This is the P2PU Archive. If you want the current site, go to www.p2pu.org!
You haven't posted any discussions yet.
Problems that are solved using a numerical computation tool such as Scilab/GNU Octave/Matlab/Python and firends were traditionally solved using programming languages such as Fortran and later on C/C++. In fact, the motivation for developing Matlab was to serve as a wrapper over well established Fortran numerical computation libraries such BLAS, LAPACK. What is your view or experience on the merits and demerits of using such tools in place of traditional programming languages?
I didn´t use any of those programs in past.
I really enjoy Mathematica, especially it's analytical computation, and infinite precision math. The wrap around is very good and useful, since you don't need to program a specific integration routin, or track errors, or worry about memory management.
By "analytical computation" do you mean numerical computation? Could you explain "infinite precision math"? Does it mean high precision numbers? In fact, Scilab is a numerical computation software and has high precision numbers. It does not have a built-in Computer Algebra System (CAS) although a toolbox is available. But CAS is one area, as far as I know, Scilab is not good at.
By "analytical computation" I meant "symbolic".
As for the infinite precision, here is a link:
http://reference.wolfram.com/mathematica/tutorial/ArbitraryPrecisionNumb...
Basically you can increase precision as much as you want, with infinite the software does calculation (with whatever precision) to obtain accuracy in the results.