bcf06ab | Eh Tan | 09 August 2007, 22:57:28 UTC | Finished the compressible Stokes solver for TALA. Two non-dimensional parameters are added: "dissipation_number" and "gruneisen" under the Solver component. One can use the original incompressible solver by setting "gruneisen=0". The code will treat this as "gruneisen=infinity". Setting non-zero value to "gruneisen" will switch to compressible solver. One can use the TALA solver for incompressible case by setting "gruneisen" to a non-zero value while setting "dissipation_number=0". This is useful when debugging the compressible solver. Two implementations are available: one by Wei Leng (U. Colorado) and one by Eh Tan (CIG). Leng's version uses the original conjugate gradient method for the Uzawa iteration and moves the contribution of compressibility to the RHS, similar to the method of Ita and King, JGR, 1994. Tan's version uses the bi-conjugate gradient stablized method for the Uzawa iteration, similar to the method of Tan and Gurnis, JGR, 2007. Both versions agree very well. In the benchmark case, 33x33x33 nodes per cap, Di/gamma=1.0, Ra=1.0, delta function of load at the mid mantle, the peak velocity differs by only 0.007%. Leng's version is enabled by default. Edit function solve_Ahat_p_fhat() in lib/Stokes_flow_Incomp.c to switch to Tan's version. | 09 August 2007, 22:57:28 UTC |
91bcb85 | Eh Tan | 03 August 2007, 19:01:02 UTC | Rearranged the code calculating element stiffness matrix | 03 August 2007, 19:01:02 UTC |
718df88 | Eh Tan | 03 August 2007, 19:00:41 UTC | Naming the functions for Uzawa iteration by the algorithm (CG or BiCG), instead of by the approximation (BA or TALA) | 03 August 2007, 19:00:41 UTC |
b3ea5b1 | Eh Tan | 03 August 2007, 18:59:06 UTC | Fixed a bug. fabs() returns a float, but abs() returns an int. | 03 August 2007, 18:59:06 UTC |
03bd866 | Eh Tan | 26 July 2007, 22:58:03 UTC | Porting r7748. Fixed a bug that *opt*.general files are mistakenly named. This bug is introduced by svn r7236 | 26 July 2007, 22:58:03 UTC |
6098ee4 | Eh Tan | 26 July 2007, 22:57:44 UTC | Update for new format of comp_nd files | 26 July 2007, 22:57:44 UTC |
86d9691 | Eh Tan | 26 July 2007, 22:03:09 UTC | Fixed a bug in ordering of mesh nodes. The bug was introduced in svn r6510. | 26 July 2007, 22:03:09 UTC |
dcb51dc | Eh Tan | 26 July 2007, 22:02:43 UTC | Fixed a bug in stable timestep | 26 July 2007, 22:02:43 UTC |
154b603 | Eh Tan | 14 June 2007, 21:40:29 UTC | Porting r7245 from trunk -- Zeroing geoid arrays before every geoid calculation. This fixes issue114 | 14 June 2007, 21:40:29 UTC |
de64ba2 | Eh Tan | 14 June 2007, 20:38:42 UTC | Porting r7236 from trunk | 14 June 2007, 20:38:42 UTC |
997dc47 | Eh Tan | 14 June 2007, 01:38:16 UTC | Porting r7221 from trunk -- Fixing issue111: Expanding %RANK in autocombine.py | 14 June 2007, 01:38:16 UTC |
ccd1ab4 | Eh Tan | 12 June 2007, 00:05:50 UTC | Fixing issue112 | 12 June 2007, 00:05:50 UTC |
0c34be6 | Eh Tan | 07 June 2007, 14:02:13 UTC | Fixed issue110 | 07 June 2007, 14:02:13 UTC |
ec719d4 | Eh Tan | 08 May 2007, 21:28:45 UTC | Changing the header format of comp_nd/comp_el ASCII outputs. The combining scripts are updated as well. The header has two lines, in the format of: ijunk n_rows time n_columns initial_bulk_composition[i] current_bulk_composition[i] (i = 0 to n_columns) | 08 May 2007, 21:28:45 UTC |
6f312c5 | Eh Tan | 07 May 2007, 21:30:13 UTC | Multi-component chemical convection. By setting tracer_flavors > 2, one can enable multi-component chemical convection. See tests/checkpoint/tracer.cfg for example input. | 07 May 2007, 21:30:13 UTC |
