Those developers old enough to remember working with punched card decks like myself probably also remember the wonders and gotchas associated with the formula translation language, aka FORTRAN. That's why I was bemused to see Microsoft promoting Visual Numberics IMSL C# Numerical Library for numerical analysis. To its entire credit, this library offers broad and sophisticated mathematical capabilities.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
I got my jollies upon seeing this reference because IMSL stands for International Mathematical and Statistical Library (though it's also sometimes decoded as International Mathematics & Statistics Library). IMSL is a body of FORTRAN code sufficiently old enough for me to have crossed paths with it in graduate school in the late 1970s and early 1980s while taking (and programming in FORTAN) for numerical analysis classes.
A quick search on IMSL on Google, in fact, indicates that this toolset is still pretty widely used in the academic community (I found links to available libraries at institutions including USC, Duke, Indiana, and my own graduate alma mater, The University of Texas, pretty much right away). That said, the Visual Numerics implementation in C# has been entirely rewritten in that language and is 100% .NET compatible -- something that can't be said for the original, all-FORTRAN implementation that's still kicking around.
In fact, Visual Numerics offers versions of IMSL for C, C#, Java and even FORTRAN, all of which support the same general functionality where the developers have paid the kind of serious attention to precision, numerical representations, accuracy, and so forth, that you'd expect in a set of library routines designed to crunch numbers precisely and delicately. Among the library's capabilities, you'll find functionality across most of the spectrum of numerical computing, including optimization tools, Fast Fourier Transforms (FFTs), interpolation, differential equations, correlations and regression analyses, time series, and so on.
IMSL is widely enough studied and used to represent something of an accepted standard in numerical computing, and those who need its services in the .NET environment should pick this toolset up with genuine appreciation and ready application. IMSL continues to be well-utilized at research labs, in academia, and in all kinds of quantitative and numerical commercial applications, which makes it a good choice for such use in the .NET environment as well.
Single-seat pricing for .NET capable versions costs between $1,175 (node locked license) and $1,765 (floating license); discounts apply when more licenses are purchased. For more information, visit Visual Numeric's IMSL home page.
Ed Tittel is a writer and trainer whose interests include XML and development topics, along with IT Certification and information security. E-mail firstname.lastname@example.org with comments, questions, or suggested topics or tools to review. Cool tools rule!