yade-dev team mailing list archive

[Bruno Chareyre <bruno.chareyre@xxxxxxxxxxx>]

On 30/08/13 07:49, Anton Gladky wrote:
> Hi Bruno, thanks for additional PFV-test.
> Could you, please, reduce the tolerance of the script?
> yade-daily fails to build [1].
>
> ___________________________________
> ###################################
> running:  DEM-PFV-check.py
> DEM-PFV: difference in bulk modulus: 269497.792393 vs. target  263673.1423
> DEM-PFV: difference in permeability: 0.0503721236865  vs. target
> 0.0512650663801
> DEM-PFV: difference in final pressure: 558.542865389  vs. target  528.554831762
> DEM-PFV: difference in final deformation 0.00254330893961  vs. target
> 0.00265188596144
> Status: FAILURE!!!
>

I have seen that. In fact the difference comes even before the fluid is
used.
It seems that the same "seed" in makeCloud(...) gives different random
numbers on different platforms. Tolerance would only hide the problem.
The results should be exactly the same.
I pushed a data file with initial positions to make things really the
same. Looking forward.

Bruno