文件列表:

disloc3d_raw\dc3d.f (47207, 2012-12-01)
disloc3d_raw\dc3omp.f (85429, 2012-12-01)
disloc3d_raw\dc3quadrant.f (85190, 2012-12-01)
disloc3d_raw\disloc3d.F (8170, 2017-03-22)
disloc3d_raw\disloc3d.m (1758, 2012-12-01)
disloc3d_raw\disloc3d.mexa64 (42074, 2012-12-01)
disloc3d_raw\disloc3domp.c (6002, 2012-12-01)
disloc3d_raw\disloc3domp.m (1790, 2012-12-01)
disloc3d_raw\disloc3domp.mexa64 (55218, 2012-12-01)
disloc3d_raw\disloc3dpm.m (45369, 2012-12-01)
disloc3d_raw\ex.m (2522, 2018-04-09)
disloc3d_raw\FaultDiagram.pdf (68139, 2012-12-01)
disloc3d_raw\mexopts.sh (12544, 2012-12-01)
disloc3d_raw (0, 2018-04-04)

disloc3d v0.1 Crustal Deformation and Fault Mechanics Group, Stanford University This package provides wrappers to the code DC3.F by Y. Okada. Cite: Okada, Y., 1992. Internal deformation due to shear and tensile faults in a half space, Bull. seism. Soc. Am., 82, 1018--1040. The wrappers differ only in software details; the math and usage are the same. In Matlab, type help disloc3d to see basic usage. Read FaultDiagram.pdf to learn the geometry conventions. In Matlab, type ex to run an example, and edit ex.m to view the code for example usage. --------- This package contains three versions of the same underlying routine. All three are used in Matlab. In increasing order of difficult of (a) getting these working and (b) speed, they are as follows: 1. The first version is disloc3dpm.m. It is pure Matlab code and so is much slower than the other two. But it can be used for small problems if you can't get either of the next two versions to build. 2a. The second version is a serial mex file that is a basic wrapper to DC3.F. To build it, type mex disloc3d.F dc3d.f If this does not work for you, then you need to set up mex and possibly install compilers. See http://www.mathworks.com/help/matlab/ref/mex.html 2b. We have implemented a fix to a numerical problem that occurs in and nearly in the plane of a rectangular element very close to the rays extending from the edges. This numerical error is essentially never an issue for most users of this code, so you should not worry about it. We provide the fix in a separate version of the code in case you wish to try it. Type mex disloc3d.F dc3quadrant.f 3. The third version is also a mex file. It uses OpenMP to parallelize the operations. It also compensates for the numerical error reported in 2b. For a variety of reasons, the code can be much harder to build than version 2. On a UNIX system, attempt to build it as follows: mex -DHMMVP_MEXSVD CXXFLAGS="\$CXXFLAGS -fopenmp" LDFLAGS="\$LDFLAGS -fopenmp" disloc3domp.c dc3omp.f --------- Version History: 0.0. 100812. AMB. First release for general use. 0.1. 111612. AMB. New method based on xi,eta>0 quadrant to get rid of cancellation errors. Implemented in dc3quadrant.f and dc3omp.f but *not* dc3d.f.

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：