PARD: Scientific Applications

• Gnuplot
• gunk
• plotutils
• SCEPTRE-90
• Scientific Applications on Linux
• Mathematics
  • Mpexpr
  • Octave

Gnuplot

Gnuplot handles both curves (2 dimensions) and surfaces (3 dimensions). Surfaces can be plotted as a mesh fitting the specified function, floating in the 3-d coordinate space, or as a contour plot on the x-y plane. For 2-d plots, there are also many plot styles, including lines, points, lines with points, error bars, and impulses (crude bar graphs). Graphs may be labeled with arbitrary labels and arrows, axes labels, a title, date and time, and a key. The interface includes command-line editing and history on most platforms.

gunk

This simulator portrays a rectangular surface coated with colored points, each representing a 68000 processor with some memory. Program memory is read-only, and identical for each processor. Because of simulation memory limitations, the memory for stack and variables is small, but controllable with a command line option.

plotutils

SCEPTRE-90

It employs a free-form input language and state variable methods to simulate problems of interest to electrical engineers.

Other highlights:

• Flexible nonlinear input. Circuit elements may be described as constants, tabular data, or abritrary functions of other network quantities.
• Flexible modeling capability. The arbitrary functional capability allows any device to be modeled to the degree of accuracy required for an analysis.
• State-of the art numerical methods. The code simulates networks in a cost-effective manner. It employs three explicit and one implicit integration routines.
• Stored models. In addition to built-in elements, SCEPTRE is capable of storing circuit models permanently, or temporarily in its own model library.
• Automatic initial conditions. The initial conditions of circuit variables are determined in two ways: automatically determined from DC steady-state solution of a circuit network, or manually from user specified initial conditions.
• Rerun. Multiple case rerun based on a single master run may be carried out automatically. The user supplies only the changes that apply from the master run for each repeated run.
• Defined Parameters. A special section has been created to enable the user to define quantities that may be output other than sources or passive currents and voltages. The user may enter systems of first-order differential equations that may or may not have anything to do with a particular electrical network.
• Minimization of computational delays. The sequencing of instructions by SCEPTRE ensures that computational delays in transient analyses will be few even when complex functional dependencies are present.
• Variety of run controls. The user may control simulations to a fine degree rather than "giving" the network to the code and losing control.
• Subprogram Capability. The user who is familiar with computer programming may write FORTRAN subroutines and insert them in otherwise conventional SCEPTRE runs. This option permits handling special situations, even though these should be rare.
• Multiple analyis available. Allow the analyst to use one code to run various analyses on circuit: DC (sensitivity, worst-case, Monte-Carlo optimization, AC, transient analyses.

A graphical user interface NGP has been added using GNUPLOT. As the source code is available, NGP may be modified due to other graphical software.

Scientific Applications on Linux

Mathematics

Mpexpr

Octave