> Good to see that you're indexing these files (all in one place and easy to
> get).  Is anyone moving them into the cvs?  Do others want that done?

With so many airplanes about to be modeled I wonder if it's time we considered 
keeping most of them in seperate packages, and what a standard package should 
look like.  I could put the files in t38.tar.gz and/or t38.zip archives that 
when extracted into $FGROOT creates directories Airplanes/T38, 
Airplanes/T38/Panels and Airplanes/T38/Models (if they don't already exist).  
I can see the Instruments, Huds and Engines directories getting out of 
control, so maybe it would be best to keep those configs within the T38 
directory.  This way new airplane packages will stay modular.  It also makes 
it easier to distribute packages that can't be GPL'd for some reason, and 
easier for FG to be sure of its GPL'dness.  FG can then link to the "off-site 
hangers" where we can keep the packages.


> As for the aero data, are you setting the values by 'tuning' and
> 'borrowing' from other aircraft?  Or are you calculating the stability
> derivatives?

Mostly tuning and borrowing.  I started with the C172 and C310 configs and 
began replacing numbers.  Some data comes from the usual textbooks, like 
Roskam, McCormick, Anderson, and Etkin, but that amounts to very little, and 
usually you have to borrow numbers from similar aircraft.    The problem is 
that most of the data is proprietary, so we can only guess.  There is a NASA 
document that has stability data on the T-38 and some other fighter-types, 
but I don't want to spend the 50+ dollars to get it.

Right now the T-38 flies very much like like a T-38 (once the yaw damper went 
in), but I'd love to put real T-38 data in anyway if it was available.  Of 
course the data problem is even worse for other airplanes, but their handling 
qualities are very T-38'ish so the T38 numbers should suffice at first.  The 
performance numbers are different, but those are mostly functions of size, 
weight and thrust.

To me the biggest obstacles to realism in the flight models is lack of "feel" 
in the joystick and lack of the usual visual/tactile/aural cues.

Dave  

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

Reply via email to