tag:blogger.com,1999:blog-3285677929777490656.post3999835786903897914..comments2023-10-31T13:16:01.375+03:00Comments on Open CASCADE notes: All-inclusive bounding boxRoman Lyginhttp://www.blogger.com/profile/18338419158437898791noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-3285677929777490656.post-34444419275745667322009-04-27T09:48:00.000+04:002009-04-27T09:48:00.000+04:00Hi Roman,
I've been following your notes for some...Hi Roman,<br /><br />I've been following your notes for some time. In my opinion "Topology and Geometry in Open Cascade" is an excellent, although short, introduction into Open Cascade concepts. I remarked you spent some time on figuring out (some of) the (speed-wise) bottlenecks of the boolean operations. As there are many posts on the OCC forum regarding the speed/robustness of the OCC booleans, I must admit I'm surprised nobody is interested in non-regularized booleans. I played with the ACIS kernel about one year ago. As I recall non-regularized booleans can be very conveniently performed with ACIS whereas no such option exists in OCC. Open Cascade is around for a while, therefore I really doubt nobody (except me :) ) was interested in these non-regularized booleans so far. I work in the field of numerical modeling of microwave circuits by BEM/FEM. The topology/geometry of the structures I need to model is relatively simple, thus I emulated the non-regularized booleans I needed (sometimes building the topology, "by hand" from vertex to solid). <br />Could you shed some light on this (regularized vs. non-regularized) matter ?<br /><br />Vali CatinaUnknownhttps://www.blogger.com/profile/11443940891178233656noreply@blogger.comtag:blogger.com,1999:blog-3285677929777490656.post-68351471192207201302009-04-17T23:12:00.000+04:002009-04-17T23:12:00.000+04:00Hi Roman,
I remember having a similar problem, wh...Hi Roman,<br /><br />I remember having a similar problem, which I can't reproduce now... I was constructing a face using BRepOffsetAPI_ThruSections based on (opened) spline curves and wanted to compute the bounding box for it. The BB was much bigger than the face itself and I guess I wasn't displaying the shape.<br /><br />PawelPawelhttps://www.blogger.com/profile/08979312019226548943noreply@blogger.comtag:blogger.com,1999:blog-3285677929777490656.post-27902105117018024492009-04-17T10:35:00.000+04:002009-04-17T10:35:00.000+04:00Maybe it would be better to calculate the first ap...Maybe it would be better to calculate the first approximate bounding box with a really coarse triangulation, that gives anyway a good bounding box. Or just use the wires to limit the boundingbox of a face.<br /><br />In effect is a bit strange :)<br /><br />There a lot of things that could be improved in OpenCascade, but the community is not really involved in that!QbPrognoreply@blogger.com