]> https://gitweb.dealii.org/ - dealii.git/commitdiff
table layout and image fixes
authorTimo Heister <timo.heister@gmail.com>
Sun, 17 Jan 2016 19:46:27 +0000 (13:46 -0600)
committerTimo Heister <timo.heister@gmail.com>
Sun, 17 Jan 2016 20:29:39 +0000 (14:29 -0600)
- fix weird case where doxygen shifts the last cell in a table downward
causing them to be not correctly aligned
- cleanup some image layout
- do not break image tags (this will break various scripts)

12 files changed:
examples/step-10/doc/results.dox
examples/step-14/doc/results.dox
examples/step-17/doc/results.dox
examples/step-2/doc/results.dox
examples/step-20/doc/results.dox
examples/step-23/doc/results.dox
examples/step-26/doc/results.dox
examples/step-27/doc/results.dox
examples/step-36/doc/results.dox
examples/step-49/doc/intro.dox
examples/step-49/doc/results.dox
examples/step-51/doc/results.dox

index dcc5dac01145ceb900348585861700a1f5b74ae8..cf8631b72eccdd813b976f1fd1734d5f46aa310c 100644 (file)
@@ -25,52 +25,38 @@ The following table shows the triangulated computational domain for
 Q1, Q2, and Q3 mappings, for the original coarse grid (left), and a
 once uniformly refined grid (right). If your browser does not display
 these pictures in acceptable quality, view them one by one.
-<table "width=80%" align="center">
+
+<table style="width:80%" align="center">
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q1_ref0.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q1_ref1.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q1_ref0.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q1_ref1.png" alt=""></td>
   </tr>
 
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q2_ref0.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q2_ref1.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q2_ref0.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q2_ref1.png" alt=""></td>
   </tr>
 
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q3_ref0.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q3_ref1.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q3_ref0.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-10.ball_mapping_q3_ref1.png" alt=""></td>
   </tr>
 </table>
+
 These pictures show the obvious advantage of higher order mappings:
 they approximate the true boundary quite well also on rather coarse
 meshes. To demonstrate this a little further, the following table
 shows the upper right quarter of the circle of the coarse mesh, and
 with dashed lines the exact circle:
-<table "width=80%" align="center">
+
+<table style="width:80%" align="center">
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-10.quarter-q1.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-10.quarter-q2.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-10.quarter-q3.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-10.quarter-q1.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-10.quarter-q2.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-10.quarter-q3.png" alt=""></td>
   </tr>
 </table>
+
 Obviously the quadratic mapping approximates the boundary quite well,
 while for the cubic mapping the difference between approximated domain
 and true one is hardly visible already for the coarse grid. You can
index 6601976f7d56a7665dc69e81db22abaa4b0588db..1cb435b22d72c97630a9a73c22dc33f9768ee46e 100644 (file)
@@ -71,38 +71,21 @@ differently.
 
 
 Looking at the grids that are produced in the course of subsequent
-refinement, these are some of them:
-<table align="center">
-  <tr>
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-0.png" alt="">
-    </td>
-
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-2.png" alt="">
-    </td>
-  </tr>
+refinement, here are some of them:
 
+<table width="80%" align="center">
   <tr>
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-4.png" alt="">
-    </td>
-
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-5.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-0.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-2.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-4.png" alt="" width="100%"></td>
   </tr>
-
   <tr>
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-7.png" alt="">
-    </td>
-
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-8.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-5.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-7.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-value.grid-8.png" alt="" width="100%"></td>
   </tr>
 </table>
+
 Note the subtle interplay between resolving the corner singularities,
 and resolving around the point of evaluation. It will be rather
 difficult to generate such a mesh by hand, as this would involve to
@@ -141,15 +124,10 @@ close to each other, even for such a complicated quantity as the point
 value:
 
 
-<table align="center">
+<table width="80%" align="center">
   <tr>
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-value.error.png" alt="">
-    </td>
-
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-value.error-estimation.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-value.error.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-value.error-estimation.png" alt="" width="100%"></td>
   </tr>
 </table>
 
