> Again to ground my understanding - if a caller sets a RetryPolicy, does that mean we completely ignore RetryOptions? (It sounds like what you're saying above, but I want to confirm I'm understanding correctly.)

Nope, we still will thread the NetworkTimeout through to the ResponseBodyPolicy. I agree that the story is a bit confusing with how we are presenting RetryPolicy as an override to RetryOptions. I would suggest we consider moving NetworkTimeout somewhere that is more sensible (perhaps TransportOptions) as @christothes suggests.

Originally posted by @JoshLove-msft in #37715 (comment)