CGAL users discussion list

[Juan Carlos Lopez Alfonso <>]

Hi all,

Thank you very much for your help, I have now a 3D triangulation with
over three millions of points in about 35 min and this si very good
for my simulations. Many thank!!!!

In my first email to this list, I asked a question related with a 
Triangulation:

I have a list of points, where no point is repeated (I am very sure of
this, exactly 3276412 points) and when I use this constructor
"Triangulation T(L.begin(), L.end());" (I am using Triangulation_3) my
resulting triangulation has  less point (using this function
"size_type vertices = T.number_of_vertices();" the triangulation has
3260084 points). Why the triangulation has less "points"? This is not
a desired behavior!!!

Best Regards and Thank you again
Juan Carlos

On Tue, Oct 25, 2011 at 11:18 PM, Monique Teillaud
<>
 wrote:
> Le 25/10/11 22:45, Juan Carlos Lopez Alfonso a écrit :
>
>> You say me that my example in your computer takes 37 min, It was in
>> release mode?
>
> yes, Andreas would only comment on running times for the release mode.
> As its name says, the debug mode is provided for debugging purposes, not for
> measuring performance. It is known to be slow.
>
> Best regards,
> --
> Monique Teillaud
> INRIA Sophia Antipolis - Méditerranée
> http://www.inria.fr/sophia/members/Monique.Teillaud/
>
>
>> On Tue, Oct 25, 2011 at 10:37 PM, Andreas Fabri
>> <>
>>   wrote:
>>>
>>> Did you build it with cmake? What generator ( nmake or .sln) ?
>>>
>>> andreas
>>>
>>> On 25/10/2011 21:53, Juan Carlos Lopez Alfonso wrote:
>>>>
>>>> Hi Andreas:
>>>>
>>>> Thank you again for your answer and your help. I have a question: Are
>>>> you running this example in a Release mode?  About that I could have
>>>> something wrong, I dont know what is it,  because I am running exactly
>>>> this example. In all cases I have a correct triangulation, but with 3
>>>> hours of computing approximately.
>>>>
>>>> best regards
>>>> Juan Carlos
>>>>
>>>> On Tue, Oct 25, 2011 at 9:40 PM, Andreas Fabri
>>>> <>
>>>>     wrote:
>>>>>
>>>>> Hi Juan Carlos,
>>>>>
>>>>> On my Intel Core 2 duo 2.8 Ghz   your program takes 37 min.
>>>>> And I bring it down to 5 min with our local fix.
>>>>>
>>>>> If it takes hours on your machine you probably do something
>>>>> wrong.
>>>>>
>>>>> best regards,
>>>>>
>>>>> andreas
>>>>>
>>>>> On 25/10/2011 15:15, Juan Carlos Lopez Alfonso wrote:
>>>>>>
>>>>>> Hi:
>>>>>>
>>>>>> Thank again for your help. In these links you can download  the
>>>>>> example code (example.cpp) and the data file (either in .rar or .txt),
>>>>>> with the set of points, where in the first line of this file is the
>>>>>> number of points and in the other lines the listed set of points. If
>>>>>> you didn't download these files, please tell me where I can put them
>>>>>> in order that you can prove my example and give me any suggestion.
>>>>>>
>>>>>> http://dl.dropbox.com/u/2516160/example.cpp
>>>>>> http://dl.dropbox.com/u/2516160/treatmentPoints.rar
>>>>>> http://dl.dropbox.com/u/2516160/treatmentPoints.txt
>>>>>>
>>>>>> In my case, I am sure that I have several group of points with the
>>>>>> same x,y and only varying the z (you can see my file of points and is
>>>>>> very easy see that), because I am trying to triangulate a TAC of a
>>>>>> radiotherapy patient with the defined organs involved. I have for all
>>>>>> slices of a TAC diferent points with the same value of x,y only
>>>>>> varying the coordinate z. I am using a set of points of a radiotherapy
>>>>>> treatment plan and from different slice of the TAC in 2D I am forming
>>>>>> a 3D domain with these points. For this reason, I have almost the same
>>>>>> points x,y only varying the z for each slice of the TAC.
>>>>>>
>>>>>> Best Regards from madrid and waiting for your answer
>>>>>> Juan Carlos
>>>>>>
>>>>>> On Tue, Oct 25, 2011 at 2:23 PM, Andreas Fabri
>>>>>> <>
