"Ring flight is... not really running off of displacement-via-applying-force? I'm not quite sure how to explain it; it's intuitive under my built-in model of physics, but my built-in model of physics diverges from the one used by your society in many ways. Her apparent velocity will scale with local conditions (inversely with air pressure, to a rather large degree) and also with the strength of her hope. At her maximum recorded hope output (which she hit last night, during her second batch of healing; a new record by about 15%) she's currently capable of hitting an apparent velocity of about Mach 23 (at 1 atmosphere of pressure), but that's not realistic to fighting conditions. In practice sessions, her hope output when stabilizing after being startled was enough for the Mach 3-5 range."
"My combat modeling indicates that the biggest limitation on her evasiveness is her reaction time and not her top speed; the augmented soldiers last night were able to hit her while she was going very fast, and she could have only dodged if she'd changed directions between when their rounds left their weapons and when they would have hit. Though, now that I'm thinking about how her flight interfaces with the ring's intent-interpretation subsystems, it may actually be possible for me to write up a flight-path assistance control that automates away individual decisions? There's existing code in here that's doing similar things for healing and complex construct generation; I can look into modifying it."