@@ -255,37 +233,20 @@ evaluation shows this:
 </table>
 This time, the grids in refinement cycles 0, 5, 6, 7, 8, and 9 look
 like this:
-<table align="center">
-  <tr>
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-0.png" alt="">
-    </td>
-
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-5.png" alt="">
-    </td>
-  </tr>
 
+<table align="center" width="80%">
   <tr>
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-6.png" alt="">
-    </td>
-
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-7.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-0.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-5.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-6.png" alt="" width="100%"></td>
   </tr>
-
   <tr>
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-8.png" alt="">
-    </td>
-
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-9.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-7.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-8.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.grid-9.png" alt="" width="100%"></td>
   </tr>
 </table>
+
 Note the asymmetry of the grids compared with those we obtained for
 the point evaluation, which is due to the directionality of the
 x-derivative for which we tailored the refinement criterion.
@@ -306,17 +267,14 @@ in the result.
 In the left part of the following chart, you again see the convergence
 of the error towards this extrapolated value, while on the right you
 see a comparison of true and estimated error:
-<table align="center">
-  <tr>
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.error.png" alt="">
-    </td>
 
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.error-estimation.png" alt="">
-    </td>
+<table width="80%" align="center">
+  <tr>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.error.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.point-derivative.error-estimation.png" alt="" width="100%"></td>
   </tr>
 </table>
+
 After an initial phase where the true error changes its sign, the
 estimated error matches it quite well, again. Also note the dramatic
 improvement in the error when using the estimated error to correct the
@@ -337,29 +295,26 @@ are better than those we had previously.
 
 First, the meshes after 9 and 10 adaptive refinement cycles,
 respectively, look like this:
-<table align="center">
-  <tr>
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.step-13.grid-9.png" alt="">
-    </td>
 
-    <td width="50%">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.step-13.grid-10.png" alt="">
-    </td>
+<table width="80%" align="center">
+  <tr>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.step-13.grid-9.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.step-13.grid-10.png" alt="" width="100%"></td>
   </tr>
 </table>
+
 The features of the solution can still be seen slightly, but since the
 solution is smooth, the roughness of the dual solution entirely
 dominates the mesh refinement criterion, and leads to strongly
 concentrated meshes. The solution after the seventh refinement step is
 like so:
-<table align="center">
+
+<table width="80%" align="center">
   <tr>
-    <td width="50%" align="center">
-      <img src="http://www.dealii.org/images/steps/developer/step-14.step-13.solution-7.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-14.step-13.solution-7.png" alt="" width="100%"></td>
   </tr>
 </table>
+
 Obviously, the solution is worse at some places, but the mesh
 refinement process should have taken care that these places are not
 important for computing the point value.
index 4da8e77bd82c23992ddc246e1563b1bfbb759ecd..19c56e1a3e6499541fed46d66d29c3f8de83223e 100644 (file)
@@ -120,19 +120,13 @@ entries locally or not.
 Here is some output generated in the 12th cycle of the program, i.e. with roughly
 300,000 unknowns:
 
-<table>
+<table align="center" style="width:80%">
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-17.12-ux.png" alt="" width="45%">
-    </td>
-    <td>
-         <img src="http://www.dealii.org/images/steps/developer/step-17.12-uy.png" alt="" width="45%">
-       </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-17.12-ux.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-17.12-uy.png" alt="" width="100%"></td>
   </tr>
 </table>
 
-
-
 As one would hope for, the x- (left) and y-displacements (right) shown here
 closely match what we already saw in step-8. As shown
 there and in step-22, we could as well have produced a
@@ -140,19 +134,13 @@ vector plot of the displacement field, rather than plotting it as two
 separate scalar fields. What may be more interesting,
 though, is to look at the mesh and partition at this step:
 
-
-<table>
+<table align="center" width="80%">
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-17.12-grid.png" alt="" width="45%">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-17.12-partition.png" alt="" width="45%">
-       </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-17.12-grid.png" alt="" width="100%"></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-17.12-partition.png" alt="" width="100%"></td>
   </tr>
 </table>
 
-
 Again, the mesh (left) shows the same refinement pattern as seen
 previously. The right panel shows the partitioning of the domain across the 8
 processes, each indicated by a different color. The picture shows that the
@@ -229,14 +217,10 @@ graphical output generated by
 this job is rather large (cycle 5 already prints around 82 MB of data), so
 we contend ourselves with showing output from cycle 4:
 
