Thanks for visiting Scientific Computing World.

You're trying to access an editorial feature that is only available to logged in, registered users of Scientific Computing World. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

Matlab 7.6

Share this on social media:

Mathworks version numbers are confusing, compounded by the numbering and combinations of packages. To be clear from the outset: this review is of Matlab version 7.6 only, concentrating on new aspects. Matlab is a core component of Mathworks, whose version R2008a was released in February of this year but is not covered here.

Starting with Matlab’s core raison d’être, there is a range of developments in mathematics scope and handling, data analysis and performance. There are several areas in which handling has been streamlined, JIT/Accelerator support is extended, and large sparse matrix multiplications in particular show a noticeable speed increase that delivers considerable benefit over homogenous high volume calculation runs. BLAS and LAPACK libraries are upgraded and a preferred version of FFTW or LAPACK can, as with BLAS, be selected in place of the default via environment variable. Multithread support is incrementally improved and, at the detail level, matrix general function (funm), Hermitian indefinite factorisation (ldl), and logarithm (logm) get new cutting edge algorithms. User interactivity gets welcome attention, with sophisticated linked data brushing and update linkage of displays to changes in underlying variables in the workspace.

Even a reluctant programmer like myself immediately notices that facilities for managing these facilities have developed greatly. Monitoring and control of large and complex operational structures is simplified by housekeeping options and expanded object orientation facilities. The environment also continues to evolve, both in facilities and the intuitiveness of their handling. Auxiliary helpers are more sophisticated, and outlining structured code in the editor is a whole new experience worth investigating just on its own. Outward (MEX) linking to externally developed routines is also expanded.

There are other developments and changes of various types, some of them platform specific. Using Matlab on live work over a period, the net overall effect is a significant gain in responsiveness, sophistication and reach.