On Thu, Aug 8, 2013 at 7:09 PM, Tom Browder <[email protected]> wrote:
> On Thu, Aug 8, 2013 at 6:06 PM, Christopher Sean Morrison
> <[email protected]> wrote:
> ...
> > The main benefit of the gdiam library is that it calculates
> object-oriented bounding boxes (OOBB), which nanoflann does not do. Still,
> it's worth testing nanoflann's kd-tree against libtie's kd-tree build to
> see if we can reduce prep time (substantially).
>
> Seems Cliff is having trouble with C++ weak ordering with gdiam--may
> be something in nanoflann to help.
>
It's weird - I suspect the simple arb point sets are probably
near-pathological cases for this algorithm, given the number of near-zero
comparisons we're winding up with. I don't know if I could just punt and
use qsort to avoid the strict weak ordering check - I may end up having to
dig into the algorithm and figure out what degenerate cases we seem to be
triggering.
Cliff
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite!
It's a free troubleshooting tool designed for production.
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk
_______________________________________________
BRL-CAD Developer mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/brlcad-devel
