Subject: CGAL users discussion list
List archive
- From: Kim <>
- To:
- Subject: Re: [cgal-discuss] 2D Boolean operation
- Date: Sun, 21 Jun 2009 14:38:21 +0200
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type; b=vqEjHTacT4UOsAeXKv1Owkeg0nwBo5aocpRSNFcKCyJNkDStc58qaHGBN8dY0me+uI 5uFnxwpKtqj0AtPBIHsKmPKDNPqKquVqanOjm94pgRxR/Y/cbaCArgJxfgAGBDrCpnje UvQynPZbX514mGUbDoG9FAz5y7LgS00tiS0m8=
Hi Alex
>>How is your speed results ?
Time : 2-3 minutes ..
cgal ++ , debug lib
XP
Mobile AMD sempron
3500+
1.79 Ghz
1.87 Gbyte ram
>>Which kernel you had used ?
typedef CGAL::Quotient<CGAL::MP_Float>
Base_nt;
typedef CGAL::Lazy_exact_nt<Base_nt> Coord_type;
struct Kernel : public CGAL::Cartesian<Coord_type> {};
typedef Kernel::Segment_2 Segment;
typedef Kernel::Point_2 Point_2;
typedef Kernel::Circle_2 Circle;
typedef Kernel::Iso_rectangle_2 Iso_rectangle;
typedef CGAL::Gps_circle_segment_traits_2<Kernel> Traits;
As you see , speed it not optimal , since my polygons in this case have a low count of segments.
(since many of the polygons are 45deg lines etc then the bounding boxes in many cases don't get as effective as wished
this above speed example was a subset of my full job ,,, if I take the full job (14000 polygons , 6 segments) ,, it will take 22 minutes.
Best regards
Kim
typedef CGAL::Lazy_exact_nt<Base_nt> Coord_type;
struct Kernel : public CGAL::Cartesian<Coord_type> {};
typedef Kernel::Segment_2 Segment;
typedef Kernel::Point_2 Point_2;
typedef Kernel::Circle_2 Circle;
typedef Kernel::Iso_rectangle_2 Iso_rectangle;
typedef CGAL::Gps_circle_segment_traits_2<Kernel> Traits;
As you see , speed it not optimal , since my polygons in this case have a low count of segments.
(since many of the polygons are 45deg lines etc then the bounding boxes in many cases don't get as effective as wished
this above speed example was a subset of my full job ,,, if I take the full job (14000 polygons , 6 segments) ,, it will take 22 minutes.
Best regards
Kim
On Sun, Jun 21, 2009 at 2:37 PM, Kim Pedersen <> wrote:
Hi AlexI had 2500 Polygons consist of approx 6 segments (with arc/curve segments) ... resulting polygon approx. 600.
>>How is your speed results ?
Time : 2-3 minutes ..
cgal ++ , debug lib
XP
Mobile AMD sempron
3500+
1.79 Ghz
1.87 Gbyte ram
>>Which kernel you had used ?
typedef CGAL::Quotient<CGAL::MP_Float> Base_nt;
typedef CGAL::Lazy_exact_nt<Base_nt> Coord_type;
struct Kernel : public CGAL::Cartesian<Coord_type> {};
typedef Kernel::Segment_2 Segment;
typedef Kernel::Point_2 Point_2;
typedef Kernel::Circle_2 Circle;
typedef Kernel::Iso_rectangle_2 Iso_rectangle;
typedef CGAL::Gps_circle_segment_traits_2<Kernel> Traits;
As you see , speed it not optimal , since my polygons in this case have a low count of segments.
(since many of the polygons are 45deg lines etc then the bounding boxes in many cases don't get as effective as wished
this above speed example was a subset of my full job ,,, if I take the full job (14000 polygons , 6 segments) ,, it will take 22 minutes.
Best regards
KimOn Sat, Jun 20, 2009 at 5:14 AM, alex <> wrote:
Hi>> You mention BSP tree - I googled a bit , it seem to be addressed against game development ? am I right ?
>> If you are planing to use cgal for game ? then I think you can drop that idear.No i am not trying to make game :) .. I am working on 3D construction project from 2D images silhoutte.How is your speed results ? Which kernel you had used ?
----- Original Message -----From:To:Sent: Friday, June 19, 2009 8:04 PMSubject: Re: [cgal-discuss] 2D Boolean operationHi Alex
I have absolut no idear what structure they use inside , but it is not as fast as I could hope .
I am struggling with speed issue , but I must say the result coming out is working.
one thing to remember , the reason I try to use the cgal are because of the Curve/arc segment handling otherwise
I think there are other packages where u can gain more speed which support only lines.
You mention BSP tree - I googled a bit , it seem to be addressed against game development ? am I right ?
If you are planing to use cgal for game ? then I think you can drop that idear.
I have static graphic , and it take me very long time to calc.
Best regards
Kim
On Fri, Jun 19, 2009 at 7:39 AM, alex <> wrote:
Dear KimThank you for reply. I have one more question.All this operation is using what kind of method ? I found in Geometeric Tools book about BSP tree. So how faster it is ? Because i have lot of operation iterations.----- Original Message -----From:To:Sent: Thursday, June 18, 2009 8:42 PMSubject: Re: [cgal-discuss] 2D Boolean operationHi
Take a look
http://www.cgal.org/Manual/3.4/doc_html/cgal_manual/Boolean_set_operations_2/Chapter_main.html
Br
Kim
On Thu, Jun 18, 2009 at 1:50 PM, alex <> wrote:
HiIn CGAL other than Nef_2 package for Boolean operation ?Because in Nef_2 Package doesnot offere inexact arithmatic and also no floating points. It only support integer .Correct me if i am wrong.
- [cgal-discuss] 3D Triangulation / surface reconstruction, Jens K. Becker, 06/04/2009
- Re: [cgal-discuss] 3D Triangulation / surface reconstruction, Bo Zhou, 06/06/2009
- Re: [cgal-discuss] 3D Triangulation / surface reconstruction, alex, 06/06/2009
- Re: [cgal-discuss] 3D Triangulation / surface reconstruction, Jens K. Becker, 06/09/2009
- [cgal-discuss] 2D Boolean operation, alex, 06/18/2009
- Re: [cgal-discuss] 2D Boolean operation, Kim, 06/18/2009
- Re: [cgal-discuss] 2D Boolean operation, alex, 06/19/2009
- Re: [cgal-discuss] 2D Boolean operation, Kim, 06/19/2009
- Re: [cgal-discuss] 2D Boolean operation, alex, 06/20/2009
- Message not available
- Re: [cgal-discuss] 2D Boolean operation, Kim, 06/21/2009
- Re: [cgal-discuss] 2D Boolean operation, Kim, 06/19/2009
- Re: [cgal-discuss] 2D Boolean operation, alex, 06/19/2009
- Re: [cgal-discuss] 2D Boolean operation, Kim, 06/18/2009
- Re: [cgal-discuss] 3D Triangulation / surface reconstruction, alex, 06/06/2009
- Re: [cgal-discuss] 3D Triangulation / surface reconstruction, Bo Zhou, 06/06/2009
- Re: [cgal-discuss] 3D Triangulation / surface reconstruction, Jens K. Becker, 06/09/2009
Archive powered by MHonArc 2.6.16.