On 13/04/14 16:08, Matthew Knepley wrote:

Thanks for the test. It was a bug in a special case:

-        coords[3] = PetscRealPart(coords[3*dim+0] - coords[0*dim+0]);
+        coords[3] = PetscRealPart(coords[3*dim+1] - coords[0*dim+1]);

I will push it to next

Thanks very much- that's fixed it.

The problems I was seeing last week with VecView() writing to VTK seem to have gone now too. Wunderbar!

Cheers, Adrian

--
Dr Adrian Croucher
Department of Engineering Science
University of Auckland
New Zealand
tel 64-9-373-7599 ext 84611

Reply via email to