Ben Jackson wrote:
> On Wed, Jul 09, 2008 at 07:02:12PM +1200, Anthony Blake wrote:
>> If anyone is interested in the topological autorouter I'm working on, 
> 
> Looks neat!  Nice presentation of the progress.
> 
> I'm curious:  What do your internal data structures look like, and how
> applicable would they be as a replacement for PCB's own clunky internals?
> 

The data structures are still pretty sketchy, but I think there are 
parts which could be put back into PCB. In particular, using the 
delaunay triangulation code we can do EMST in O(nlogn) time rather than 
the O(n^2) with the current code. When moving only a few points, we 
would only end up doing the computation for those points, and then 
generating the MST in linear time. This should improve the rats-nest code.

TBH, I'm not really familiar with a lot of PCB's internals, so it's hard 
for me to say whether or not it would be useful to replace parts.


_______________________________________________
geda-dev mailing list
[email protected]
http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev

Reply via email to