r21908 - in /trunk: docs/latex/dispersion.tex specific_analyses/relax_disp/parameters.py -- December 09, 2013 - 14:33

Author: bugman
Date: Mon Dec  9 14:33:05 2013
New Revision: 21908

URL: http://svn.gna.org/viewcvs/relax?rev=21908&view=rev
Log:
Removed the pC <= pB constraint from the 3-site dispersion models.

This is important for the linear models where a violation of this constraint 
is reasonable.  This
has been replaced by the pC <= pA constraint.


Modified:
    trunk/docs/latex/dispersion.tex
    trunk/specific_analyses/relax_disp/parameters.py

Modified: trunk/docs/latex/dispersion.tex
URL: 
http://svn.gna.org/viewcvs/relax/trunk/docs/latex/dispersion.tex?rev=21908&r1=21907&r2=21908&view=diff
==============================================================================
--- trunk/docs/latex/dispersion.tex (original)
+++ trunk/docs/latex/dispersion.tex Mon Dec  9 14:33:05 2013
@@ -1460,10 +1460,9 @@
     \dwHAB \geqslant 0, \\
     \dwHBC \geqslant 0, \\
     \pA\dw^2 \geqslant 0, \\
-    \pA \geqslant 0, \\
-    \pB \geqslant 0, \\
-    \pC \geqslant 0, \\
-    \pC \leqslant \pB \leqslant \pA \leqslant 1, \\
+    0 \leqslant \pA \leqslant 1, \\
+    0 \leqslant \pB \leqslant \pA, \\
+    0 \leqslant \pC \leqslant \pA, \\
     \pA \geqslant 0.85 \quad (\textrm{the skewed condition, } \pA \gg \pB), 
\\
     0 \leqslant \kex \leqslant 2e^6, \\
     0 \leqslant \kexAB \leqslant 2e^6, \\

Modified: trunk/specific_analyses/relax_disp/parameters.py
URL: 
http://svn.gna.org/viewcvs/relax/trunk/specific_analyses/relax_disp/parameters.py?rev=21908&r1=21907&r2=21908&view=diff
==============================================================================
--- trunk/specific_analyses/relax_disp/parameters.py (original)
+++ trunk/specific_analyses/relax_disp/parameters.py Mon Dec  9 14:33:05 2013
@@ -638,11 +638,11 @@
                     b.append(-1.0 / scaling_matrix[param_index, param_index])
                     j += 1
 
-                    # Then the pB >= pC constraint (rearranged as pA + 2pB 
> = 1).
+                    # Then the pA >= pC constraint.
                     A.append(zero_array * 0.0)
                     A[j][param_index2] = 1.0
-                    A[j][param_index] = 2.0
-                    b.append(1.0 / scaling_matrix[param_index, param_index])
+                    A[j][param_index] = -1.0
+                    b.append(0.0)
                     j += 1
                     break