Re: Getting Matlab geophysics code running in Octave

[Juan Pablo Carbajal]

On Sun, Aug 4, 2013 at 12:17 PM, Lester Anderson <address@hidden> wrote:
> Hello,
>
> I have an issue getting some Matlab code running in Octave (3.6.4) and have
> managed to get part way into the problems, but it is still not functioning
> as expected, i.e. the output is not coming up for one.
>
> A lot of code has - defval('T',35e3) for example, so I just changed lines
> like that to T=35e3, however I am not sure this is the intention?
>
> Other times I have seen code written like here:
>
> % Te=[40 30 20 10 5 0]; <----------------------------  user input of Te
> values as a vector
> % [l,Zb,G2b]=mckenzie([0 2670 2670+670],[0 35],[1 0],Te(4)); % Top loading
> % [l,Zb,G2b]=mckenzie([0 2670 2670+670],[0 35],[0 1],Te(4)); % Bottom
> loading
> %
> % See FORSYTH, MCKENZIE1, MCKENZIE2, LOADING
> %
> % Last modified by fjsimons-at-alum.mit.edu, October 21st, 2003
>
> defval('R',[1030 2600 2900 3300])
> defval('T',[      0     15   35])
> defval('F',[    0.6    0.4    0])
> defval('Te',17)   <----------------------  A value of Te is given here, but
> above a range of values is given!
> defval('E',1.4e11);
> v=0.25;
> g=9.81;
>
> Again, I can change the defav to R=[......], and it will plot sometimes, bu
> only one value (in this case Te=17 and not the range [40 30 20 10 5 0].
>
> I am not an expert on Matlab, but there are obvious differences with Octave
> (and Scilab too). I would be grateful if I could get help to get the
> attached code to run and generate the output (image - coheref2.gif). The
> code is freely available at:
> http://geoweb.princeton.edu/people/simons/software.html.
>
> The function coheref2.m calls forsyth.m and both are attached in the
> original Matlab.
>
> The lines like:
>
>   defval('g',fralmanac('GravAcc'));
>   G=fralmanac('GravCst'); % calls a file with a load of constants
>
> can be replaced with the constants:
>
> g=9.81 % gravity acceleration
>
> G=6.67e-11 % Universal gravitational constant
>
> The big advantage of Octave over Scilab, is that generally most code will
> run straight "out of the box" whereas in Scilab it has to be converted (and
> tweaked).
>
> If I can see how it should be written then it will be easier for me to sort
> the rest :)
> Thanks for any help and guidance, apologies for the long-winded query.
>
> Cheers
>
> Lester
>
>
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://mailman.cae.wisc.edu/listinfo/help-octave
>

Checking mathworks help I can't find a function "defval" (though it
may be due to my horrible connection and their heavy site). If you do
not find the help of a defval function it means the person who created
the package you want to use has it. In that case is not a Matlab -
Octave problem but just that the code you received is incomplete or
the paths are not well defined.

Please double check. If you find a mathworks defval let us know what
it does and we can suggest alternatives or implement it.

Cheers