Sorry Christophe. You are right. The complete second elements are produced in *Aster* from incomplete ones. When writing result files, we turn the quad9 into quad8.<div><br><div>Another information : a brand new verson 3.0 with more features will be released in the next months.<br>
<div><br></div><div>Good night,</div><div>Nicolas<br><br><div class="gmail_quote">2011/3/13 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 class="im">On 11/03/11 13:08, Karin&NiKo wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Yes they are.<br>
BTW, using incomplete second order elements is OK.<br>
<br>
</blockquote>
<br></div>
Hmm... In the version of MED we have here (MED 2.3.4) this is what's defined in med.h:<br>
<br>
typedef enum {MED_POINT1=1, MED_SEG2=102, MED_SEG3=103, MED_TRIA3=203,<br>
MED_QUAD4=204, MED_TRIA6=206,MED_QUAD8=208,MED_TETRA4=304,<br>
MED_PYRA5=305, MED_PENTA6=306, MED_HEXA8=308, MED_TETRA10=310,<br>
MED_PYRA13=313, MED_PENTA15=315, MED_HEXA20=320,<br>
MED_POLYGONE=400, MED_POLYEDRE=500, MED_NONE=0} med_geometrie_element;<br>
<br>
Should we switch to a newer version?<br>
<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">
Nicolas<br>
<br>
2011/3/11 Christophe Geuzaine <<a href="mailto:cgeuzaine@ulg.ac.be" target="_blank">cgeuzaine@ulg.ac.be</a><br></div><div class="im">
<mailto:<a href="mailto:cgeuzaine@ulg.ac.be" target="_blank">cgeuzaine@ulg.ac.be</a>>><br>
<br>
On 11/03/11 09:47, Karin&NiKo wrote:<br>
<br>
Dear Gmsh-ers,<br>
<br>
I would like to report a bug that appears with gmsh 2.5.0 and<br>
also in<br>
the nightly build.<br>
When trying to write the mesh generated by the attached script, a<br>
segmentation fault occurs :<br>
<br>
/opt/gmsh-2.5.1-svn-Linux/bin/gmsh -2 -format med -o carre-Q2.mmed<br>
carre.geo<br>
<br>
Please notice that :<br>
- no problem occurs when using a first order mesh<br>
- the unv format for the 2nd order mesh can be written but, when<br>
read<br>
back in gmsh, the mesh is wrong<br>
<br>
<br>
Hi Nico - Indeed, it looks like 9-node quads I/O is not implemented<br>
for MED (nor UNV). I don't remember why I did not implement this...<br>
Are 9-node quads supported by MED?<br>
<br>
<br>
Nicolas<br>
<br>
<br>
<br>
_______________________________________________<br>
gmsh mailing list<br></div>
<a href="mailto:gmsh@geuz.org" target="_blank">gmsh@geuz.org</a> <mailto:<a href="mailto:gmsh@geuz.org" target="_blank">gmsh@geuz.org</a>><div class="im"><br>
<a href="http://www.geuz.org/mailman/listinfo/gmsh" target="_blank">http://www.geuz.org/mailman/listinfo/gmsh</a><br>
<br>
<br>
<br>
--<br>
Prof. Christophe Geuzaine<br>
University of Liege, Electrical Engineering and Computer Science<br>
<a href="http://www.montefiore.ulg.ac.be/~geuzaine" target="_blank">http://www.montefiore.ulg.ac.be/~geuzaine</a><br></div>
<<a href="http://www.montefiore.ulg.ac.be/%7Egeuzaine" target="_blank">http://www.montefiore.ulg.ac.be/%7Egeuzaine</a>><br>
<br>
<br>
</blockquote><div><div></div><div class="h5">
<br>
<br>
-- <br>
Prof. Christophe Geuzaine<br>
University of Liege, Electrical Engineering and Computer Science<br>
<a href="http://www.montefiore.ulg.ac.be/~geuzaine" target="_blank">http://www.montefiore.ulg.ac.be/~geuzaine</a><br>
</div></div></blockquote></div><br></div></div></div>