# [cig-commits] r15720 - doc/geodynamics.org/benchmarks/trunk/geodyn

luis at geodynamics.org luis at geodynamics.org
Wed Sep 30 15:07:27 PDT 2009

Author: luis
Date: 2009-09-30 15:07:25 -0700 (Wed, 30 Sep 2009)
New Revision: 15720

Modified:
doc/geodynamics.org/benchmarks/trunk/geodyn/index.html
doc/geodynamics.org/benchmarks/trunk/geodyn/index.rst
Log:
Fixes to geodyn

Modified: doc/geodynamics.org/benchmarks/trunk/geodyn/index.html
===================================================================
--- doc/geodynamics.org/benchmarks/trunk/geodyn/index.html	2009-09-30 22:07:17 UTC (rev 15719)
+++ doc/geodynamics.org/benchmarks/trunk/geodyn/index.html	2009-09-30 22:07:25 UTC (rev 15720)
@@ -296,13 +296,13 @@
electrical insulators and the magnetic field on the boundaries matches with
appropriate potential fields in the exterior that imply no external sources
of the field.</p>
-<p>In both cases the Ekman number is $E = 10^{3}$ and the Prandtl number is
-$Pr = 1$. The Rayleigh number is set to $Ra = 100000$. Note that the
+<p>In both cases the Ekman number is [;E = 10^{3};] and the Prandtl number is
+[;Pr = 1;]. The Rayleigh number is set to [;Ra = 100000;]. Note that the
definition of the Rayleigh number differs from the one in the published
-cases [6] by a factor of Ekman number, i.e., $Ra=frac{Ra}{E}$. The
+cases [6] by a factor of Ekman number, i.e., [;Ra=frac{Ra}{E};]. The
magnetic Prandtl number is zero in the non-magnetic convection case 0, and
-is $Pm = 5$ in case 1. The spherical harmonic expansion is truncated at
-degree $ell_{max}=32$ and a four-fold symmetry is assumed in the
+is [;Pm = 5;] in case 1. The spherical harmonic expansion is truncated at
+degree [;]ell_{max}=32;] and a four-fold symmetry is assumed in the
longitudinal direction (<tt class="docutils literal"><span class="pre">param.f</span></tt> should be linked to <tt class="docutils literal"><span class="pre">param32s4.f</span></tt>
when you compile MAG). The input parameter files are <tt class="docutils literal"><span class="pre">par.bench0</span></tt>
for case 0, and <tt class="docutils literal"><span class="pre">par.bench1</span></tt> for case 1; both files reside in the
@@ -315,11 +315,11 @@
relatively short run of MAG</p>
<table border="1" class="docutils">
<colgroup>
-<col width="16%" />
+<col width="17%" />
+<col width="27%" />
+<col width="13%" />
<col width="28%" />
<col width="14%" />
-<col width="28%" />
-<col width="15%" />
</colgroup>
<tbody valign="top">
<tr><td>&nbsp;</td>
@@ -328,27 +328,27 @@
<td>Case 1 Suggested Value</td>
<td>Mag Case 1</td>
</tr>
-<tr><td>$E_{kin}$
-$E_{mag}$
-$T$
-$mu_{phi}$
-$B_{theta}$
-$omega$</td>
-<td><p class="first">$58.348 pm 0.050$</p>
-<p>$0.42812 pm 0.00012$
-$-10.1571 pm 0.0020$</p>
-<p class="last">$0.1824 pm 0.0050$</p>
+<tr><td>[;E_{kin};]
+[;E_{mag};]
+[;T;]
+[;mu_{phi};]
+[;B_{theta};]
+[;omega;]</td>
+<td><p class="first">[;58.348 pm 0.050;]</p>
+<p>[;0.42812 pm 0.00012;]
+[;-10.1571 pm 0.0020;]</p>
+<p class="last">[;0.1824 pm 0.0050;]</p>
</td>
<td><blockquote class="first">
58.35</blockquote>
<p class="last">-10.80</p>
</td>
-<td>$30.733 pm 0.020$
-$626.41 pm 0.40$
-$0.37338 pm 0.00040$
-$-7.6250 pm 0.0060$
-$-4.9289 pm 0.0060$
-$-3.1017 pm 0.0040$</td>
+<td>[;30.733 pm 0.020;]
+[;626.41 pm 0.40;]
+[;0.37338 pm 0.00040;]
+[;-7.6250 pm 0.0060;]
+[;-4.9289 pm 0.0060;]
+[;-3.1017 pm 0.0040;]</td>
<td><p class="first">30.72
627.15</p>
<p class="last">-7.84</p>
@@ -362,9 +362,9 @@
<p>In this benchmark, we produce a magnetic field reversal using MAG. The
input parameter in the source directory for this case is <cite>~/src/par.Rev</cite>.
There is no longitudinal symmetry in this case, so when you compile MAG,
-use <cite>param32s1.f</cite> linking to <cite>param.f</cite>. The Ekman number is $E=0.02$, the
-Prandtl number is $Pr=1$ and the magnetic Prandtl number is $Pm=10$. The
-Rayleigh number is $Ra=12000$.</p>
+use <cite>param32s1.f</cite> linking to <cite>param.f</cite>. The Ekman number is [;E=0.02;], the
+Prandtl number is [;Pr=1;] and the magnetic Prandtl number is [;Pm=10;]. The
+Rayleigh number is [;Ra=12000;].</p>
<div class="section" id="results-and-discussions">
<h2>Results and Discussions</h2>
<p>This case was run on 32-bit and 64-bit Intel processors. Figure