6177b48 | Eh Tan | 07 May 2007, 21:23:11 UTC | Forgot to remove 'composition.ireset_initial_composition' from the header file | 07 May 2007, 21:23:11 UTC |
fc5decb | Eh Tan | 03 May 2007, 00:02:06 UTC | Incorporating the effect of density/expansivity/gravity profiles when computing geoid | 03 May 2007, 00:02:06 UTC |
89f7438 | Eh Tan | 02 May 2007, 20:06:42 UTC | Forgot to update the makefile | 02 May 2007, 20:06:42 UTC |
067f35e | Eh Tan | 02 May 2007, 00:00:31 UTC | Updated Cookbook5 so that it uses only 2 processors | 02 May 2007, 00:00:31 UTC |
61465a8 | Eh Tan | 30 April 2007, 23:22:34 UTC | Using tic_method=-1 to read initial termpature from files | 30 April 2007, 23:22:34 UTC |
d07bf8c | Eh Tan | 30 April 2007, 23:22:15 UTC | Expand datadir_old when post_processing=on or tic_method=-1 | 30 April 2007, 23:22:15 UTC |
70fd7f4 | Eh Tan | 27 April 2007, 19:31:10 UTC | Added a functional test for checkpoint | 27 April 2007, 19:31:10 UTC |
0af6033 | Eh Tan | 27 April 2007, 19:24:44 UTC | Init'd E->mat in initial_conditions() and read_checkpoint() | 27 April 2007, 19:24:44 UTC |
ec80762 | Eh Tan | 26 April 2007, 00:46:11 UTC | Updating CoupledSolver for the checkpoint and tracers. The checkpoint stuff is not finished. | 26 April 2007, 00:46:11 UTC |
1b26944 | Eh Tan | 26 April 2007, 00:45:51 UTC | Removed the restart hack from Solver.py, which was required by solver coupling. The hack is still useful in CoupledSolver.py | 26 April 2007, 00:45:51 UTC |
a7111d6 | Eh Tan | 26 April 2007, 00:45:09 UTC | Clean up restart in lith_age | 26 April 2007, 00:45:09 UTC |
dea4ac2 | Eh Tan | 26 April 2007, 00:44:38 UTC | Setting solver.ic.tic_method=-1 will read the temperature field from old velo files. | 26 April 2007, 00:44:38 UTC |
55a6bec | Eh Tan | 25 April 2007, 20:17:19 UTC | Append timestep to the checkpoint filename. Write/read timing information correctly. | 25 April 2007, 20:17:19 UTC |
6962d60 | Eh Tan | 25 April 2007, 19:30:44 UTC | Clean up the old way of tracers/composition restart. Setting solver.ic.restart=on won't read the *.tracer.* and *.comp_el.* files anymore. Setting solver.tracer.ic_method=2 will read the old *.tracer.* files as before, but the composition field will be calculated according to the tracers, not read from the *.comp_el.* files. The parameter solver.tracer.reset_initial_composition becomes obsolete. | 25 April 2007, 19:30:44 UTC |
12e75b9 | Eh Tan | 25 April 2007, 19:03:00 UTC | Fixed a bug in tracer setup | 25 April 2007, 19:03:00 UTC |
f8acab5 | Eh Tan | 24 April 2007, 22:16:50 UTC | Implemented writing and reading (binary) checkpoints. The checkpoint file is called [datafile].chkpt.[rank] and is outputed with an interval of controller.checkpointFrequency. Before writing the checkpoint file, the previous checkpoint file (if exists) is renamed to [datafile].chkpt.[rank].bak. To read back the checkpoint file, set solver.ic.restart=on. The old checkpoint file, [datafile_old].chkpt.[rank] will be read in. The timestep information is stored in the checkpoint file. There is no need to specify solution_cycles_init, as the old way of restarting. The old way of restarting (read in temperature from the velo files) will still be available through solver.ic.tic_method=-1. (To do.) Several things to do: clean up the old way of restarting tracers/composition, documentation, does mat_control need checkpoints? | 24 April 2007, 22:16:50 UTC |
