Bill, Thanks for the updates. I have a couple of questions: 1: You said (userDefinedMotion.C) the object params has been renamed to parameters. I made some changes in the userDefinedMotion.C, can I still use the modified subroutine in the new version? 2: Will cgmp consider the viscosity change with the temperature? I made the change in cgins before. Thanks, Yongsheng On Mon, Apr 6, 2009 at 3:02 PM, Bill Henshaw <henshaw@xxxxxxxx> wrote: > Dear Overture users: > > Announcing the release of Overture.v23 and cg.v23. > This version should compile with the newer versions > of gcc (I use 4.3.2). > > A list of some of the major changes to Overture and cg are > given below (from the Overture/CHANGES and cg/Changes files). > > Please let me know if you have troubles building or using > the new version or if I have forgotten to fix some bug that > has been previously reported. > > > Regards, > #6. > > > ====================================================================================================== > Overture.v23 - released March 13, 2009 > > Changes: > > o NOTE: Comments in command files can now begin with # or * > > o The new version of A++/P++ supports arrays that can have 2^63 = 9e18 > elements, the previous version was > limited to 2^31 = 2e9 elements. > > o ** Data base files now contain the Overture version number they were > created with. If you are unable > to read an old file with a newer version of Overture, Overture will tell > you which older version to use. > o fixed bug in plotting sequences in the show file. > o fixed a bug in Oges -- the initial guess for iterative solvers was not > being assigned properly > for the case of systems of equations. > o there is a new example in Overture/primer/gridPrint.C that reads a grid > from a data base and > outputs the grid information to a plain text file. > o commands generated by perl in command files can now contain a comment > line (starting with a '*' or '#') > o the 2d contour plotter now has an option for flat-shading (toggle > button) where the colour of each > grid cell is constant (based on the average values at the 4 vertices) > o fixed a bug in ogen that caused the hole-cutting to fail in some special > cases in 2D > (cutHolesNew.bC, line 2241) > o fixed the hyperbolic grid generator so that it handles start curves that > are "derivative" periodic > o fixed a bug in Oges.C for predefined equations with mixed-BC's in which > sometimes the system was > wrongly thought to be singular. > o fixed a bug in ReductionMapping.C -- the inverseMap was not correct. > o fixed a bug in the mixed/aDotGradu/generalized-mixed BC on curvilinear > grids -- a matrix was built > and saved to hold the BC but this matrix was not updated if the > coefficents of the mixed BC changed. > o the option was added to the grid plotter to colour grids by domain > number (for multi-domain grids) > o there is a coefficient matrix example in tests/tcmge.C, that shows how > to fill in arbitrary equations. > o there is a new displayCoeff function to display elements in a > coefficient matrix in a nice way. > o the new version of Ogen can be used to generate grids in parallel for > many cases (some hard cases > that require backup interpolation will not work yet). > > > ============================================================================================== > cg.v23 -- released March 13, 2009 > > Changes: > > > o userDefinedMotion.C : the object params has been renamed to parameters. > o userDefinedBoundaryValues.C : the getBoundaryData(..) functions are now > in parameters.getBoundaryData(...) > o There is now an option to save solution values over time at specified > points using "probes". This information > is written to a text file that can be plotted with the matlab script > plotProbes.m > o There have been many changes to cgmp. There now many examples of > conjugate heat transfer problems that > couple fluid flows and heat transfer. > o The solid mechanics solver cgsm is starting to acquire some > capabilities. > >