Modified: doc/geodynamics.org/benchmarks/trunk/geodyn/index.rst
===================================================================
--- doc/geodynamics.org/benchmarks/trunk/geodyn/index.rst	2009-09-30 22:07:17 UTC (rev 15719)
+++ doc/geodynamics.org/benchmarks/trunk/geodyn/index.rst	2009-09-30 22:07:25 UTC (rev 15720)
@@ -10,13 +10,13 @@
appropriate potential fields in the exterior that imply no external sources
of the field.

-In both cases the Ekman number is $E = 10^{3}$ and the Prandtl number is
-$Pr = 1$. The Rayleigh number is set to $Ra = 100000$. Note that the
+In both cases the Ekman number is [;E = 10^{3};] and the Prandtl number is
+[;Pr = 1;]. The Rayleigh number is set to [;Ra = 100000;]. Note that the
definition of the Rayleigh number differs from the one in the published
-cases [6] by a factor of Ekman number, i.e., $Ra=\frac{Ra}{E}$. The
+cases [6] by a factor of Ekman number, i.e., [;Ra=\frac{Ra}{E};]. The
magnetic Prandtl number is zero in the non-magnetic convection case 0, and
-is $Pm = 5$ in case 1. The spherical harmonic expansion is truncated at
-degree $\ell_{max}=32$ and a four-fold symmetry is assumed in the
+is [;Pm = 5;] in case 1. The spherical harmonic expansion is truncated at
+degree [;]\ell_{max}=32;] and a four-fold symmetry is assumed in the
longitudinal direction (param.f should be linked to param32s4.f
when you compile MAG). The input parameter files are par.bench0
for case 0, and par.bench1 for case 1; both files reside in the
@@ -29,16 +29,16 @@
and case 1, the values listed were obtained with low resolution and a
relatively short run of MAG

-+--------------+------------------------+------------+------------------------+-------------+
-|              | Case 0 Suggested value | Mag Case 0 | Case 1 Suggested Value | Mag Case 1  |
-+--------------+------------------------+------------+------------------------+-------------+
-| $E_{kin}$    | $58.348 \pm 0.050$     |  58.35     | $30.733 \pm 0.020$     | 30.72       |
-| $E_{mag}$    |                        |            | $626.41 \pm 0.40$      | 627.15      |
-| $T$          | $0.42812 \pm 0.00012$  |            | $0.37338 \pm 0.00040$  |             |
-| $\mu_{\phi}$ | $-10.1571 \pm 0.0020$  | -10.80     | $-7.6250 \pm 0.0060$   | -7.84       |
-| $B_{\theta}$ |                        |            | $-4.9289 \pm 0.0060$   |             |
-| $\omega$     | $0.1824 \pm 0.0050$    |            | $-3.1017 \pm 0.0040$   |             |
-+--------------+------------------------+------------+------------------------+-------------+
++----------------+-------------------------+------------+--------------------------+-------------+
+|                | Case 0 Suggested value  | Mag Case 0 | Case 1 Suggested Value   | Mag Case 1  |
++----------------+-------------------------+------------+--------------------------+-------------+
+| [;E_{kin};]    | [;58.348 \pm 0.050;]    |  58.35     | [;30.733 \pm 0.020;]     | 30.72       |
+| [;E_{mag};]    |                         |            | [;626.41 \pm 0.40;]      | 627.15      |
+| [;T;]          | [;0.42812 \pm 0.00012;] |            | [;0.37338 \pm 0.00040;]  |             |
+| [;\mu_{\phi};] | [;-10.1571 \pm 0.0020;] | -10.80     | [;-7.6250 \pm 0.0060;]   | -7.84       |
+| [;B_{\theta};] |                         |            | [;-4.9289 \pm 0.0060;]   |             |
+| [;\omega;]     | [;0.1824 \pm 0.0050;]   |            | [;-3.1017 \pm 0.0040;]   |             |
++----------------+-------------------------+------------+--------------------------+-------------+

Reversal Dynamo Case
@@ -46,9 +46,9 @@
In this benchmark, we produce a magnetic field reversal using MAG. The
input parameter in the source directory for this case is ~/src/par.Rev.
There is no longitudinal symmetry in this case, so when you compile MAG,
-use param32s1.f linking to param.f. The Ekman number is $E=0.02$, the
-Prandtl number is $Pr=1$ and the magnetic Prandtl number is $Pm=10$. The
-Rayleigh number is $Ra=12000$.
+use param32s1.f linking to param.f. The Ekman number is [;E=0.02;], the
+Prandtl number is [;Pr=1;] and the magnetic Prandtl number is [;Pm=10;]. The
+Rayleigh number is [;Ra=12000;].

Results and Discussions