[vtkusers] Why is vtkDelaunay2D so slow in VC++ 2005

[Godofredo]

Hi, I haven't tried it yet under another platform. Sure it's slow by itself
but I don't think that it should be that slow. Take a look at this
unanswered post (quite old) regarding to VC7 vs VC6 and vtkDelaunay2D:

http://public.kitware.com/pipermail/vtkusers/2002-December/064429.html

As soon I've time, I'll try under another platform to evaluate the algorithm
performance. Meanwhile any comment is wellcomed. Thanks to all.


santana wrote:
> 
> Hi, Delaunay is slow algorithm in general, it has to work point by point 
> and make a search through the point set with each insertion. I am not 
> sure about VTK but I have seen the time complexity of other 
> implementations is O(n*log(n)). It sounds like you by hand approach is 
> O(n) so would be much faster. You mentioned that you think it is related 
> to VC++ 2005 but I think it is the delaunay algorithm itself rather than 
> the platform which is the problem. have you tried under another platform?
> Burlen
> 
> 
> Godofredo wrote:
>> Hi to all, I've been doing some triangulations with big datasets and
>> found
>> out that vtkDelaunay2D is extremely slow in VC++ 2005. I've done the
>> triangulation by hand (as my data came from a range scanner I just join
>> the 
>> verts in order) and It takes only seconds to triangulate what with
>> vtkDelaunay2D takes almost 2 minutes. I've found in the forums a post
>> that
>> talked about the same problem but had no replys. Anyone could put some
>> light
>> into this? Many thanks.
>>   
> 
> _______________________________________________
> This is the private VTK discussion list. 
> Please keep messages on-topic. Check the FAQ at:
> http://www.vtk.org/Wiki/VTK_FAQ
> Follow this link to subscribe/unsubscribe:
> http://www.vtk.org/mailman/listinfo/vtkusers
> 
> 
> -----
> http://quaoar.sr.unh.edu 
> 

-- 
View this message in context: http://www.nabble.com/Why-is-vtkDelaunay2D-so-slow-in-VC%2B%2B-2005-tp14576119p14581000.html
Sent from the VTK - Users mailing list archive at Nabble.com.