yade-dev team mailing list archive

[Janek Kozicki <janek_listy@xxxxx>]

I will look at that next week, when I'm back from COSPAR 2010
conference on 26.07. I hope to find the reason for those recently
occurring unexpected NaNs.



Janek Kozicki said:     (by the date of Thu, 15 Jul 2010 15:40:26 +0200)

> I wonder if this can be related to problems with O.load(), because
> here I am NOT using O.load. I am only generating a new sample using
> somewhat more complex method, including a call to O.reset().
> 
> I am attaching a stripped down script, but still a bit complex.
> 
> After I run that script, in the end I can get a following thing in python:
> 
> Yade [1]: O.bodies[0].state.dict()
>  ->  [1]: 
> {'accel': Vector3(0,0,0),
>  'angAccel': Vector3(0,0,0),
>  'angMom': Vector3(0,0,0),
>  'angVel': Vector3(0,0,0),
>  'blockedDOFs': 0,
>  'inertia': Vector3(1.680768072756119e-10,1.680768072756119e-10,1.680768072756119e-10),
>  'mass': 9.7475048955972981e-05,
>  'refOri': Quaternion((1,0,0),0),
>  'refPos': Vector3(0.046777726741579144,0.064726888517427708,0.060929861505333097),
>  'se3': (Vector3(0.046777726741579144,0.064726888517427708,0.060929861505333097),
>          Quaternion((1,0,0),0)),
>  'vel': Vector3(0,0,0)}
> 
> Yade [2]: O.run()
> Yade [3]: 48474 FATAL yade.ThreadRunner /home/janek/20-Programowanie/10-cpp/50-Yade/Code/HEAD/trunk/core/ThreadRunner.cpp:31 run: Exception occured: 
> Body #0 has velocity==NaN!
> 
> 
> Now - I have just checked that velocity is 0,0,0. And for some reason
> it is NaN. why?
> 
> -- 
> Janek Kozicki                               http://janek.kozicki.pl/  |


-- 
Janek Kozicki                               http://janek.kozicki.pl/  |