yade-users team mailing list archive

[Bruno Chareyre <bruno.chareyre@xxxxxxxxxxx>]

> (BTW you're probably the first person besides me to read that code.)
>
>   
Hehe. I'm sure Gallizzi read it before, and I read it before you too.

> >  Is this contact detection method more efficient in terms of search
> > time 
> > than other methods based on spatial subdivisions? 
> >     
>
> No.
>
>   
It might be better in some circumstances, nobody really knows.

Keep in mind than in single thread and in a typical step with 50k 
element, the contact detection  is 3% of the CPU time.
With 5 threads, it is up to 10% (see 
https://yade.hmg.inpg.fr/index.php/Triaxial_Test_Parallel).

So, yes, it can be better, especially if we speak about very big 
simulations and massive parallelism. But currently, this algorithm works 
quite well with respect to overall performance.

Bruno


> _______________________________________________
> Mailing list: https://launchpad.net/~yade-users
> Post to     : yade-users@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~yade-users
> More help   : https://help.launchpad.net/ListHelp
>
>   


--

_______________
Chareyre Bruno
Maître de Conférences

Grenoble INP
Laboratoire 3SR - bureau E145
BP 53 - 38041, Grenoble cedex 9 - France
Tél : 33 4 56 52 86 21
Fax : 33 4 76 82 70 43
________________