yade-users team mailing list archive

Thread
Date

Re: [Question #660585]: Why avoid reseting particle positions in DFNFlow?

To: yade-users@xxxxxxxxxxxxxxxxxxx
From: Bruno Chareyre <question660585@xxxxxxxxxxxxxxxxxxxxx>
Date: Wed, 15 Nov 2017 09:34:47 -0000
Reply-to: question660585@xxxxxxxxxxxxxxxxxxxxx
Sender: bounces@xxxxxxxxxxxxx

Question #660585 on Yade changed:
https://answers.launchpad.net/yade/+question/660585

    Status: Open => Answered

Bruno Chareyre proposed the following answer:
Hi,
What justified the existence of a specific function was the need to make it do nothing... There was probably a better way than inserting one line in a duplicated code but at least this one did not touch the code of another engine. 

Why this option was needed is a bit less clear. The idea of not triangulating was initially to gain cpu time for stiff materials, as you suggest Robert, nothing else  IIRC. We subsequently forgot to write code compatible with updatePosition=true, I guess (showing that premature optimizations are not a good idea in general).
Possible problems I suspect are related to the algorithm scanning the edges of the triangulation to check which ones have a crack, and changing permeability accordingly. Not sure it works very well if the triangulation itslef is changing.
I don't see a fundamental reason why it should not work, it seems more like a blindspot of implementation.

Bruno

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