Must We Test? Yes, We Must!

"Test We Must," cried High Fidelity's erstwhile editor, Michael Riggs, in a January 1989 leader article condemning the growth of subjective testing. (See the sidebar for Peter Mitchell's obituary of HF magazine, now effectively merged with Stereo Review.) With the exception of loudspeakers, where it is still necessary to listen, he wrote, "laboratory testing (properly done) can tell us pretty much everything we need to know about the performance of a typical piece of electronics...We know what the important characteristics are, how to measure them, and how to interpret the results."

With respect to Mr. Riggs who, all things aside, is still a man who apparently believes that magazines should contain leaders and is therefore a cut above the run-of-the-mill editor, this last statement is ridiculous. Ridiculous. I defy anyone to look at a comprehensive set of CD player or amplifier specifications and correctly predict its sound quality (footnote 1). I was reminded of this editorial, however, when I attended a press conference hosted by NAD at the 1989 SCES. (A full report from the Stereophile commando squad appears in this issue.)

If there is one fundamental amplifier specification that ought to give an unambiguous correlation with what is heard, it is its power rating. An amplifier rated at 100W should, naturally, go louder (by 3dB, for all you techno-buffs) than one capable of pumping 50W into the same load. Yet at CES, NAD put on a convincing demonstration that their 50W FTC-rated 7100 receiver would swing more clipping-free volts into a typical load (represented by a B&W loudspeaker) than even a 90W competing product. The implication was that the 50W NAD product would go louder without distortion than another amplifier officially rated as being almost twice as powerful. "Where's your neat correlation between power rating and loudness capability now?" cried the Wicked Witch of the West. (I think I'm getting The Ten Commandments mixed up with The Wizard of Oz, with which NAD had been showing off the capabilities of their impressive new 13" TV monitor immediately prior to the press conference.)

The real situation, of course, is rather more complicated than at first it might seem. (Take no notice of that man behind the curtain.) The FTC rated power, intended to be an amplifier primary rating, concerns maximum continuous power. There is a secondary power rating, the IHF "Dynamic Headroom," which is the ratio between the maximum power the amplifier can put out for a burst just 20ms long, and the continuous power. An amplifier that can deliver 200W for 20ms but 50W continuously will have a dynamic headroom of 6dB. In combination with the FTC rating, this secondary rating is supposed to indicate how loud an amplifier will play into a specified load. (This, of course, is a resistor, not a loudspeaker, which is another whole can of worms.)

So what is NAD's beef with the FTC? Mainly, as I understand it, that mass-market consumers—MMCs, for short—make their buying decisions on the number of FTC watts alone. This puts NAD's commutating amplifiers—as with products from Proton and Carver, use of multiple-rail power supplies means that they can deliver many times their FTC-rated power for music transients lasting up to at least one second—at a commercial disadvantage. (Heaven forbid that any of these MMCs actually listen to what they're about to splurge their increasingly hard-earned dollars on (footnote 2).

The FTC continuous rating, however, was not arbitrarily chosen. It was imposed on a hi-fi industry that had proved totally irresponsible in how it rated its products. Such mystical numbers as "Peak Power" and "Music Power" were used willy-nilly to produce sales-oriented ratings that had nothing to do with reality. (Today's in-car industry, which has no such regulations governing advertised power ratings, and where an amplifier that sinks at the most maybe 6W RMS into a 4-ohm load is advertised as offering more than 20W (footnote 4), stands as an example of the degree of precision typical of the consumer electronics industry.)

NAD proposes a new primary amplifier rating, one that at least would correlate with how loud an amplifier will actually play on music. Such a rating, called, perhaps, "Burst Power," "Music Transient Power," or "Equivalent Loudness Power"—NAD suggests the mouthful "Musically Effective Burst Power"—would now be the primary quoted rating of a power amplifier. It would be specified as the maximum power into a standard load, measured using a 200ms toneburst superimposed on a -12dB background tone. A 10% duty cycle (ie, the burst peak repeats every two seconds) would give a peak:mean ratio of 15%, equivalent to the highest ratio found in music. The 200ms period was arrived at following research performed by Stereophile's Peter Mitchell, who found that this was the longest transient typically found in music. It is also equivalent to the average note length in music, interspersed by periods 6dB lower in power level.

Footnote 1: Maybe Stereophile should offer a prize?

Footnote 2: I know, I know: I ended a sentence in a preposition. It must be the proximity of the Lone Star State to Stereophile's New Mexico office. I really do know where my syntax is at! (footnote 3)

Footnote 3: Sure you do.—Copy Editor

Footnote 4: To make the grade as a car-amplifier salesman, you need first to understand how an RMS power rating is arrived at. (Know your enemy!) Running from, say, a 14V car supply, a typical single-ended amplifier can swing around 12V peak-peak into a 4-ohm loudspeaker. This voltage is equivalent to an RMS voltage of 12 / (2 x root 2) = 4.24V RMS. This is the DC voltage equivalent to that peak-peak AC voltage when considered as a source for resistive heating. By Ohm's Law, into a 4- ohm load, that 4.24 RMS voltage will raise 4.24 x 4.24 / 4W, ie, just 4.5W! I know, I know, you're not going to sell many 4.5W amplifiers come crank-it-up season, so let's use the power that would be generated if that peak-peak voltage were, instead, an RMS figure. (Your customers are hardly likely to be triple-E graduates.) That gives, let's think up a fancy label, a "Maximum Peak Burst Power" rating, for example—no, replace the word "Burst," it implies that the amp can't keep it up—a "Maximum Peak Music Power" of 36W! Now that sounds more like a man's car amplifier! You're on your way to making your first sale. Joe Isuzu would be proud of you.

MattJ's picture

I like the idea of using dB for power ratings. Benchmark has similar ideas to use dB for distortion tests.