If this is a duplicate, my apologies; having a few minor email problems today and not sure what was sent.
> > Well, this is where it gets interesting :-) We really need to define the > > auto-throttle property nodes, so that the FMC/MCDU can set (some of ) > > those. Or, on the 757/767, this is done via a physical panel, apparently > > (if I understand Thomas' comments). (the 777 uses a THRUST LIM page). > > And on the airbuses (at least the A320) it's again done via the MCDU. > > Presumably John W is going to need to be able to set this too, from his > > external FMC impl. > > > > I would figure on the FMC (or some other source, like a panel) computing > > the appropriate EPR values internally, and setting them via properties. > > On the "real" metal, the engine control units do the calcs and send the data > up. One less thing to worry > about with mismatching what's hanging on the wing. > > > Then the auto-throttle code just needs to track those. > > Yes, there is a page on the MCDU as you described. With regards to the FG > interface, OpenGC works > throught the native_ctrls.[ch]xx code and sends control inputs based on the > class defined in raw_ctrls.hxx. > All other calcs (nav, FMC, autopilot, modes, etc) are done on the OpenGC > side based on the state data > received in the data packet defined in opengc_data.hxx in the Network > directory. I guess the only item to > really "worry" about is that we don't do something to preclude (lock-out) > the external interfaces. With that > it mind, it seems the FG side of developing the FMC/MCDU should be able to > proceed independently and > vice versa on the OpenGC without running into each other. Does that make > sense? > > And while on the subject of interfaces... > > Bernie made some changes (wrappers?) on the socket code. Downloaded the > latest CVS last night. Something is > not working in the TCP connections that was fine in 0.7.10. The read routine > in sg_socket is returning > "Warning: read() not enough bytes." which crumps the control channel from > OpenGC to FG. Well, a quick peek at > the net traffic last night indicates the bytes (chars) are lining up with > the right size, but for some reason the > reported length of the read stream may not be agreeing with the expected > length. > > Haven't had a chance to dig through it, but any suggestions ideas from the > author of the change....? > > Regards > John W. > > > > > _______________________________________________ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
