All the graphs are measurements performed on the Audio Precision Model One with the AUX-0025 Switching Amplifier Measurement Filter in place when needed.

Important!
Do not to put all your faith in measurements. There are tangible variables which can't fully be explained by current measurement standards, and every test bench and product will measure a little differently. Readers should take this as a single point on which to further their own understanding. While these numbers are certainly comparable to each other, they shouldn't be directly compared to any other tests run elsewhere without keeping in mind the possible differences between test setups.

Signal/Noise: These measurements were taken using the switch mode power supply that comes with the Super-T amp, except for the AMP6 which has an onboard linear supply. The Super-T power supply was found to be as quiet or better than a lab grade 8A linear supply. Results are thus typical of what most users will encounter, as most of the spectrum noise is below the threshold of most CD resolution.
Example: the Fenice shows a noise level of -112dB thru much of the audio band. If your system peak listening level is 110dB (loud!), then most noise is 2dB below the threshold of hearing. The amps prove their reputation for an inky black, quiet background.

The AMP6 shows a little noise at 60 Hz and a few harmonics. This is likely from the onboard power supply. The noise is not audible in any of my listening tests.

Crosstalk: Usually we see crosstalk given as a simple number at a given frequency. In these measurements you will see crosstalk for both channels as plotted against frequency. Notice that the tighter the PCB layout, the more the crosstalk rises in the upper frequencies. This is especially noticeable on the Sonic Impact 5066.

The peaks in crosstalk on the AMP6 measurement are an artifact of the AC common mode noise. Some of the crosstalk in the other measurements may be as well.

Power Curves: Look at this graph closely. The measurement shows POWER vs THD. As power rises, so does distortion, up to a point where the audio quality is no longer acceptable. Power shown is RMS sine wave, so peaks will be 3dB higher (twice the power).

All the curves show a" break point" where they go rapidly up. I will try to determine what causes this, clipping, soft clipping, other - and post the results.

The power of the AMP6 is higher than the other amps because it is supplied by 14V as opposed to 12V for the other 3 amplifiers.