yade-users team mailing list archive

[Bruno Chareyre <bruno.chareyre@xxxxxxxxxxx>]

> 4 - I have make a rapid tour in cachegrind/kcachegrind, and if I 
> understand well, boost operations take a significant part of the 
> computation cost (particularly "/shared_ptr<something> sh_ptr = 
> existing_shared_ptr/" and /shared_ptr<something>~/ ). Don't you think 
> it could be a good idea to write specific parts of Yade, like 
> PersistentSAPCollider, with classical pointers? I seriously think 
> about this type of optimisations.

Forget this! My understanding of the callgrind output was erroneous! I 
summed inclusive costs of all boost::something. Summing _recursive_ 
costs gives a far more acceptable result...

> More generally, I would really appreciate having your point of view on 
> where I should start if I plan to get faster computations.
This question holds...

Bruno

--
_______________
Chareyre Bruno
Maître de conférence

Institut National Polytechnique de Grenoble
Laboratoire 3S (Soils Solids Structures) - bureau I08
BP 53 - 38041, Grenoble cedex 9 - France     
Tél : 04.56.52.86.21
________________

_______________________________________________
Yade-users mailing list
Yade-users@xxxxxxxxxxxxxxxx
http://lists.berlios.de/mailman/listinfo/yade-users