McCreanor, Duncan writes:
> When we were developing the Multiplayer code, we tested a crude but
> simple replay mechanism.
> 
> Using tcpdump we captured the multiplayer broadcast packets to a file.
> We then used a simple program that read the captured data (used libpcap)
> and put the packets back on the network as if Flightgear had generated
> them.
> 
> Multiple different flights were recorded with each flight having a
> different callsign. We then replayed the data files in parallel to
> multiple instances of Flightgear on separate computers. The effect was
> to see multiple aircraft flying around a control tower view. Each
> Flightgear instance was set show a different 45 degree view from the
> tower. We had four computers set to provide a 180 degree view of the
> airport.
> 
> The update rate of the recorded aircraft was controlled by the rate of
> multiplayer output and input used for recording and playback.

Yes, this is an interesting alternate approach to doing an instant
replay sort of feature.  I can think of a couple other ways to do
instant replay as well.  I guess this is in the grand tradition of
computerdom ... having 101 ways to accomplish the same thing. :-)

<OT> Unfortunately, in today's world, this sort of flexibility often
goes the same way as command line options and command terminals.  Too
often we get one way to do something via a gui, or if the program
authors didn't consider what you want to do, you get no way to do it.
That's the price of progress I guess.  When people talk about being
progressive, I always wonder what they are progressing towards. :-)
</OT>

Curt.
-- 
Curtis Olson   IVLab / HumanFIRST Program       FlightGear Project
Twin Cities    curt 'at' me.umn.edu             curt 'at' flightgear.org
Minnesota      http://www.menet.umn.edu/~curt   http://www.flightgear.org

_______________________________________________
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel

Reply via email to