I misunderstood you. Now it makes sense.

For the test, you are, of course, right again. Binned data is what is used in the test so the cdf differences will do it.

Thanks for the help, I think I've finally got things straightened out