Re: [Flightgear-devel] Eye candy

[Josh Babcock]

Actually, I was thinking of playing with wing flexing with the model 
that I am working on now, but I figured that I already had too many 
bells and whistles.  I was going to divide each wing into about four 
sections and put them into a tree-like structure with all the engines, 
etc hanging off the group objects with the wing segments.  Then you can 
rotate the inner group by a factor of the g-load, and again for each 
section.  The effect should be cumulative at the tip, but I never tested 
it.  There might be a problem with the pivot points not matching up when 
the wing is flexed though, you might have to do some translating too.  
This would be a great feature if anyone ever modeled the Voyager.  Then 
the trick would be modeling the wingtip scrapes and deleting the wingtip 
lights, just like in real life!
Josh

Lee Elliott wrote:

On Friday 13 February 2004 22:27, Jon S Berndt wrote:
 

Any chance of modeling wingtip vortices (when CL is high enough above
some threshhold) and rocket engine exhaust?
:-)

Jon
   

I've thought about trying this but I think it could only be really effective 
in level flight.  As soon as you start banking the 'end' of whatever trail 
you're simulating will rotate (echos of the VRP issues:).

For a very short trail this might not be noticable but for longer ones I think 
it's a bit of a show stopper.

It would be possible to include curved trails in the model, and the select 
anim function could be used to select the appropriate model object, but then 
you'd need a wide range of trail objects.

At first glance, this doesn't seem too bad, but then I think you'd really need 
several versions of each trail type so that you could switch between them to 
give the impression that the trail is changing throught time.

Hmm...   I don't think I've explained that bit very well...

Consider an a/c in stable level flight where votices are being generated from 
the wing-tips.  Even though the vortex trail may not change in length or 
shape, it's appreance will change from frame to frame, so you'd need to 
switch between several identically shaped and sized objects that are textured 
slightly differently to give the impression of change.

If we could actually modify the model objects themselves then we could 'bend' 
the trails, and cut down on the number needed.

But then we could also do wing flexing too.

Before anyone starts thinking seriously about this, it would be very 
non-trivial to do, at least for the wings, where some objects would need to 
be 'bent' i.e. the wings and control surfaces, whereas other objects would 
have to simply translate i.e. wing mounted engine nacelles and U/C.

For a simple object, like a vortex trail, bending might be feasible, and 
combined with scaling (which I think we already have), it might work ok.

Another possiblity would be some sort of particle object handling where 
temporary objects could be generated, left for a while and then culled.  We 
could then 'drop' a stream of these objects behind the a/c that're culled 
after a certain time.  It'd probably need a lot of objects to work though, 
and it would also push up the object count of course.

Both methods would need some carefull texturing.

LeeE

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



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