My experience has been poor (biamping). never seemed to make a difference. But in my case I did it using different power amps (seems like I only tested biamping when I somehow ended up with an extra set of amps and was bored - I have never really tried to test it with identical amps.
Ed: have you tried using the Nakamichi PA5 to test this and see if there is a detectable difference?
I have had better luck with bi-wiring, that seems to be easier to get right (I guess that would be obvious
).
pepe