Results 1 - 10 of 4175
Results 1 - 10 of 4175. Search took: 0.026 seconds
|Sort by: date | relevance|
[en] FPP is a preprocessor which aids in porting Fortran source code across differing platforms. It provides conditional compilation features to enable or disable sections of code, and can modify file names in INCLUDE statements to a syntax suitable for a target platform. FPP is written Fortran 77, and runs on VM/CMS, VAX/VMS, UNIX, and PC/DOS SYSTEMS
[en] More than about 50% of stars are in binaries, but most stellar population studies take single star stellar population (ssSSP) models, which do not take binary interactions into account. In fact, the integrated peculiarities of ssSSPs are different from those of stellar populations with binary interactions (bsSSPs). Therefore, it is necessary to investigate the effects of binary interactions on the Lick indices and colors of populations in detail. We show some formulae for calculating the difference between the Lick indices and colors of bsSSPs, and those of ssSSPs. Twenty-five Lick indices and 12 colors are studied in this work. The results can be conveniently used for calculating the effects of binary interactions on stellar population studies and for adding the effects of binary interactions into present ssSSP models. The electronic data and fortran procedures in the paper can be obtained on request from the authors. (research papers)
[en] Metalinguistic description is presented of FORTRAN syntax made with the help of Bacus notations. Unambiguity of the language syntax enables us to regard the paper as a reference book for FORTRAN studies. Due to a large number of the language vernaculars the authors admit some deviations from the syntax rendered and warn users against an automatic application of determinations in the course of the program elaboration
[en] GRESS is a FORTRAN precompiler and run time library which automatically processes computer models and adds derivative-taking capabilities to the normal calculated results. The GRESS system is under development at ORNL and work is proceeding in upgrading test versions of the precompiler to allow processing by GRESS of most of the ANSI X3.9-1978 FORTRAN syntax. Heretofore, a major limitation has been the inability of GRESS to process arrays with more than two dimensions. Consequently, a significant amount of human effort was required to convert three- and higher-dimensional arrays into two-dimensional arrays for processing with GRESS. This report describes the addition of ten operation codes to the GRESS precompiler and run time library to allow application to FORTRAN programs with three-dimensional arrays. The new operation codes were tested on a sample problem. A comparison of CPU time is made for sample problems run on a VAX 8600 computer with one-, two-, and three-dimensional arrays. It is recommended that GRESS be further modified to accommodate higher-dimensional arrays up to the ANSI X3.9-1978 limit of seven dimensions
[en] OpenMP (13) is the dominant programming model for shared-memory parallelism in C, C++ and Fortran due to its easy-to-use directive-based style, portability and broad support by compiler vendors. Similar characteristics are needed for a programming model for devices such as GPUs and DSPs that are gaining popularity to accelerate compute-intensive application regions. This paper presents extensions to OpenMP that provide that programming model. Our results demonstrate that a high-level programming model can provide accelerated performance comparable to hand-coded implementations in CUDA.