You clearly do not know me. You've a learning curve ahead, and it may be rather steep. Ask questions if you do not understand, as you've been shooting first, yet asking no questions.
A scientific standard is one that is invariant of the conditions.
A DBT using interconnects requires the changing of the wires via some method. If one uses simple swaps, then one is subject to the resistance of the contact connections and the loop area of the wires..NEITHER are controlled.
Two interconnects between two chassis form a ground loop which allows the currents to take the path of lowest impedance. There is NO control, period.
And both loop area and contact resistance are significant in terms of the current path that the source's signal will take getting back. And that's only for class 2 devices. It gets worse with safety bonded units. Going differential removes a very significant amount, but even there inductive coupling is ignored.
Now, consider the use of an ABX box. How does that control the path of the return current?
Quite honestly, everything I've seen to date has been done with virtually NO control over current path.
cheers, jnjn