As already explained (search the archive), if you use cundall's damping,
the rounding errors can propagate infinitely quickly.
I agree that damping discontinuity can make the difference grow faster
in general. But in special cases where you can proove damping has no
effect on the result, it is a bit short.
It seems many multithread runs of the same simulation with damping=0.02
gives higher variability than monothread runs with damping =0.02, 0.1,
0.3. But again, I used initial clouds instead of dense packings to
repeat the test, it might be the reason. I'll know more sooon.
Bruno
_______________________________________________
Mailing list: https://launchpad.net/~yade-dev
Post to : [email protected]
Unsubscribe : https://launchpad.net/~yade-dev
More help : https://help.launchpad.net/ListHelp
