Mario Rodriguez <biomates <at> telefonica.net> writes:
>
> Hello Reco,
>
> > Being so if i change the labels , and do zlabel="Y axis" ,
> > ylabel="Z label" (in the draw3d instruction), i get what i want ,
> > but , to maintain consistency , i still need to swap the values -m-
> > and -n-, in all vector definitions!And it is a prone error method , i
> > 'd like to know if there is a way to simply swap , axis drawn in the
> > gnuplot window , say swap z with y , for example.
> >
>
> You can't do that from Maxima's syntax. A quick look at the Gnuplot
> manual didn't give me any idea on how to do that.
>
> Some time ago, someone suggested me the idea to introduce the
> possibility of making transformations before rendering, like rotations
> or symmetries. For example, you define a basic function like x^2+y^2 and
> then apply to it translations and/or rotations in order to orientate it
> in any direction. Maybe such a feature could be of some help in cases
> like yours.
>
> > The output of the former expresssion , that originates vector vx ,
> > ( = linsolve_by_lu(A,b) ) returns what seems a vector , as maxima
> > interprets it.So when i try to use this value to draw this vector
> > (with the former draw3d() instrcution) i get the error :
> >
> > draw (vector): coordinates are not correct
> >
>
> Graphic object 'vector' expects two lists as arguments, and the output
> of linsolve_by_lu contains a column matrix. You have to transform it
> into a list; try
>
> (%i1) display2d:false $
> (%i2) linsolve_by_lu(matrix([5,7],[9,6]), matrix([1],[1]));
> 0 errors, 0 warnings
> (%o2) [matrix([1/33],[4/33]),false]
> (%i3) first(transpose(first(%)));
> (%o3) [1/33,4/33]
>
> --
> Mario
>
Hello Mario
Thank you very much for your answer.
In deed one of my first approaches was to transform the , already calculated ,
vectors, and swap the second and third direction cosines (as referred in the
original message).And ironically all this work is to study quaternions and
rotations in 3d.
I implemented an algorithm for structural analysis in Fortran95 (or at least im
trying to ... ).But during the Newton.Raphson resolution method (that is used in
the program) the solution starts diverging(?) and I needed to validate some
parts of the program using maxima , because I still didn't find the cause for
the problem.
Anyway I'm elucidated now ,
Thanks ,
FG