Ctioga is deprecated

As of now (Mon Jan 31 2011), the use of Ctioga is deprecated, in favor of its full rewrite ctioga2. Development had stopped already two years ago. Please use ctioga2, which is now mature enough to serve as a full replacement of Ctioga, but provides much more features, such as XYZ color maps.

You may be interested in this page to help you transition to ctioga2.

Why should I use ctioga ?

Well, I’ll answer that by saying why I did write it. I’m a command-line guy, and I used to work a lot with gnuplot. But I was somehow unsatisfied with it. Here is what I think ctioga has for advantages over gnuplot.

Fast to use

If you want to look at the contents of a file, with gnuplot, you need the following

gnuplot
plot 'file.dat'

Well, that doesn’t look awful. But there’s no automatic completion, as you’re within gnuplot. And when you’ve got a lot of datafiles with long names, the shell autocompletion makes a difference. With ctioga, it is enough to type…

ctioga file.dat

And there is autocompletion.

Based on Tioga

Tioga is a powerful plotting library from one of the authors of the Postscript language, Bill Paxton. I believe the plots produced by Tioga look way nicer than any plot made by gnuplot, even with quite a lot of fine-tuning.

Smooth integration within LaTeX documents

As the text typeset inside the graphes is processed with pdflatex, you have complete control over it. In particular, it is easy to make publication-quality graphes where it is impossible to tell the difference between the text inside the graph and the text around.

Support for arbitrary data sources

Ctioga is based on the concept of Backends, little black boxes which are in charge of gathering the data referred to on the command-line. The architecture is completely transparent and it is really easy to add a new backend — and to support new data files.

Easy to script

With ctioga, it is easy to write scripts that update a whole set of plots in a single run — very practical when you realize at the moment of submitting an article that your plots should have been black and white instead of color…