Content area
Full Text
Displaying scientific data by computer has become essential to reduce the complexity of large data sets so they can be encompassed by the mind's eye. Data in digital form can be processed both quantitatively and graphically, allowing both accuracy and comprehension. In this review I describe and evaluate five scientific graphics packages running on desktop computers. They offer a variety of visualization methods.
Since there are dozens of such graphics packages on the market, I imposed two criteria to limit those reviewed. First, the software was required to operate on at least Windows 95 or Power Macintosh systems. The second requirement was that each package cover a broad range of graphics likely to be needed by many scientists. These criteria were met by five packages: Axum by MathSoft, Noesys by Fortner Research, Origin by Microcal, SigmaPlot by SPSS (formerly a Jandel product) and TecPlot by Amtec Engineering. (Noesys is pronounced "noesys," presumably from the Greek noesis, for the exercise of reason.)
Features. All the graphics packages reviewed here are completely adequate for typical applications in the sciences. Although computerized graphics relieve much of the tedium of plotting, scientists still have responsibility for consistent and logical style, as emphasized by Engineering columnist Henry Petroski (American Scientist, January-February 1995). We discuss here features that might make one package more suitable than others for a given kind of task.
In Axum the main interface is very similar in layout, icons and function to Microsoft Word, except that Axum is designed primarily for graphics rather than text. It is therefore very useful for preparing reports that have many graphics interspersed with text, since then the Axum and Word interfaces are consistent. Axum also provides a programming language resembling C that can be used to write commands for manipulating columns of data before plotting them.
The color graphics in Noesys are of very high quality. In their recent book Number by Colors (Springer, New York, 1997), Brand Fortner and Theodore Meyer provide an introduction to using color for understanding technical data. Both authors are at Fortner Research, the home of Noesys. This package, in either Macintosh or Windows, is very useful for visualizing large and complicated multiparameter sets of data. It is particularly good for three-dimensional and color displays.
Origin...