With resistive passives controlling the interconnect is more critical. TVCs and AVCs less so (Ralph has this thing about transformers too - so that probably wouldn't be a convincing argument to him). However John Chapman's new design (6 input version) incorporates a buffer stage that can be turned on/off for each input. He comments that using it would be in extreme cases where a source could not drive the passive. No buffer on the output though, which the B1 incorporates in its resistive passive design (along with the input as well). Interestingly enough John chose to just make his TAP resistive passive mini modules with short pigtails that you place near the amp, eliminating the long cable on the output. Not sure how this sounds, but it's similar to the concept of the EVS nude attenuators which you just plug into the amps inputs (I have a pair which I should revisit using).
I'd have to say Nelson Pass's design is interesting and could be worth a try. It's certainly priced right. Just wish it had balanced inputs and outputs. Also, noticed that Nelson commented that a regulated power supply would be better than battery.