Index: /trunk/DOC/documentation/cpdet.tex
===================================================================
--- /trunk/DOC/documentation/cpdet.tex	(revision 2330)
+++ /trunk/DOC/documentation/cpdet.tex	(revision 2331)
@@ -112,24 +112,31 @@
 \section{Pratical aspects in the code}
 
-A specific file has been added to the dynamical core, \textsf{cpdet.F}, which includes 
+A specific file has been added to the dynamical core, \textsf{dyn3d\_common/cpdet\_mod.F90}, which includes 
 all the needed routines to take the $C_p(T)$ possibility into account.
-These routines take advantage of the keyword \textsf{planet\_type} to 
-implement different expressions of $C_p(T)$. These routines are:
+These routines take advantage of the boolean flag \textsf{cpofT} (declared in \textsf{gcm.def}, default is T for Venus, F otherwise)
+to implement different expressions of $C_p(T)$. 
+It could also use the keyword \textsf{planet\_type} if needed to distinguish different planets.
+
+These routines are:
 \begin{itemize}
 \item \textsf{ini\_cpdet}: initializes the parameters $\nu$ (\textsf{nu\_venus})
 and $T_0$ (\textsf{t0\_venus}) to either the Venus values, or zero. 
 It is called just once at the begining of the main routine.
-These parameters are declared in \textsf{comconst.h}.
+These parameters are declared in \textsf{comconst\_mod.F90}.
 \item \textsf{cpdet}: function, that computes $C_p$ for a given $T$. 
 For other planets than Venus, it is just \textsf{cpdet(T)=cpp}.
 \item \textsf{t2tpot}: converts a temperature vector to a potential temperature
-vector.
+vector. Serial \textsf{t2tpot} and parallel \textsf{t2tpot\_p} versions.
 \item \textsf{tpot2t}: converts a potential temperature vector to a temperature
-vector.
+vector. Serial \textsf{tpot2t} and parallel \textsf{tpot2t\_p} versions.
 \end{itemize}
 
-In the routines, instead of using the constants \textsf{cpp} (dynamical core) 
-or \textsf{RCPD} (physical module), we need to use the 
-function \textsf{cpdet(T)} where T is the temperature at a given point.
+In the physics, these routines are also defined \textsf{phyvenus/cpdet\_phy\_mod.F90}. 
+They are initialized with \textsf{init\_cpdet\_phy} that is called from the dynamics/physics interface, 
+passing the \textsf{cpp}, \textsf{t0\_venus} and \textsf{nu\_venus} variables. 
+The value of \textsf{t0\_venus} (0 or not) is used to determine the expression of $C_p(T)$.
+
+For Venus, in the dynamics/physics interface, the routine \textsf{suphec(cpp)} needed to be called to initialize \textsf{RCPD=cpp} in the physics.
+In the routines, instead of using the constants \textsf{cpp} (dynamical core) or \textsf{RCPD} (physical module), we need to use the function \textsf{cpdet(T)} where T is the temperature at a given point.
 
 Since the potential temperature is the variable used in the dynamical core,