> And that was before numerical python existed. Now, MATLAB compatibility is about the only reason to use Octave. But a better idea is to get out of the MATLABi

> And that was before numerical python existed. Now, MATLAB compatibility is about the only reason to use Octave. But a better idea is to get out of the MATLABiverse altogether. Personally, I do a lot of numerical calculus and prefer octave than numpy, just because the language is less verbose and more comfortable to use. I have converted many times ugly python+numpy code into beautiful octave. If your main and only goal is to perform matrix computations, then octave is excellent. I have never used nor needed the commercial matlab. The python numerical stack has a rather absurd feeling in my eyes.