Key Differences Between Octave & Matlab

Key differences of which we are aware

	Octave	Matlab
character string syntax	str='hi' str="hi"	str='hi' str="hi"     % gives error
for loop syntax	for   ... endfor	for   ... end
Hack using ~ as a blank variable. For example in the line [A B]=max(C), if the variable A is not needed you could write the line as [~ B]=max(C) and the the first output will not be created. This yields an error in Octave (and probably should in matlab).	not supported	[~ B]=max(C)
The input function in Octave does not allow the use of slash commands.	input('Input \nis:') yields: Input \nis:	input('Input \nis: ') yields: Input is:
There seems to be some issues with placing text in a plot on the complex plane using the text command.	not supported, use x,y instead.	text() can take complex inputs and separate the real and imaginary part for a 2 point coordinate.
Some I/O functions such as textscan()	Get recent version, 3.4.0 or later	Included
Some image file reading functions	Some functions not in basic mac distribution. Not sure about windows.	Included in basic distribution or various optional toolboxes


The wiki page http://en.wikibooks.org/wiki/MATLAB_Programming/Differences_between_Octave_and_MATLAB documents some minor differences--mainly cases where Octave works and matlab does not.

Installing special functions in Octave

In some cases, functions are needed that are not included in the regular Octave distribution. These are often ones that are in the Matlab toolboxes such as the Signal Processing Toolbox, which you should have bundled with your student version of Matlab. If you are using Octave you will need the Signals Package to use these functions.

Also make sure to note that if you are using Octave, this program's version of the function "firpm" (Remez Parks-McClellan Optimal FIR Filter Design), might be called "remez" and should work the same as "firpm" works in Matlab.

If you get something like the following error message:
      ??? Undefined function or method 'fft' for input arguments of type 'double',
it means you do not have the function installed in Octave or Matlab. So if you see this error while trying to access any of the signal processing functions, if you are using Octave, do the following instructions to install the Signals Package (and all of its dependency packages), if you are using Matlab and get this message, you will need to purchase (if for some reason you did not get the student version) and/or install the Signal Processing Toolbox, if you can use the functions, you should disregard the rest of this message. Some of your Octave programs might have come with these packages pre-installed.

To install the Signal Package (and all of its dependency packages) in Octave:

1. Go to: http://octave.sourceforge.net/
2. Go to the Packages section
3. Download the "miscellaneous", "struct", "specfun", "optim", and "signal" packages
4. Start up Octave
5. Install those packages in the order given in step 3., but changing your working directory to where you saved the packages (i.e. "cd Downloads") and typing: "pkg install package_file_name.tar" where package_file_name is the name of the package you are installing.
As long as no errors occurred, you should now be able to use the signal processing functions needed for this problem, remember the "firpm" function might be called "remez" in your Octave installation.