-<table>
+<table width="80%" align="center">
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-17.4-3d-partition.png" alt="" width="45%">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-17.4-3d-ux.png" alt="" width="45%">
-       </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-17.4-3d-partition.png" width="100%" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-17.4-3d-ux.png" alt="" width="100%"></td>
   </tr>
 </table>
 
index 9066dbb5b7bad60c4a5f48b444621d801a0c964c..6fde7c7cc2cc1f2f709fcaa517aa8d8ebb5fb522 100644 (file)
@@ -9,15 +9,10 @@ zero or not depends on the equation under consideration, but the
 indicated positions in the matrix tell us which shape functions can
 and which can't couple when discretizing a local, i.e. differential,
 equation):
-<TABLE WIDTH="60%" ALIGN="center">
+<table style="width:60%" align="center">
   <tr>
-    <td ALIGN="center">
-      <img src="http://www.dealii.org/images/steps/developer/step-2.sparsity-1.svg" alt="">
-    </td>
-
-    <td ALIGN="center">
-      <img src="http://www.dealii.org/images/steps/developer/step-2.sparsity-2.svg" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-2.sparsity-1.svg" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-2.sparsity-2.svg" alt=""></td>
   </tr>
 </table>
 
index 20ece4a3e2eb3747ae5e95dff6a91b71cd08ee7a..1dcd7b176f5c6b080cc37acc4f489ae6bdc31df8 100644 (file)
@@ -219,10 +219,12 @@ Remember that the function returns the inverse of the permeability tensor.
 With a significantly higher mesh resolution, we can visualize this, here with
 x- and y-velocity:
 
-
-<img src="http://www.dealii.org/images/steps/developer/step-20.u-wiggle.png" alt="">
-<img src="http://www.dealii.org/images/steps/developer/step-20.v-wiggle.png" alt="">
-
+<table style="width:60%" align="center">
+  <tr>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-20.u-wiggle.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-20.v-wiggle.png" alt=""></td>
+  </tr>
+</table>
 
 It is obvious how fluids flow essentially only along the middle line, and not
 anywhere else.
@@ -298,9 +300,12 @@ looks as follows:
 With a permeability field like this, we would get x-velocities and pressures as
 follows:
 
-
-<img src="http://www.dealii.org/images/steps/developer/step-20.u-random.png" alt="">
-<img src="http://www.dealii.org/images/steps/developer/step-20.p-random.png" alt="">
+<table style="width:60%" align="center">
+  <tr>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-20.u-random.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-20.p-random.png" alt=""></td>
+  </tr>
+</table>
 
 We will use these permeability fields again in step-21 and step-43.
 
index 1c86e3eb7d36a91c8691fb72dad7df5be8fabb14..f630d8b8cbdf936807633577fe072457ad3fdece 100644 (file)
@@ -65,9 +65,7 @@ In addition to the screen output, the program writes the solution of each time
 step to an output file. If we process them adequately and paste them into a
 movie, we get the following:
 
-<img
-src="http://www.dealii.org/images/steps/developer/step-23.movie.gif"
-alt="Animation of the solution of step 23.">
+<img src="http://www.dealii.org/images/steps/developer/step-23.movie.gif" alt="Animation of the solution of step 23.">
 
 The movie shows the generated wave nice traveling through the domain and back,
 being reflected at the clamped boundary. Some numerical noise is trailing the
index 58a105902d162b838b95c3da086cec7bb9769e2a..92c5396721a8f4568c3b9021382b20851f3fa059 100644 (file)
@@ -41,9 +41,7 @@ Number of degrees of freedom: 2109
 Maybe of more interest is a visualization of the solution and the mesh on which
 it was computed:
 
-<img
-src="http://www.dealii.org/images/steps/developer/step-26.movie.gif"
-alt="Animation of the solution of step 26.">
+<img src="http://www.dealii.org/images/steps/developer/step-26.movie.gif" alt="Animation of the solution of step 26.">
 
