Hi Tuukka,

On Sun, Apr 15, 2012 at 1:58 PM, Tuukka Verho <[email protected]> wrote:
> Hi Dave,
>
> Your algorithm for finding the shortest bond vector doesn't work right for 
> unit
> cells with angles other than 90deg.

You're right! I saw that method in a patch that went into openbabel
and thought it was clever -- mapping the end position into a
Wigner-Seitz cell around the start position is far less expensive than
checking 27 images explicitly. But yes, this will only work for cubic
lattices. Thanks for catching that -- I'll patch up OpenBabel with a
more reliable method, too.

> I changed it to something that should
> yield correct results but I can't compile it because I get:
>
> /home/tuukka/code/dlonie/libavogadro/src/engines/bsdyengine.cpp:35:34: fatal
> error: avogadro/obeigenconv.h: No such file or directory

Oops -- This branch does depends on this old patch on gerrit:

http://review.source.kitware.com/#/c/4238/

Maybe one of the other core devs will find some time to review the
backlog of patches soon? For now, just copying the obeigenconv.h
header from libavogadro/src/extensions/crystallography to some place
that bsdyengine.cpp can find it will let the code compile. I had a
stale copy in my working tree, so I hadn't noticed it missing from the
patch ;-)

> However I made a patch of my changes, you can see if it compiles.

Thanks -- I tweaked it a little to make it a bit more efficient and
use Eigen vectors/matrices instead of OB's linear algebra
implementation. I've put the new version up on github.

Dave

------------------------------------------------------------------------------
For Developers, A Lot Can Happen In A Second.
Boundary is the first to Know...and Tell You.
Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
http://p.sf.net/sfu/Boundary-d2dvs2
_______________________________________________
Avogadro-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/avogadro-devel

Reply via email to