Octave vs Matlab

Have assignments for my university in Matlab, the program is provided for us, but would like to look into the FOSS alternative Octave. Other than syntax differences is there any other advantage of Octave other than being open and respecting?

Other urls found in this thread:

en.wikipedia.org/wiki/Mathcad
en.wikibooks.org/wiki/MATLAB_Programming/Differences_between_Octave_and_MATLAB
jakevdp.github.io/PythonDataScienceHandbook/
twitter.com/NSFWRedditGif

>any other advantage of Octave
Octave's usage is to be somewhat compatible with Matlab. It's slower (and no JIT), misses whole Simulink, misses some GUI creation tools... and it follows in everything it provides. There is no advantage beside the obvious being free of price and licensing hell and ported to many systems.

Just to be clear am I able to Mathlabs syntax in Octave? Or are there many roadblocks?

Octave is generally slower and is essentially a clone of Matlab without the toolboxes, so it's always playing catch up to an already shitty """"""language""""". The only advantage I can think of is that you can run Octave scripts on multiple machines in parallel because of Matlab's absurd license.

If you really want FOSS alternatives you can look into Python (SciPy and NumPy) even though using Python for Matlab stuff is verbose as fuck, or look into Julia, which is what Matlab would look like if it wasn't cobbled together by a bunch of greedy incompetent academics that never learned any programming after the 1980s.

Get Mathcad.
en.wikipedia.org/wiki/Mathcad

Overall yes, it should just work
en.wikibooks.org/wiki/MATLAB_Programming/Differences_between_Octave_and_MATLAB

Iirc there are some details that you have to keep in mind. Octave syntax is much more permissive than Matlab so Matlab stuff should always run in Octave but the other way around not always.

>posting proprietary (just like matlab) and single platform only (unlike matlab and octave) as an alternative to either of them.
you're not the smartest one in your family are you?

>Python (SciPy and NumPy)
I'll look into it, but will only use it on my own time. At this point I just want to get my assignments done.
Thanks user!

I haven't used Octave before, but I would not be surprised if MATLAB have claimed proprietary API function names. I remember that you can't even use their 2D color mappings for plotting outside MATLAB (check the video about how the numpy color mapping 'viridis' came to existance)

it is so strange that matlab took off so much.

it is absolutely terrible for large projects. a separate file for every functions makes it a pain in the ass to debug and maintain. they added the next gen functuonality of defining a function in a script file literally like a year ago. for small calculations the heavy ui is unnecessary. it's too slow for serious high performance simulations. simulink is objectively inferior to labview. it's expensive as fuck. 2 versions ago they changed the way settings are stored and so some 3rd party toolboxes dont work.

i cant wait until this meme goes away and is replaced by python at least.

Excuse my ignorance, but what does SciPy even do?

if you ever decide to look on python as alternative, this is handful link
jakevdp.github.io/PythonDataScienceHandbook/

SciPy is a library for Python to do science things that programmer don't usually need, like signal processing, optimization, differential equations ,etc.
Matlab is never going away no matter how hard I wish they would. Their entire business model is they give away their software to schools for low cost so that the only programming that engineers and scientists knows coming out of school is Matlab.

also they are not even similar
Mathcad is cas like Mathematica, not numerical computing software as Matlab is

just use python

Why can you not use R?

Neat, my pirated license for r2011a is expired. Sounds like the new versions are retarded too.

Because his assignments are in Matlab.
But if given choice I'm with R as well.

>Other than syntax differences is there any other advantage of Octave other than being open and respecting?
Yes. It sometimes even is a better language, since eg. you can do:
ones(10)(1,1)
And thus creating and selecting from a Matrix at the same time, this does not work in matlab, so be weary if you are trying to port code backwards.

There are also are a few drawbacks, octave is generally slower (matlab is quite fast when ti comes to matrix computations) and it lacks a few nice features, such as the symbolic toolbox and other stuff.

Octave is garbage and it does the free software world a disservice by its mere existence. The "there's always octave" meme leads many students to believe there's an alternative once they've finished their studies and no longer have a valid matlab license. It would be better if this argument could not be used, that they knew that they were shafted once they're done with their studies and they aren't employed by one of the handful of shitty companies that actually use matlab. This is closer to the truth, too, since the toolboxes are what make matlab powerful, not the actual program.

>matlab is quite fast when ti comes to matrix computations
This is all LINPACK/LAPACK code. Is there any evidence that matrix computations are faster in matlab than in octave (or numpy etc)? I'm willing to bet not, unless you mean special proprietary edge-case algorithms that very few people will actually use

I actually looked it up and it appears that you are right, matlab and octave vary wildly in speed (in both directions), but both are very comparable when it comes to actual matrix computations.
At least that was the case in the (quite old) benchmarks I looked at.

It's almost like essentially every matrix operation calls down to the same C/fortran libraries... like any reasonable author of scientific software will have close to the same performance in matrix code if the matrix model is reasonable?
what a concept.