Martin Spott wrote:
To my knowledge there _are_ aircraft in FlightGear that are build upon
real data. Right ?
I think this is always the case. Take the B-29 for instance. Josh has obviously done a ton of research to get the dimensions and proportions down exactly right ... that's a key component of dynamics modeling. Along with that you want to figure out the mass of the aircraft and refine the mass distribution throughout the airframe (engines, fuel, pilots, cargo, balance, etc.) Then of course you need flight data, but even with limited flight data, you can start to craft a very plausible model based on aircraft mass, dimensions, power plants, etc.
I don't think anyone sits down and makes up an entire aircraft from scratch based on ficticious number and pure guessing. We use guesses just do that to fill in the gaps for things we don't know. But often the guesses are educated and at least in the right ball park.
In the end, the quality of the model depends greatly on the quality and amount of data the aircraft designer can find.
For Level 3 FTD certification, the FAA has a long lists of flight tests you need to run to validate against real aircraft performance. It is very difficult/expensive to get all the data required for the FAA certification, and no FlightGear model has gone to those extremes that I'm aware of. However, someone with the resources, time, and data could certainly go through the process. There's nothing in our code or infrastructure that would prevent or limit us from being able to do this.
To my experience what makes the most significant difference between
FlightGear and a real aircraft is the limited view angle in FlightGear
- as long as you don't have a a simulator with multiple screens.
But for the price tag of a couple of large TFT displays you can
afford your own PPL ....
Compared to that the difference in the flight 'behaviour' of a real
C172 (or PA-28) and the one in FlightGear is neglectible. At least
these two get really close to reality - pretty much close enough to use
FlightGear for trainig final approaches for example,
Yes to get to a good level of realism you need:
1. A cockpit mockup with all the gauges, switch, lights, controls etc. in approximately the right place.
2. A reasonable field of view on your visual system (often accomplished with multiple projectors, or large screens.)
3. Refined/validated flight dynamics.
4. Motion isn't required at the lower levels of FAA certification, but it is a really nice thing to have when done right. The problem is that it is hard to do right, and there is a cascading effect on other elements of the simulator.
It's actually a lot of fun to play around at this next level up, but it's also a lot of work, and there's enough issues and problems that it's difficult for a single person to do everything well themselves. It often becomes less of a hobby and more of a sickness (or business.) :-)
Curt.
-- Curtis Olson http://www.flightgear.org/~curt HumanFIRST Program http://www.humanfirst.umn.edu/ FlightGear Project http://www.flightgear.org Unique text: 2f585eeea02e2c79d7b1d8c4963bae2d
_______________________________________________ Flightgear-devel mailing list [email protected] http://mail.flightgear.org/mailman/listinfo/flightgear-devel 2f585eeea02e2c79d7b1d8c4963bae2d