 The movie shows how the two sources switch on and off and how the mesh reacts
 to this. It is quite obvious that the mesh as is is probably not the best we
index bb721c292ebfdc12b7d287041cd0adee47223b2c..bc82a7067b7a8113bf01ee6f936f07cd0a62962b 100644 (file)
@@ -58,57 +58,34 @@ solution of the problem:
 
 Secondly, let us look at the sequence of meshes generated:
 
-<table align="center" border="1" cellspacing="3" cellpadding="3">
+<table width="60%" align="center">
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.mesh-0.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.mesh-1.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.mesh-2.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.mesh-0.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.mesh-1.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.mesh-2.png" alt=""></td>
   </tr>
-
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.mesh-3.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.mesh-4.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.mesh-5.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.mesh-3.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.mesh-4.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.mesh-5.png" alt=""></td>
   </tr>
 </table>
+
 It is clearly visible how the mesh is refined near the corner singularities,
 as one would expect it. More interestingly, we should be curious to see the
 distribution of finite element polynomial degrees to these mesh cells:
-<table align="center" border="1" cellspacing="3" cellpadding="3">
+
+<table width="60%" align="center">
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-0.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-1.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-2.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-0.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-1.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-2.png" alt=""></td>
   </tr>
 
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-3.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-4.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-5.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-3.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-4.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.fe_degree-5.png" alt=""></td>
   </tr>
 </table>
 
@@ -123,29 +100,17 @@ This arrangement of polynomial degrees of course follows from our smoothness
 estimator. Here is the estimated smoothness of the solution, with blue colors
 indicating least smoothness and red indicating the smoothest areas:
 
-<table align="center" border="1" cellspacing="3" cellpadding="3">
+<table width="60%" align="center">
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-0.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-1.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-2.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-0.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-1.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-2.png" alt=""></td>
   </tr>
 
   <tr>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-3.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-4.png" alt="">
-    </td>
-    <td>
-      <img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-5.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-3.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-4.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-27.smoothness-5.png" alt=""></td>
   </tr>
 </table>
 
index 0644023a6b879effb8f96688602ee469c3821b0b..dfe659761aaa0b766e59d4f61b8af5864cebe47b 100644 (file)
@@ -34,31 +34,17 @@ correspond to pairs $(m,n)=(1,1)$, $(1,2)$ and $(2,1)$, $(2,2)$, and
 $(3,1)$. A visualization of the corresponding eigenfunctions would
 look like this:
 
-<TABLE WIDTH="100%">
+<table width="80%">
 <tr>
-<td>
-  <img src="http://www.dealii.org/images/steps/developer/step-36.default.eigenfunction.0.png" alt="">
-</td>
-<td>
-  <img src="http://www.dealii.org/images/steps/developer/step-36.default.eigenfunction.1.png" alt="">
-</td>
+<td><img src="http://www.dealii.org/images/steps/developer/step-36.default.eigenfunction.0.png" alt=""></td>
+<td><img src="http://www.dealii.org/images/steps/developer/step-36.default.eigenfunction.1.png" alt=""></td>
+<td><img src="http://www.dealii.org/images/steps/developer/step-36.default.eigenfunction.2.png" alt=""></td>
 </tr>
 
 <tr>
-<td>
-  <img src="http://www.dealii.org/images/steps/developer/step-36.default.eigenfunction.2.png" alt="">
-</td>
-<td>
-  <img src="http://www.dealii.org/images/steps/developer/step-36.default.eigenfunction.3.png" alt="">
-</td>
-</tr>
-
-<tr>
-<td>
-  <img src="http://www.dealii.org/images/steps/developer/step-36.default.eigenfunction.4.png" alt="">
-</td>
-<td>
-</td>
+<td><img src="http://www.dealii.org/images/steps/developer/step-36.default.eigenfunction.3.png" alt=""></td>
+<td><img src="http://www.dealii.org/images/steps/developer/step-36.default.eigenfunction.4.png" alt=""></td>
+<td></td>
 </tr>
 </table>
 
@@ -120,31 +106,17 @@ circle of the potential):
 
 The first five eigenfunctions are now like this:
 
