<div class="gmail_quote">2010/8/3 Christophe Geuzaine <span dir="ltr"><<a href="mailto:cgeuzaine@ulg.ac.be">cgeuzaine@ulg.ac.be</a>></span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div><div></div><div class="h5">On 31/07/10 20:31, Alexey Vasyukov wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi all.<br>
<br>
I'm trying to use gmsh to construct 3D mesh from initial 2D surface<br>
mesh. With some initial surface meshes it works and with others gmsh<br>
crashes or hangs. You can find example meshes and steps to reproduce<br>
problems in Trac - <a href="https://geuz.org/trac/gmsh/ticket/88" target="_blank">https://geuz.org/trac/gmsh/ticket/88</a><br>
<br>
Could anybody point me in right direction how to fix it?<br>
<br>
</blockquote>
<br></div></div>
I've looked at your files a bit more closely. They all work OK with the Delaunay. They cannot work with the Frontal algorithm because the surface mesh should be oriented with outward pointing normals in that case. Gmsh tries to orient the mesh correctly, but it does so "surface by surface". In your example the boundary of the volume is a single surface ("1") with non-coherent triangle orientations, so Gmsh cannot orient the triangles correctly.</blockquote>
<div><br></div><div><br></div><div>I fixed normals orientation in input mesh and tried version 2.4.2 (binary, Linux) and nightly builds of 2.5.0. I got strange result. If I launch GUI and start 3D meshing from GUI - everything works fine. But if I launch it from console (gmsh tetr-element-5.geo -3) - it fails with the same error as I reported to Track.</div>
<div><br></div><div>Could you please point me what I am doing wrong? How can I make console mode produce the same result that GUI mode produces?</div><div><br></div><div><br></div><div>Thanks in advance for your time and help.</div>
<div><br></div><div>Best regards,</div><div>Alexey Vasyukov</div></div>