7867dae | Eh Tan | 16 April 2007, 22:38:30 UTC | Compute adiabatic cooling correctly. Assuming gravity is constant. Slightly modified viscous heating computation to improve the efficiency. | 16 April 2007, 22:38:30 UTC |
1c88d2f | Eh Tan | 16 April 2007, 22:37:27 UTC | Compute Tmass matrix with density (missing heat capacity) | 16 April 2007, 22:37:27 UTC |
17ff14a | Eh Tan | 16 April 2007, 22:36:30 UTC | Fixed a typo | 16 April 2007, 22:36:30 UTC |
fccccc5 | Eh Tan | 16 April 2007, 22:35:39 UTC | Removed some unused variables in global_defs.h | 16 April 2007, 22:35:39 UTC |
93687ef | Eh Tan | 16 April 2007, 22:34:53 UTC | Added indentation and comments to mass_matrix() | 16 April 2007, 22:34:53 UTC |
a34172d | Eh Tan | 10 April 2007, 23:43:47 UTC | Merging r6541-r6542 from trunk | 10 April 2007, 23:43:47 UTC |
de3d198 | Eh Tan | 06 April 2007, 23:07:02 UTC | Merging r6041 to r6510 from trunk | 06 April 2007, 23:07:02 UTC |
0fb9bb0 | Eh Tan | 06 April 2007, 23:00:39 UTC | Merging r6365 from trunk | 06 April 2007, 23:00:39 UTC |
0753462 | Eh Tan | 06 April 2007, 22:57:44 UTC | Merging r6364 from trunk | 06 April 2007, 22:57:44 UTC |
7bc7173 | Eh Tan | 08 February 2007, 00:19:05 UTC | Porting r5956 to r5980 from trunk | 08 February 2007, 00:19:05 UTC |
494622d | Eh Tan | 02 February 2007, 20:32:09 UTC | Porting r5952 from trunk | 02 February 2007, 20:32:09 UTC |
9d9a7dd | Eh Tan | 31 January 2007, 20:58:23 UTC | Fixed two bugs in heating | 31 January 2007, 20:58:23 UTC |
134d38a | Eh Tan | 31 January 2007, 20:50:30 UTC | Porting r5942 from trunk | 31 January 2007, 20:50:30 UTC |
3581735 | Eh Tan | 30 January 2007, 21:32:04 UTC | Porting r5930-5935 from trunk | 30 January 2007, 21:32:04 UTC |
2008acc | Eh Tan | 30 January 2007, 19:10:00 UTC | Porting r5921-5928 from trunk | 30 January 2007, 19:10:00 UTC |
0504ed7 | Eh Tan | 30 January 2007, 02:07:30 UTC | First cut on computing adibatic/viscous/latent heating. Most of the code is copied from CitcomCU r2546 and refactored later. It is not tested yet. | 30 January 2007, 02:07:30 UTC |
60884b1 | Eh Tan | 25 January 2007, 23:41:54 UTC | Renamed the input parameter for dissipation number to be consistent with CitcomCU | 25 January 2007, 23:41:54 UTC |
b5cdfc8 | Eh Tan | 25 January 2007, 23:41:15 UTC | Porting r5896 from trunk | 25 January 2007, 23:41:15 UTC |
5a6a5be | Eh Tan | 25 January 2007, 01:45:20 UTC | Porting r5886-5887 from trunk | 25 January 2007, 01:45:20 UTC |
fa6fd60 | Eh Tan | 24 January 2007, 02:17:10 UTC | Removed unused functions and struct members | 24 January 2007, 02:17:10 UTC |
a8c708a | Eh Tan | 24 January 2007, 01:10:13 UTC | Fixed a bug when computing pressure correction. Using full spherical solver, Di=0, init T perturbation (l,m)=(3,2), and default values for other parameters, the differences in CG and BiCGstab solvers are 0.5% in velocity and stress, 0.2% in pressure. | 24 January 2007, 01:10:13 UTC |
a8b86a8 | Eh Tan | 19 January 2007, 20:14:41 UTC | Porting r5744-5837 from trunk | 19 January 2007, 20:14:41 UTC |
678a6a7 | Eh Tan | 19 January 2007, 00:14:37 UTC | BiCGStab algorithm for TALA Stokes solver Compared with the CG algorithm solver, the velocity differs by ~0.1% but the stress is way wrong. | 19 January 2007, 00:14:37 UTC |
7d3eb44 | Eh Tan | 19 January 2007, 00:10:53 UTC | Refactoring Stokes solver | 19 January 2007, 00:10:53 UTC |
b519e61 | Eh Tan | 12 January 2007, 00:06:11 UTC | Cleaned up, beautified and added some comments | 12 January 2007, 00:06:11 UTC |
