly directional, which has to be taken into account in the speaker's voicing.

Going beyond those points, a cost-no-object speaker that covers the same range—for example Wilson Benesch's A.C.T. (Circa $15,000/pair, depending on finish), which also has a claimed –6dB point of 32Hz—is going to do a better job of getting out of the way of the music and creating the illusion of thereness we all crave. Compared to my vivid recollection of the A.C.T., the S8e sounded slightly veiled.

Playing music with some deep bass content moderately loudly on the S8e, I could feel the top of the cabinet vibrating—not grossly, but not imperceptibly, either. The A.C.T.'s hugely expensive high-tech design and construction seem to banish cabinet resonances—it seems to let the music emerge from a quieter space. The A.C.T. is also extraordinarily coherent; the join between midrange and tweeter is not audible, as far as I can hear. And so on and so forth. But remember: the A.C.T. costs five times as much as the S6e.

So, despite its technical limitations, an 8" two-way can still provide a very cost-effective, perhaps even optimal speaker solution for a music lover. My own experience seems to bear this out. My first "high-end" speaker (1977) was Bud Fried's Q/2, an 8" two-way with a soft-dome tweeter and a primitive transmission-line substitute for cabinet loading. (My first high-end equipment epiphany was the extraordinary change effected by getting the Q/2s off milk crates and onto Fried's dedicated tilt-back stands, which, by raking the speakers back about 20°, got the midrange out of the shag rug and involved the ceiling in dispersing the treble.)

I later bought Spendor's SP1s, which were 8" two-ways with the addition of a supertweeter. A pair of used SP1s in good condition would make a fine start on a budget system, by the way. Spendor's US importer, QS&D, reports that they can still get or recone all the drivers for SP1s.

Not only on The Complete Norman Granz Jam Sessions but also on Telarc's recording of Brahms' Ein deutsches Requiem, with Robert Shaw and the Atlanta Symphony, the MD-208/S8e combination acquitted itself very well indeed. The Magnum Dynalab receiver ran out of drive before the speakers ran out of dynamics, which is good. Although it indicates that spending more money for a more powerful amp would not be a waste, on a day-to-day basis it probably would be a case of steeply diminishing returns.

As far as a source goes for our hypothetical music teacher's system (keeping in mind all the free music a good FM tuner provides), that can be any decent digital source, from Sony's DVP-NS975V SACD player ($300) to Marantz' SA-8260 ($1100).

At this point in the evolution of the system I make a bold claim: Without doubt, our music lover will be better off getting his or her interconnect and speaker cables on the cheap from a professional and broadcast source such as Markertek, and spending comparatively serious money on acoustical treatments, than popping for audiophile-approved name-brand cables and neglecting room acoustics. (Markertek is a large stocking dealer for just about everything made by Belden, Canare, and Neutrik. Their custom-fabrication services are excellent. In-stock 3' RCA–RCA interconnects run from $13 to $30/pair, and speaker cables are priced correspondingly. Markertek's catalog is a great "dream book," in any case.)

Whether or not to spend the money for an entry-level Rives Audio consultation (see "The Fifth Element," August 2005, Vol.28 No.8) will depend on how problematic the room's bass response is. If the bass is not bad, it would be perfectly reasonable simply to emplace RPG's Binary Amplitude Diffusor (BAD) panels, or improvised or home-made equivalents, on the room's first-reflection points. Those points can be located by using the simple technique described in "The Fifth Element," May 2001, Vol.24 No.5.

If a room is not burdened with too many hard, reflective surfaces, a dozen BAD panels—four each on the front and sidewall first-reflection points—should make for wonderful improvements in focus and articulateness. The retail price of a dozen 2' by 2' BAD panels will be about $800 or more, depending on the fabric and other options selected. Half a dozen RPG Skylines (ca $1000) on the ceiling, in my experience-based opinion, should make for more of a positive contribution to your room's sound than any expensive cables I have ever heard.

When it came time to select wire goods, I sought the counsel of Vince Galbo, former US importer of Plinius (footnote 2). Vince's immediate suggestion was that I try Analysis Plus's Oval 9 speaker cables ($439/8' pair) and their corresponding Copper Oval interconnects ($219/0.5m, $296/m; ). He strongly believes them to be the price-performance leaders. Vince's own listening biases run to higher-powered solid-state amplifiers and moderately efficient speakers with long-throw voice coils; Harbeth's Super HL5 is a particular favorite of his. So that Vince likes a speaker cable with "a lot of metal in it" comes as no great shock.

The particular Analysis Plus products I received here have, over the past few years, fared very well among the critics, including Michael Fremer (January 2001, Vol.24 No.1). Compared to the inexpensive cables I'd been using, the Analysis Plus wire goods had a cleaner, crisper sound, with more treble extension—a positive synergy with both the MD-208 and the S8es.

The bottom line: Yes, you can spend a lot more money. But a lot of music lovers never will, and it is more important for the survival of high-end audio that we make the tent bigger rather than more exclusive (read: snobby). Please do all you can to get the word out: Good audio equipment is not just for audiophiles anymore!

Paeans of praise

---

Footnote 2: His change of status must now mean that Vince is Pliny the Former the Younger, Pliny the Former the Elder being Victor Goldstein. Sorry, I couldn't resist.