"Norman Vine" <[EMAIL PROTECTED]> wrote: > If you have a time series of positions, you have a time series of velocities > ie the velocity is just the first difference ( Pi - P(i-1)) / dT; > And acceleration is just the 2nd differeance
Oh, yes, I know this numerical stuff quite well because I did massive use for my diploma thesis. But the DIS protocol goes one step further and they would not have included this additional stuff into an IEEE standard if it were of no use. The 'classical' numeric methods, as you already know, allow some sort of prediction based on the _past_ movement of an object. You can optimize your prediction by different weighing of the past waypoints based on empirical experience. I developed such an algorithm - it was uncommon, sort of 'simple' but very effective for my use :-) The DIS protocol is one step ahead. It tells the client in which direction the object will _definitely_ move at which speed. From this information the client can do a much better prediction. The server itself does this prediction on the same information _too_ and this is the reason why the server always knows what position and velocity of the object the client is currently assumin. The server delays sending updates to the client until it's own (the servers) prediction differs from the actual position and speed of the object. This is quite tricky, isn't it :-))) Martin. -- Unix _IS_ user friendly - it's just selective about who its friends are ! -------------------------------------------------------------------------- _______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