d9a2d84 | Eh Tan | 11 January 2007, 21:01:06 UTC | 1. Stored the Gruneisen parameter in its inverse 2. Calculated reference state 3. Removed contribution of ref. temperature from buoyancy | 11 January 2007, 21:01:06 UTC |
e66e0c6 | Eh Tan | 10 January 2007, 22:01:50 UTC | Taking variable density field into the calculation of thermal buoyancy | 10 January 2007, 22:01:50 UTC |
ccd7d8c | Eh Tan | 10 January 2007, 21:59:32 UTC | Minor refactoring | 10 January 2007, 21:59:32 UTC |
419bb1f | Eh Tan | 09 January 2007, 22:20:08 UTC | 1. Added two non-dimensional number: dissipation # and gruneisen parameter 2. Modified stiffness matrix for TALA 3. Modified stress calculation for TALA | 09 January 2007, 22:20:08 UTC |
3f8a9fe | Eh Tan | 09 January 2007, 22:01:20 UTC | Porting r5743 from trunk to branch | 09 January 2007, 22:01:20 UTC |
509fb47 | Eh Tan | 09 January 2007, 20:04:02 UTC | Branch for compressible (Truncated Anelastic Liquid Approximation) convection | 09 January 2007, 20:04:02 UTC |
1be8307 | Eh Tan | 08 January 2007, 23:09:47 UTC | Another document fix for issue75, related to "how to specify time-invariant velocity boundary conditions?" | 08 January 2007, 23:09:47 UTC |
0bfa938 | Eh Tan | 08 January 2007, 22:19:35 UTC | 1. Fixed a reference error, "Section 1.6" should be "Equation 1.6" 2. Velocity BC input and geoid output are dimensional. The units were undocumented in previous release. This fix the issue75. 3. More information about geoid output: the meaning of each column and how to compute the geoid field. | 08 January 2007, 22:19:35 UTC |
8fb120b | Eh Tan | 08 January 2007, 22:19:14 UTC | Truncate the temperature grd file at specified min/max Sometimes the temperature field has overshoot/undershoot, which would be leave a blank area in the contour plot. This is a simple workaround. | 08 January 2007, 22:19:14 UTC |
bf5e730 | Eh Tan | 05 January 2007, 22:33:03 UTC | Fix for issue73 "problems with autocombine.py in single workstation mode" $HOSTNAME is not defined in the non-interactive shell in Ubuntu 6.10. Need to put a quote around it. | 05 January 2007, 22:33:03 UTC |
7d3ef92 | Leif Strand | 05 January 2007, 00:06:11 UTC | Fix for issue72: "'make install' broken for non-pyred CitcomS". | 05 January 2007, 00:06:11 UTC |
9b67fc5 | Sue Kientz | 20 December 2006, 22:13:44 UTC | added note about updating to 2.10.3 | 20 December 2006, 22:13:44 UTC |
0b9c4d2 | Eh Tan | 09 December 2006, 00:26:38 UTC | Regenerated ChangeLog upto r5565 | 09 December 2006, 00:26:38 UTC |
545c4df | Sue Kientz | 09 December 2006, 00:22:28 UTC | gurnis final changes; image resize; ready for release | 09 December 2006, 00:22:28 UTC |
47bdc2d | Sue Kientz | 08 December 2006, 23:50:10 UTC | small text changes; image resize | 08 December 2006, 23:50:10 UTC |
7da60aa | Eh Tan | 08 December 2006, 23:45:26 UTC | Added "created by OpenDX" on several figure captions | 08 December 2006, 23:45:26 UTC |
03df21a | Eh Tan | 08 December 2006, 23:36:57 UTC | Updated MayaVi2 section | 08 December 2006, 23:36:57 UTC |
8f5c823 | Eh Tan | 08 December 2006, 23:30:39 UTC | Renamed the MayaVi launching script and move it to visual/ | 08 December 2006, 23:30:39 UTC |
dd70690 | Eh Tan | 08 December 2006, 23:26:02 UTC | Added executable bit to file permission | 08 December 2006, 23:26:02 UTC |
742ed51 | Luis Armendariz | 08 December 2006, 22:44:00 UTC | Add plugins for MayaVi2 to installation | 08 December 2006, 22:44:00 UTC |
92235c0 | Luis Armendariz | 08 December 2006, 22:27:26 UTC | Added README files | 08 December 2006, 22:27:26 UTC |