-<TABLE WIDTH="100%">
-<tr>
-<td>
-  <img src="http://www.dealii.org/images/steps/developer/step-36.mod.eigenfunction.0.png" alt="">
-</td>
-<td>
-  <img src="http://www.dealii.org/images/steps/developer/step-36.mod.eigenfunction.1.png" alt="">
-</td>
-</tr>
-
+<table width="80%">
 <tr>
-<td>
-  <img src="http://www.dealii.org/images/steps/developer/step-36.mod.eigenfunction.2.png" alt="">
-</td>
-<td>
-  <img src="http://www.dealii.org/images/steps/developer/step-36.mod.eigenfunction.3.png" alt="">
-</td>
+<td><img src="http://www.dealii.org/images/steps/developer/step-36.mod.eigenfunction.0.png" alt=""></td>
+<td><img src="http://www.dealii.org/images/steps/developer/step-36.mod.eigenfunction.1.png" alt=""></td>
+<td><img src="http://www.dealii.org/images/steps/developer/step-36.mod.eigenfunction.2.png" alt=""></td>
 </tr>
 
 <tr>
-<td>
-  <img src="http://www.dealii.org/images/steps/developer/step-36.mod.eigenfunction.4.png" alt="">
-</td>
-<td>
-</td>
+<td><img src="http://www.dealii.org/images/steps/developer/step-36.mod.eigenfunction.3.png" alt=""></td>
+<td><img src="http://www.dealii.org/images/steps/developer/step-36.mod.eigenfunction.4.png" alt=""></td>
+<td></td>
 </tr>
 </table>
 
index f6b14ff9ad643e648674a917fb37f05296e79a92..83f3239b767bbd90e024735f3ae4a16aa96dbffd 100644 (file)
@@ -147,6 +147,7 @@ given mesh using a smooth function. An example of its use is also given in the
 results section of step-38 but let us show a simpler example here:
 In the function <code>grid_5()</code> of the current program, we perturb the y
 coordinate of a mesh with a sine curve:
+
 <TABLE WIDTH="60%" ALIGN="center">
   <tr>
     <td ALIGN="center">
@@ -176,6 +177,7 @@ of this tutorial:
 Finally, the function GridTools::distort_random allows you to move vertices in the
 mesh (optionally ignoring boundary nodes) by a random amount. This is
 demonstrated in <code>grid_7()</code> and the result is as follows:
+
 <TABLE WIDTH="60%" ALIGN="center">
   <tr>
     <td ALIGN="center">
@@ -211,20 +213,12 @@ to make sure that cells line up correctly and no unpaired nodes exist in the
 merged Triangulation.
 
 These are the input meshes and the output mesh:
-<TABLE WIDTH="80%" ALIGN="center">
-  <tr>
-    <td ALIGN="center">
-        <img
-       src="http://www.dealii.org/images/steps/developer/step-49.grid-2a.png"
-       alt="" height="200px"> input mesh 1
-    </td>
-    <td ALIGN="center">
-        <img src="http://www.dealii.org/images/steps/developer/step-49.grid-2b.png" alt="" height="200px"> input mesh 2
-    </td>
 
-    <td ALIGN="center">
-        <img src="http://www.dealii.org/images/steps/developer/step-49.grid-2.png" alt="" height="200px"> merged mesh
-    </td>
+<table width="80%" align="center">
+  <tr>
+    <td ALIGN="center"><img src="http://www.dealii.org/images/steps/developer/step-49.grid-2a.png" alt="" height="200px">input mesh 1</td>
+    <td ALIGN="center"><img src="http://www.dealii.org/images/steps/developer/step-49.grid-2b.png" alt="" height="200px">input mesh 2</td>
+    <td ALIGN="center"><img src="http://www.dealii.org/images/steps/developer/step-49.grid-2.png" alt="" height="200px">merged mesh</td>
   </tr>
 </table>
 
index 31824d9630427b40b8a2f79c37b92bb09e32bada..ed20909c73642a78de57d2c1e91b6300fb397125 100644 (file)
@@ -16,18 +16,14 @@ the coarse mesh. Several of the meshes shown in the introduction section fall
 into this category. For example, for this mesh the central hole is supposed to
 be round:
 
-<img
- src="http://www.dealii.org/images/steps/developer/step-49.grid-2a.png"
- alt="" height="200px">
+<img src="http://www.dealii.org/images/steps/developer/step-49.grid-2a.png" alt="" height="200px">
 
