[cig-commits] r12197 - long/3D/Gale/trunk/documentation

[sue at geodynamics.org]

Author: sue
Date: 2008-06-13 15:16:29 -0700 (Fri, 13 Jun 2008)
New Revision: 12197

Modified:
   long/3D/Gale/trunk/documentation/gale.lyx
Log:
added back times signs

Modified: long/3D/Gale/trunk/documentation/gale.lyx
===================================================================
--- long/3D/Gale/trunk/documentation/gale.lyx	2008-06-13 21:59:40 UTC (rev 12196)
+++ long/3D/Gale/trunk/documentation/gale.lyx	2008-06-13 22:16:29 UTC (rev 12197)
@@ -1,4 +1,4 @@
-#LyX 1.5.3 created this file. For more info see http://www.lyx.org/
+#LyX 1.5.1 created this file. For more info see http://www.lyx.org/
 \lyxformat 276
 \begin_document
 \begin_header
@@ -16898,10 +16898,10 @@
 \end_inset
 
  shows the error scaled with resolution.
- We can see that the green medium resolution error is always between the
- blue high resolution error and the red low resolution error.
- This means that the scaled high resolution error is closer to the medium
- resolution error than the low resolution error.
+ We can see that the green medium-resolution error is always between the
+ blue high-resolution error and the red low-resolution error.
+ This means that the scaled high-resolution error is closer to the medium-resolu
+tion error than the low-resolution error.
  In other words, the error is following the expected relation
 \end_layout
 
@@ -16942,8 +16942,19 @@
 \end_inset
 
 Error in the strain rate for the kinetic friction benchmark.
- Red is low resolution (16x16), green is medium resolution (32x32), and
- blue is high resolution (64x64).
+ Red is low resolution (16
+\begin_inset Formula $\times$
+\end_inset
+
+16), green is medium resolution (32
+\begin_inset Formula $\times$
+\end_inset
+
+32), and blue is high resolution (64
+\begin_inset Formula $\times$
+\end_inset
+
+64).
 \end_layout
 
 \end_inset
@@ -16984,10 +16995,21 @@
 
 Error in the strain rate scaled with resolution for the kinetic friction
  benchmark.
- Red is low resolution (16x16), green is medium resolution (32x32), and
- blue is high resolution (64x64).
- The high resolution error is closer to the medium resolution error than
- the low resolution error, indicating proper linear convergence.
+ Red is low resolution (16
+\begin_inset Formula $\times$
+\end_inset
+
+16), green is medium resolution (32
+\begin_inset Formula $\times$
+\end_inset
+
+32), and blue is high resolution (64
+\begin_inset Formula $\times$
+\end_inset
+
+64).
+ The high-resolution error is closer to the medium-resolution error than
+ the low-resolution error, indicating proper linear convergence.
 \end_layout
 
 \end_inset
@@ -17114,8 +17136,11 @@
 
 \end_inset
 
- shows runs with low resolution (256x32), runs with more accurate nonlinear
- solves (
+ shows runs with low resolution (256
+\begin_inset Formula $\times$
+\end_inset
+
+32), runs with more accurate nonlinear solves (
 \begin_inset Formula $2\cdot10^{-3}$
 \end_inset
 
@@ -17167,8 +17192,31 @@
 \end_inset
 
 Strain rate invariant for the numerical extension models after 5 cm of extension.
- The resolutions of the various models are: I2ELVIS: 400x75, LAPEX-2D: 301x71,
- Microfem: 201x61, SloMo: 401x71, Sopale: 401x71, Gale: 512x64.
+ The resolutions of the various models are: I2ELVIS: 400
+\begin_inset Formula $\times$
+\end_inset
+
+75, LAPEX-2D: 301
+\begin_inset Formula $\times$
+\end_inset
+
+71, Microfem: 201
+\begin_inset Formula $\times$
+\end_inset
+
+61, SloMo: 401
+\begin_inset Formula $\times$
+\end_inset
+
+71, Sopale: 401
+\begin_inset Formula $\times$
+\end_inset
+
+71, Gale: 512
+\begin_inset Formula $\times$
+\end_inset
+
+64.
 \end_layout
 
 \end_inset
@@ -17209,8 +17257,15 @@
 \end_inset
 
 Strain rate invariant for the extension model after 5 cm of extension for
- a) the reference run, b) a lower resolution run (256x32 rather than 512x64),
- c) a run with more accurate nonlinear solves (
+ a) the reference run, b) a lower resolution run (256
+\begin_inset Formula $\times$
+\end_inset
+
+32 rather than 512
+\begin_inset Formula $\times$
+\end_inset
+
+64), c) a run with more accurate nonlinear solves (
 \begin_inset Formula $2\cdot10^{-3}$
 \end_inset
 
@@ -17336,13 +17391,23 @@
 \begin_inset Formula $10^{-5}$
 \end_inset
 
-) is required in order to get the benchmark for high resolution (512x64)
- to run to completion (14 cm of shortening).
+) is required in order to get the benchmark for high resolution (512
+\begin_inset Formula $\times$
+\end_inset
+
+64) to run to completion (14 cm of shortening).
  In general, this benchmark is more challenging to run, so we only have
  the sensitivity analysis for an earlier time (10 cm of shortening rather
  than 14 cm).
- Figure shows runs with lower resolution (256x32 rather than 512x64), runs
- with less accurate linear solves (
+ Figure shows runs with lower resolution (256
+\begin_inset Formula $\times$
+\end_inset
+
+32 rather than 512
+\begin_inset Formula $\times$
+\end_inset
+
+64), runs with less accurate linear solves (
 \begin_inset Formula $10^{-5}$
 \end_inset