See? The measurements were taken at 1KHz which means that a sine wave signal injector set at 1000 cycles (Hz) was used as input signal. Now if the injectors center frequency would be sweeped higher then those THDs will surely be higher for that particular power and vice-versa.
Although some manufacturers give a spec with a "20Hz-20Khz" range, Pioneer's specs would have been lower if they used this technique (20HZ-20KHz) because it averages the THD levels over the whole frequency spectrum.
Cheers!
I am a bit confused with what you said. You said the higher the frequencies injected, the higher the THDs. So if they had used various injected frequencies across the entire spectrum, how sure are you that they'd get lower THD when averaged as opposed to getting it only at one frequency. You have only 980 frequency points below 1khz with lower THDs, and around 19,000 frequency points above 1khz with higher THDs. Now if they use 1 octave intervals after 20hz, maybe, since there's one more octave below 1khz than above. Even then, taht depends on how much greater the THDs above 1khz are compared to how much lower the THD's are below 1khz.
But that may be beside the point. THD is not averaged. See those amps with the "<" sign before the THD figure, the THD over the entire 20khz bandwidth is not an average figure, but the HIGHEST obtained. That's why they specify "<0.06%, 20hz - 20Khz, so you don't expect the THD to be any higher at that power level at any point across the audible spectrum.
But, here I go again, if you want to hype your specs, you can always average it. In the case of those specs given by
jerrix, the THD is measured ONLY at the 1Khz point. Not averaged. It can be MORE at higher frequencies. Hence, while other conservatively rated euipment will tell you not to expect any higher THD reading on their specs, the pioneer specs opens up the possiblity you can get higher THDs above 1k.
And not only is the 1 khz point a measurement condition that can yield a higher power rating vis-a-vis a full 20hz-20khz measurement condition, but the fact that the measurement is done with only two channesl driven (steeo mode. I assume both channels driven) is already a right step in the overhyping direction for a multichannel receiver. Compare that to the specs of a conservatively rated ROTEL RSX1067:
100 watts/ch (20-20k Hz, <0.05% THD, 8 ohms)
(Seven Channel Driven)
(taken from the rotel site)
The measurement is made across the entire BANDWIDTH with ALL channels driven. And the THD you should expect across that bandwith is less that 0.05%. Not more, but less. Not averaged.
I have other caveat against power measurements done at 1khz instead of full bandwidth or with only one or two channels driven in a multichannel receiver. But that's already out of topic.