 On the other hand, if you simply refine it, the Triangulation class can not
 know whether you wanted the hole to be round or to be an octagon. The default
 is to place new points along existing edges. After two mesh refinement steps,
 this would yield the following mesh, which is not what we wanted:
 
-<img
- src="http://www.dealii.org/images/steps/developer/step-49.grid-2d-refined.png"
- alt="" height="200px">
+<img src="http://www.dealii.org/images/steps/developer/step-49.grid-2d-refined.png" alt="" height="200px">
 
 What needs to happen is that you tell the triangulation that you in fact want
 to use a curved boundary. The way to do this requires three steps:
@@ -50,9 +46,7 @@ The goal was to generate (and use) a geometry that describes a
 microstructured electric device. In a CAD program, the geometry looks like
 this:
 
-<img
- src="http://www.dealii.org/images/steps/developer/step-49.yuhan.1.png"
- alt="">
+<img src="http://www.dealii.org/images/steps/developer/step-49.yuhan.1.png" alt="">
 
 In the following, we will walk you through the entire process of creating a
 mesh for this geometry, including a number of common pitfalls by showing the
@@ -233,9 +227,7 @@ void create_3d_grid (Triangulation<3> &triangulation)
 
 With this code, you get a mesh that looks like this:
 
-<img
- src="http://www.dealii.org/images/steps/developer/step-49.yuhan.2.png"
- alt="">
+<img src="http://www.dealii.org/images/steps/developer/step-49.yuhan.2.png" alt="">
 
 The next step is to teach each of the top surfaces that they should be
 curved. We can do this by creating CylinderBoundary objects that
@@ -291,9 +283,7 @@ describe this. A first attempt looks like this:
 
 With this code, we get a mesh that looks like this:
 
-<img
- src="http://www.dealii.org/images/steps/developer/step-49.yuhan.3.png"
- alt="">
+<img src="http://www.dealii.org/images/steps/developer/step-49.yuhan.3.png" alt="">
 
 This is clearly not correct: The new vertices that have been entered at
 mid-edge and mid-face points are not where they should have been. Upon some
@@ -315,9 +305,7 @@ axis:
 
 This yields an improvement, though it is still not quite correct:
 
-<img
- src="http://www.dealii.org/images/steps/developer/step-49.yuhan.4.png"
- alt="">
+<img src="http://www.dealii.org/images/steps/developer/step-49.yuhan.4.png" alt="">
 
 Looking closely at this mesh, we realize that the new points on mid-face
 vertices are where they should be, though the new vertices inserted at
@@ -329,9 +317,7 @@ easily fixed by using the function TriaAccessor::set_all_boundary_ids()
 instead of TriaAccessor::set_boundary_id() used above. With this change,
 the grid now looks like this:
 
-<img
- src="http://www.dealii.org/images/steps/developer/step-49.yuhan.5.png"
- alt="">
+<img src="http://www.dealii.org/images/steps/developer/step-49.yuhan.5.png" alt="">
 
 This is already better. However, something is still going wrong on the
 front left face. On second look, we can also see that the faces where
@@ -386,15 +372,11 @@ code does the trick:
 
 With this, we finally get a mesh that looks good:
 
-<img
- src="http://www.dealii.org/images/steps/developer/step-49.yuhan.6.png"
- alt="">
+<img src="http://www.dealii.org/images/steps/developer/step-49.yuhan.6.png" alt="">
 
 We can then refine the mesh two more times to see in more detail what
 happens to the curved part of the boundary:
 
- <img
- src="http://www.dealii.org/images/steps/developer/step-49.yuhan.7.png"
- alt="">
+ <img src="http://www.dealii.org/images/steps/developer/step-49.yuhan.7.png" alt="">
 
  So, yes!, this is finally what we were looking for!
index 28300a4d9d4ffb8de1cfaac299f8704d55b304d6..06ef541ea1264865eb283973f32b131caa5b529f 100644 (file)
@@ -31,22 +31,14 @@ the interior variables do not exactly satisfy boundary conditions. On the
 lower and left boundaries, we set Neumann boundary conditions, whereas we set
 Dirichlet conditions on the right and top boundaries.
 
