You are here

SPECTRUM--A review


Brian Pankuch

Note: This article was scanned using OCR from the Fall 1993 CCCE Newsletter. Please contact us if you identify any OCR errors.
          This set of tools is primarily for use in analytical chemistry. I spent hours exploring and it is really interesting to use. Having done many of the operations by hand, well with a calculator point by point, it is a great feeling of power to run an absorption or NMR spectra through the many available operations and transformations. If you've done it by hand and haven't tried it by computer you are in for a treat.
          By transferring the files on the program diskette to your hard drive and double clicking on the program icon you are up and running. I'd suggest going through a few pages of the instruction manual to get a feel for the program. The program is straight forward to use. It is very intuitive. The 97 page manual is excellentmostly screen shots showing the before and after using a transformation or operation. The manual is very easy to read and follow.
          Despite playing 'what happens if I do this?', I only got Spectrum to crash twice. Considering that I can do this with every program I've ever used without trying, this means that the program is quite stable and robust. Although primarily for those handling data from any type of spectra or graph it is easy enough to use that I'm considering using it for data manipulation and teaching some basics. Of course it would help if we had computers to use in the lab.
          The program is designed for advanced and graduate level analytical courses. Strong organic students using gas chromatography, etc., would find parts of the tutorial quite worthwhile.lfthe data could be transferred directly from the instrument to Spectrum it would be worth spending sometime showing students some oft he rudiments of cleaning up a signal. Although using and understanding all the features would certainly take a lot of expertise, just using some of the basics shown in the following figures (from signals supplied with the program) with less advanced students would seem worthwhile.
          The tutorial has a variety of signals and suggested operations to demonstrate signal processing. The tutorial is informative. Having done some of these techniques by hand I can really appreciate the power at my fingertips. I'd recommend getting this program even if you don't have an immediate application. It is good enough and easy enough to use that you will minimally learn a bit and probably find some uses that I Add constant Multiply by constant haven't thought of. Data files may be prepared in a text editor, spreadsheet, telecommunications programs, plotting or data acquisition program, and saved in ASCII (text only) format. You can also use the New command under File in the menu to generate synthetic signals. The signals can have optional noise on Gaussian, Lorentzian or sine waves.
          The program is extendible, i.e., you can add your own functions. On the Mac you would need Borland's, Turbo Pascal1.1 and Turbo Pascal Numeric Methods Toolbox. Since I have neither of the above nor a copy of the source code (the actual line by line code for the program as opposed to the compiled code that is what we usually use) for Spectrum, I was not able to actually check the actual procedure. You would also neerl ResEdit from Apple. Suggestions for making changes looked straight forward if you are a Pascal programmer.
          SPECTRUM was the 1990 NCR I PTAL winner. It is priced at $49.95 for a single copy, $249 for a department, $395 for all campus sites. It is available from: Academic Software Develpment Group, Computer Science Center, Room 3357, University of Maryland, College Park, MD 20742. Phone: 301-405-7600.


02/01/93 to 02/05/93