Slowly but surely, the benefits of speaker efficiency are gaining wider acceptance. You could argue that it’s a move that started with the advent of surround-sound and home theatre, with its demands for high-levels, wide dynamic range and assumption that there’d be a(n active) sub-woofer to handle the grunt work. Either way, the dynamic demands of challenging music programme suddenly got a shot in the arm, as hardened audiophiles, wed to flat frequency response and brutally reactive crossovers, suddenly realised what they’d been missing. Of course, a shift from an average speaker efficiency of 88dB to 91dB isn’t exactly seismic – at lest not for those touting horn-based systems, but as soon as a speaker creeps over the 90dB threshold things, musical things, start to happen. The further it ventures, the more obvious and significant those ‘things’ become, until once you start flirting with 93dB, or even 95dB, a whole new vista of performance opportunities emerges. And let’s be clear, by “performance opportunities” I’m referring to the musical performance recorded on the disc or file, and its reproduction via your system, in your home.
But all of this assumes that the amplifier is a fixed commodity. We’re all aware that more efficient speakers open the way to lower powered amplifiers, but we’re less clear on whether that is or could be a good thing. Of course, the horn brigade will point to flea-powered SETs as the ultimate musical solution, the logical end-point of this argument, but I’m not sure I agree. I’ve heard a lot of SET amps and I’ve heard a lot of them driving big horn systems. As a general approach it’s horribly flawed – not to mention tremendously impractical. I’ve heard great sound, genuinely astonishing sound, from SET amps and horns, but it’s rare: just as rare as similarly impressive results from other approaches and technologies. I’m not convinced that four-and-a-half Watts of NOS triode is a general solution. In fact, there are just as few, genuinely good triode amps as there are solid-state ones: just as few really good full-range horns as panels or box speakers. The virtues of this argument lie not at the extremes, but much closer to the centre ground. The real importance of the shift in efficiency from 88dB to 92dB has little to do with the speakers themselves – and everything to do with the fact that it makes the amplifier’s job easier! The true significance of the shift lies in the realisation that it’s easier to build efficiency into a speaker than it is power into an amplifier…
You might think, “So what?”, happy to take the win, regardless of where it’s sourced. But understanding the nature of the change is important, because in it lies a whole new world of opportunity. Just as increasing the efficiency of a speaker eases the load on the driving amp, so too does bi-amping, albeit by a rather different and (arguably) even more effective mechanism.
The root of all evil…
It’s not hard to understand why so many people think that passive crossovers are the source of much that’s wrong with audio systems. They’re referred to as “subtractive” for a reason. They reduce amplitude, they absorb energy and all too often they impact musical coherence and phase relationships within the speaker’s pass-band. Nor is the expression ‘passive’ particularly appropriate for something that has a very real capacity to bite back. Between the crossover and the drive units, the loudspeaker is a far from ‘passive’ partner, simply waiting to be pushed around by the amplifier. Varying impedance, especially sharply varying impedance, delivers some serious push-back, with a very real and musically significant impact on amplifier performance. Alter the load impedance and the amplifier has to respond by delivering significantly more (or less) current – at different points in the frequency range, occasionally uncomfortably adjacent to each other. Throw in the affects of back-EMF, both on the driving amplifier and (potentially) the crossover and it’s a jumble of almost schizophrenic demands. But one thing you can say for sure. Reduce or segment the bandwidth and you’ll definitely reduce the complexity of the problem – at least from the amplifier’s point of view.
