On Tue, 9 Mar 2010, Vikram Garg wrote:

> Yes, removing existing boundary info makes it work. But I guess what
> needs to be fixed in the long term is the support for multiple ids
> you were mentioning.

Definitely.  At the very least a more informative assertion test would
be helpful.

> Also, I dont see why the boundary ids are already set?

Convenience - if you're building a simple domain and you don't already
have a convention for "this id means no-slip, that id means Dirichlet
inlet velocity, etc." then you can just use the existing ids in your
physics.

> As a user, there seems to be no way for me to know that.

Hmm... not a good way, at least.  Example 13 mentions that
build_square sets boundary ids, but the documentation for build_square
itself doesn't mention it.  I'll fix that.
---
Roy

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Libmesh-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-users

Reply via email to