-<table align="center" border="1" cellspacing="3" cellpadding="3">
+<table align="center">
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.sol_2.png" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.sol_3.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.sol_2.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.sol_3.png" alt=""></td>
   </tr>
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.sol_4.png" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.sol_8.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.sol_4.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.sol_8.png" alt=""></td>
   </tr>
 </table>
 
@@ -57,22 +49,14 @@ cycle two, it looks much better for cycles three and four. As shown by the
 convergence table below, we find that is also converges more quickly to the
 analytical solution.
 
-<table align="center" border="1" cellspacing="3" cellpadding="3">
+<table align="center">
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.post_2.png" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.post_3.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.post_2.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.post_3.png" alt=""></td>
   </tr>
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.post_4.png" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.post_8.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.post_4.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.post_8.png" alt=""></td>
   </tr>
 </table>
 
@@ -82,14 +66,10 @@ to the linear solution (not post-processed) at cycle 8 with 4,096
 cells. This clearly shows the superiority of high order methods for smooth
 solutions.
 
-<table align="center" border="1" cellspacing="3" cellpadding="3">
+<table align="center">
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.sol_q3_2.png" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.post_q3_2.png" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.sol_q3_2.png" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.post_q3_2.png" alt=""></td>
   </tr>
 </table>
 
@@ -218,14 +198,10 @@ TrilinosWrappers::PreconditionAMG. For the HDG part, a wrapper around
 ChunkSparseMatrix for the trace variable has been used in order to utilize the
 block structure in the matrix on the finest level.
 
-<table align="center" border="1" cellspacing="3" cellpadding="3">
+<table align="center">
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.2d_plain.png" width="400" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.2dt_plain.png" width="400" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.2d_plain.png" width="400" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.2dt_plain.png" width="400" alt=""></td>
   </tr>
 </table>
 
@@ -239,14 +215,10 @@ by $p=1^*$ for example). We now see a clear advantage of HDG for the same
 amount of work for both $p=3$ and $p=6$, and about the same quality
 for $p=1$.
 
-<table align="center" border="1" cellspacing="3" cellpadding="3">
+<table align="center">
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.2d_post.png" width="400" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.2dt_post.png" width="400" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.2d_post.png" width="400" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.2dt_post.png" width="400" alt=""></td>
   </tr>
 </table>
 
@@ -257,14 +229,10 @@ $p+1$. If we do this, we get the convergence curves below. We see that
 CG with second order polynomials is again clearly better than HDG with
 linears. However, the advantage of HDG for higher orders remains.
 
-<table align="center" border="1" cellspacing="3" cellpadding="3">
+<table align="center">
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.2d_postb.png" width="400" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.2dt_postb.png" width="400" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.2d_postb.png" width="400" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.2dt_postb.png" width="400" alt=""></td>
   </tr>
 </table>
 
@@ -294,30 +262,18 @@ degree $p$. There are fewer degrees of freedom on the skeleton variable
 for FE_FaceP for a given mesh size, but the solution quality (error vs. number
 of DoFs) is very similar to the results for FE_FaceQ.
 
-<table align="center" border="1" cellspacing="3" cellpadding="3">
+<table align="center">
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.3d_plain.png" width="400" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.3dt_plain.png" width="400" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.3d_plain.png" width="400" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.3dt_plain.png" width="400" alt=""></td>
   </tr>
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.3d_post.png" width="400" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.3dt_post.png" width="400" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.3d_post.png" width="400" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.3dt_post.png" width="400" alt=""></td>
   </tr>
   <tr>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.3d_postb.png" width="400" alt="">
-    </td>
-    <td>
-        <img src="http://www.dealii.org/images/steps/developer/step-51.3dt_postb.png" width="400" alt="">
-    </td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.3d_postb.png" width="400" alt=""></td>
+    <td><img src="http://www.dealii.org/images/steps/developer/step-51.3dt_postb.png" width="400" alt=""></td>
   </tr>
 </table>
 

In the beginning the Universe was created. This has made a lot of people very angry and has been widely regarded as a bad move.

Douglas Adams


Typeset in Trocchi and Trocchi Bold Sans Serif.