>>>>>>       wrote:
>>>>>>>
>>>>>>> If you want you can post your code,
>>>>>>> and put the data on a server so that we can
>>>>>>> download it and try to reproduce/fix the problem.
>>>>>>>
>>>>>>> It might also be the data. We had a user where
>>>>>>> all points where on a 3D grid, or all points
>>>>>>> where organized in k groups with identical x,y
>>>>>>> and only varying z coordinate, where CGAL was
>>>>>>> rather slow. We have a not yet released solution
>>>>>>> for that.
>>>>>>>
>>>>>>> best regards,
>>>>>>>
>>>>>>> andreas
>>>>>>>
>>>>>>> On 25/10/2011 14:14, Juan Carlos Lopez Alfonso wrote:
>>>>>>>>
>>>>>>>> Hi:
>>>>>>>>
>>>>>>>> Thank you again for your answer. Yes, you are right I am using
>>>>>>>> Exact_predicates_inexact_constructions_kernel. For example with 4618
>>>>>>>> points, when I use the Triangulation_3 the triangulation is computed
>>>>>>>> in 3 sec and with Delaunay_triangulation_3 in 20 sec.
>>>>>>>>
>>>>>>>> But my real simulation is with more than 3500 000 of points and 3.5
>>>>>>>> hours is much time for my time real simulations only computing the
>>>>>>>> triangulation.
>>>>>>>>
>>>>>>>> Any comment about how I can do to improve my simulation time?
>>>>>>>>
>>>>>>>> Best regards agian, and sorry for your time
>>>>>>>> Juan Carlos
>>>>>>>>
>>>>>>>> On Tue, Oct 25, 2011 at 1:49 PM, Marc 
>>>>>>>> Glisse<>
>>>>>>>>  wrote:
>>>>>>>>>
>>>>>>>>> On Tue, 25 Oct 2011, Juan Carlos Lopez Alfonso wrote:
>>>>>>>>>
>>>>>>>>>> Thank you for your answer. About my second question about the
>>>>>>>>>> computing time, you say that the problem would be my memory, but
>>>>>>>>>> could
>>>>>>>>>> you tell me any suggestion about how I can obtain better
>>>>>>>>>> performance?
>>>>>>>>>> I am using CGAL 3.9 with visual studio 2010, in a Pentium(R)
>>>>>>>>>> Dual-Core
>>>>>>>>>> CPU 2.10 GHz and 2GB RAM. On the other hand, I have the maximum
>>>>>>>>>> value
>>>>>>>>>> (8190 MB) of virtual memory in my windows 7 option.
>>>>>>>>>
>>>>>>>>> To check whether it is memory, you should try with fewer points and
>>>>>>>>> see
>>>>>>>>> how
>>>>>>>>> the running time evolves.
>>>>>>>>>
>>>>>>>>> But first:
>>>>>>>>> * I assume you are using
>>>>>>>>> Exact_predicates_inexact_constructions_kernel,
>>>>>>>>> right?
>>>>>>>>> * Are you using Triangulation_3? Did you try
>>>>>>>>> Delaunay_triangulation_3?
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Marc Glisse
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> You are currently subscribed to cgal-discuss.
>>>>>>>>> To unsubscribe or access the archives, go to
>>>>>>>>> https://lists-sop.inria.fr/wws/info/cgal-discuss
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Andreas Fabri, PhD
>>>>>>> Chief Officer, GeometryFactory
>>>>>>> Editor, The CGAL Project
>>>>>>>
>>>>>>> phone: +33.492.954.912    skype: andreas.fabri
>>>>>>>
>>>>>>> --
>>>>>>> You are currently subscribed to cgal-discuss.
>>>>>>> To unsubscribe or access the archives, go to
>>>>>>> https://lists-sop.inria.fr/wws/info/cgal-discuss
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Andreas Fabri, PhD
>>>>> Chief Officer, GeometryFactory
>>>>> Editor, The CGAL Project
>>>>>
>>>>> phone: +33.492.954.912    skype: andreas.fabri
>>>>>
>>>>> --
>>>>> You are currently subscribed to cgal-discuss.
>>>>> To unsubscribe or access the archives, go to
>>>>> https://lists-sop.inria.fr/wws/info/cgal-discuss
>>>>>
>>>>>
>>>>
>>>
>>>
>>> --
>>> Andreas Fabri, PhD
>>> Chief Officer, GeometryFactory
>>> Editor, The CGAL Project
>>>
>>> phone: +33.492.954.912    skype: andreas.fabri
>>>
>>> --
>>> You are currently subscribed to cgal-discuss.
>>> To unsubscribe or access the archives, go to
>>> https://lists-sop.inria.fr/wws/info/cgal-discuss
>>>
>>>
>>
>
>
>
> --
> You are currently subscribed to cgal-discuss.
> To unsubscribe or access the archives, go to
> https://lists-sop.inria.fr/wws/info/cgal-discuss
>
>