[ODE] more air friction

[Andrew Aye]

I think you're worrying too much about accuracy or "realism".  If you wanted
to be "accurate" then you would also have to take into account that air
"friction", has as a principal component fluidic drag - which is dependent
on not only the square of the velocity - but more importantly on the
cross-sectional area of the object compared to the direction of motion.  ie.
Any air friction calculation that you're thinking of doing is most likely a
huge kludge/approximation - so trying to distinguish between t0 and t1 for
reasonable size dt is going to be relatively insignificant error when
compared to the errors introduced by the design/approximation of the system
in the first place.  As a first order kludge to get objects to simply stop
rolling, I just applied a counter force and moment to imitate a linear
resistance.

 - Andrew Aye

-----Original Message-----
From: Ivan Bolcina [mailto:ivan.bolcina at snt.si]
Sent: Saturday, April 24, 2004 1:38 PM
To: ode at q12.org
Subject: Re: [ODE] more air friction



Another thing on air friction.
If I calculate the force that friction produces on a current moment with 
current speed, then this is only current air friction force.
If I then run simulation for a given dt, now this force applies for all 
dt. But air friction force should be diferent at,for example 0.5dt, 
since it has already 0.5dt time to slow down object.
So I should pass to ODE not constant force, but rather formula of air 
friction. or should I canculate avarage airfiction force, which applies 
for all dt, not just current time.


bye, ivan


_______________________________________________
ODE mailing list
ODE at q12.org
http://q12.org/mailman/listinfo/ode