c60971f | Sue Kientz | 08 December 2006, 22:00:21 UTC | fixed image size, text size and other small text changes | 08 December 2006, 22:00:21 UTC |
8c92de8 | Luis Armendariz | 08 December 2006, 21:16:47 UTC | Removing reduce filter from comments, as it is currently obsolete. | 08 December 2006, 21:16:47 UTC |
4511a3d | Luis Armendariz | 08 December 2006, 21:15:17 UTC | Added MayaVi screenshot to manual, along with a couple of other fixes. | 08 December 2006, 21:15:17 UTC |
13061f6 | Luis Armendariz | 08 December 2006, 20:03:11 UTC | Last set of patches to citcoms_plugins * Display filename in mayavi pipeline * Changed glob pattern so that only HDF5 timestep files can be selected | 08 December 2006, 20:03:11 UTC |
af17846 | Luis Armendariz | 08 December 2006, 19:47:43 UTC | Forgot to rename data filters. | 08 December 2006, 19:47:43 UTC |
de6f70d | Luis Armendariz | 08 December 2006, 17:53:20 UTC | Added MayaVi2 screenshot for the manual. | 08 December 2006, 17:53:20 UTC |
75bb12b | Luis Armendariz | 08 December 2006, 17:44:47 UTC | Various patches. | 08 December 2006, 17:44:47 UTC |
b3423fa | Luis Armendariz | 08 December 2006, 17:40:08 UTC | Forking original plugins | 08 December 2006, 17:40:08 UTC |
9c83a26 | Luis Armendariz | 08 December 2006, 17:32:25 UTC | Archiving original plugins. | 08 December 2006, 17:32:25 UTC |
42b3b3e | Luis Armendariz | 08 December 2006, 17:31:25 UTC | Modify plugins directory | 08 December 2006, 17:31:25 UTC |
3c488f3 | Luis Armendariz | 08 December 2006, 17:29:46 UTC | Explain contents of each directory. | 08 December 2006, 17:29:46 UTC |
849cea5 | Leif Strand | 08 December 2006, 02:29:09 UTC | Made 'install_deps' actually do what 'configure' says it's doing (*missing* deps, not *all*). | 08 December 2006, 02:29:09 UTC |
8465d3f | Sue Kientz | 07 December 2006, 23:26:14 UTC | reinstated a few lost changes; changed caption position on fig 5.3 on request; enlarged fig 5.4; regen pdf | 07 December 2006, 23:26:14 UTC |
79a8733 | Eh Tan | 07 December 2006, 22:34:26 UTC | chapter 5 * Added MayaVi2 in the introduction * Added links to MayaVi2 installation | 07 December 2006, 22:34:26 UTC |
6e256a7 | Sue Kientz | 07 December 2006, 22:19:37 UTC | fixed image size; regen pdf | 07 December 2006, 22:19:37 UTC |
e7c2ea3 | Sue Kientz | 07 December 2006, 22:09:17 UTC | text fixes to recent additions | 07 December 2006, 22:09:17 UTC |
b7662aa | Eh Tan | 07 December 2006, 21:30:53 UTC | * Put back the HDFView sections * Added a screenshot of HDFView * Added a warning on "-O3" compiler flag | 07 December 2006, 21:30:53 UTC |
7bf8a4b | Sue Kientz | 07 December 2006, 05:03:23 UTC | text edits to appendices and bib | 07 December 2006, 05:03:23 UTC |
cf68821 | Sue Kientz | 07 December 2006, 00:12:17 UTC | small changes to chapt 6; standardized cross refference style; sstandardized code spacing | 07 December 2006, 00:12:17 UTC |
8910ace | Sue Kientz | 06 December 2006, 23:31:57 UTC | enumerating several sections of the instructions in chap 5 for clarity, and a few minor text fixes | 06 December 2006, 23:31:57 UTC |
cd4ffb9 | Sue Kientz | 06 December 2006, 20:59:49 UTC | a few text fixes in chapter 3 | 06 December 2006, 20:59:49 UTC |
bb3c41c | Sue Kientz | 06 December 2006, 20:33:12 UTC | a few text fixes, preface to end of chapt 2 | 06 December 2006, 20:33:12 UTC |
6f35a8c | Sue Kientz | 06 December 2006, 18:59:56 UTC | gurnis corrections and some other fixes | 06 December 2006, 18:59:56 UTC |