<p>
If run, the program prints the following output, explaining what it is
-doing all that time:
+doing during all that time:
<code>
<pre>
examples/step-18> time make run
<table width="100%">
<tr width="100%">
- <td>
+ <td width="33%">
<a href="step-18.data/solution-0002.0000.png" target="_top">
<img src="step-18.data/solution-0002.0000.png"
width="100%"></a>
Time = 2
</td>
- <td>
+ <td width="33%">
<a href="step-18.data/solution-0005.0000.png" target="_top">
<img src="step-18.data/solution-0005.0000.png"
width="100%"></a>
Time = 5
</td>
- <td>
+ <td width="33%">
<a href="step-18.data/solution-0007.0000.png" target="_top">
<img src="step-18.data/solution-0007.0000.png"
width="100%"></a>
</tr>
<tr width="100%">
- <td>
+ <td width="33%">
<a href="step-18.data/solution-0008.0000.png" target="_top">
<img src="step-18.data/solution-0008.0000.png"
width="100%"></a>
Time = 8
</td>
- <td>
+ <td width="33%">
<a href="step-18.data/solution-0009.0000.png" target="_top">
<img src="step-18.data/solution-0009.0000.png"
width="100%"></a>
Time = 9
</td>
- <td>
+ <td width="33%">
<a href="step-18.data/solution-0010.0000.png" target="_top">
<img src="step-18.data/solution-0010.0000.png"
width="100%"></a>
forces and boundary deflections are symmetric) but the effect is
probably physically correct anyway: in reality, small inhomogeneities
in the body's material properties would lead it to buckle to one side
-to evade the forcing; in numerical simulation, small perturbations
+to evade the forcing; in numerical simulations, small perturbations
such as numerical round-off or an inexact solution of a linear system
-by an iterative solver could have the same effect.
\ No newline at end of file
+by an iterative solver could have the same effect. Another typical source for
+asymmetries in adaptive computations is that only a certain fraction of cells
+is refined in each step, which may lead to asymmetric meshes even if the
+original coarse mesh was symmetric.