yade-users team mailing list archive

[Chareyre <question692392@xxxxxxxxxxxxxxxxxxxxx>]

Question #692392 on Yade changed:
https://answers.launchpad.net/yade/+question/692392

Chareyre posted a new comment:
@Leonard
The rationale for not saving/loading Un is that it is always unambiguously
deduced from positions and sizes.
We try to store only independent variables, not derived variables.
What you report is not actually a problem, or it is only a problem of the
way you output penetrationDepth.
Insert a PyRunner to print Un, after NewtonIntegrator. The apparent issue
will disappear.
B


On Thu, 20 Aug 2020 at 17:48, Bruno Chareyre <bruno.chareyre@xxxxxxxxxxxxxxx>
wrote:

> > (I did not really get your comment :-)
>
> I was pointing the fact that the notion of "value of X at iteration N" is
> vague in the first place, since X can change at any point within one
> time-loop.
> Virtually you could insert multiple O.save() or any form of snapshot
> within an iteration, interleaved with the usual engines, and you would get
> different numbers in each of them depending on their relative positions in
> the sequence of engines.
> The above save/load issue is nearly the same thing I would say.
> B
>

-- 
You received this question notification because your team yade-users is
an answer contact for Yade.