One can easily get a much narrower skirt at the bottom by doing synchronized measurements, which seem to be the norm. Simply connect input to output of the ADI and you get that. But these measurements are quite misleading (read too good to be true), so I used two units, where one is fed analog into a second unit, unsynchronized, which is a much better real-world setup. By doing so the skirt gets widened, but that doesn't matter at all. What matters is that there are zero (0) sidebands visible. The widening represents lowest frequency jitter at very low levels that stay definitely inaudible.

I have measured many units even with special low phase noise crystals. While most show a bit more narrow skirt, they suffer from lots of sidebands, peaking out of the noise floor up to 30 dB, simply because they have no effective jitter reduction in the audio band.

Finally: if the widening would have any meaning, the ADI-2 Pro would not have the specs that it has - and these have been confirmed by several magazines already.

Hello Matthias,

Many thanks for the explanation which helps a lot.
I've recently bought the ADI 2 Pro for home use and I think it's a phenomenal product.

Just wondering:
The first graph here with signal related sidebands but in the manual no sidebands - both are done with 44.1 kHz?

Greetings
Andreas