> An R2R scheme might be promising... Hmm, off to the lab... How about a resistor T network? You could get some very big effective resistances with relatively small resistors (or pots), and still use relatively small (i.e., non-leaky) caps. Just a random thought.