On February 18, 2005 08:02 am, Melchior FRANZ wrote:
> I'll find out anyway. I can't stand people borrowing things from me and
> returning them like *that*!
>   http://members.aon.at/mfranz/exhibit_A.jpg
> m.

Yes, that's a very nice touch. =)

On February 18, 2005 12:45 pm, Christian Mayer wrote:
> This reminds me, that we don't have a nice crash simulation yet. (It
> doesn't need to be realistic).
> We probably could do this, when an crash occures:
> 1) switch to external view (that should be easy)
> 2) split up the scenegraph of the model and solve a simple motion
> equation for the parts (so that they "bounce" on the ground - that needs
> a heavy damping though)
> 3) put a fireball in the middle
> The result isn't realistic but much "nicer" than the current approach
> (it looks more like an arcade game though...)
> Step 1 and 2 shouldn't be too hard to do with the current code.
> CU,
> Christian

Every landing is basically a crash... or every crash is basically a landing. 
The two terms just differ in how much damage the aircraft received. ;-)

In my opinion, it will be better in the long run to model fatigue instead.  By 
modelling fatigue:
* damages can be accumlated.  For example, the aircraft can make 19 hard 
landings without incident, but has its landing gears break off in the 20th.
* parts can still break off right away if enough force is applied, like what 
you've suggested in #2.

For visualizations, different effects can be created for different materials 
contacting with some other material.  For contacts with concrete or asphalt: 
rubber will create smoke, metal will create sparks, fuel tanks will create 
fireballs and black smoke, etc.

Putting all the above into a fatigue-routine or a fatigue-class, we can make a 
call to it everytime the aircraft makes a contact with the ground.  We can 
make a call to it every second, too.  Afterall, aicrafts can disintegrate at 
any time.


Flightgear-devel mailing list

Reply via email to