[Gfs-users] strange results from GfsInit

[Blake H. <sta...@gm...>]

Hello all:

I've noticed some unusual results from GfsInit.  I'm not sure if my  
expectations are wrong, or what's going on.  Anyways, for the  
vorticity.gfs example, if I use GfsOutputScalarStats for time = 0,  
some of the velocities are outside of the expected range,  
specifically the maximum and minimums of U and V are greater than 0.5  
and less than -0.5, respectively.   At the start of the simulation,  
all values of U and V should be between -0.5 and 0.5, correct? I  
thought this could be a problem with the rand()/(double)RAND_MAX part  
of the function, so I stripped GfsInit down to just U=0.5 and V=0.5,  
and the results get more extreme:

U time: 0 min:  5.000e-01 avg:  5.000e-01 |  4.478e-06 max:  5.000e-01
V time: 0 min: -7.785e-05 avg:  3.296e-05 |  4.346e-05 max:  1.094e-04

and it looks like the V component isn't being set at all.  This is  
basically the case for any non-zero initialization values I provide.   
Anyways, I've pasted my simulation file below, I'm curious if other  
users get the same results.   My expectations are that we should be  
able to set the V component to a constant value, but that doesn't  
seem to be the case.



As always, I appreciate any suggestions, and thanks in advance.

Blake Huff
sta...@gm...

1 1 GfsSimulation GfsBox GfsGEdge {}{
   GfsTime { end = 10 }
   GfsRefine 6
   GfsInit {} {
#	V = 0.5
	U = 0.5 - (rand()/((double)RAND_MAX))
	V = 0.5 - (rand()/((double)RAND_MAX))
   }
   GfsOutputScalarStats {istart = 0 istep = 1 iend = 5} stdout {v = U}
   GfsOutputScalarStats {istart = 0 istep = 1 iend = 5} stdout {v = V}
}
GfsBox {}
1 1 right
1 1 top