> From: "Torsten Dreyer" > > Loosing the ability to control the atmosphere within FlightGear would be > a massive loss of functionality. We currently control atmosphere in > several different ways like real weather based on METAR. The "local > weather" project has just started to create a complex weather simulation > system and I am (slowly but steadily) working on adding more live > weather data, including aloft wind, temp and dewpoint for any point of > this world. > It would be a huge degression to be able to fly within ICAO standard > atmosphere, only. > > Torsten
I should have worded this better, because it's not my intention to propose removing functionality with FlightGear. What I mean to ask is what is the current (and planned) way that FlightGear interacts with atmosphere modeling (not referring to winds, at the moment)? The JSBSim standard atmosphere models the U.S. Standard Atmosphere, and supports user-supplied adjustments to the temperature. In the FlightGear/JSBSim interface there is an allowance made to drive the atmosphere model by setting temperature, pressure, and density. At this point (with FlightGear driving STP), any calculations done by the JSBSim standard atmosphere model because superfluous. In the case where FlightGear is to drive the atmosphere model for JSBSim aircraft, there should be what amounts to a null atmosphere model that only provides the C++ interface to storage and common atmosphere calculations (viscosity calcs, for instance). Trying to make the JSBSim standard atmosphere model serve both purposes has resulted in code that is convoluted and difficult to read, and error prone. So, the question that remains is to consider how to provide the capabilities that both parts need. I'm beginning to think that FlightGear should provide its own atmosphere model, and then just copy values into a null JSBSim atmosphere model when integrated with FlightGear, or be happy with the JSBSim implementation of the standard atmosphere with temperature and pressure offsets. Winds and turbulence are another matter. We have a couple of turbulence models and the recently added Milstd and Tustin wind models are looking pretty good. In a discussion on the JSBSim developer list, I proposed separating out the wind model and the atmosphere model. It's all open for discussion ... JB ------------------------------------------------------------------------------ What Every C/C++ and Fortran developer Should Know! Read this article and learn how Intel has extended the reach of its next-generation tools to help Windows* and Linux* C/C++ and Fortran developers boost performance applications - including clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
