Re: 3d Plot

[Doug Stewart]

On Sat, Jul 11, 2015 at 9:29 AM, Doug Stewart <address@hidden> wrote:

On Sat, Jul 11, 2015 at 2:51 AM, Tatsuro MATSUOKA <address@hidden> wrote:

----- Original Message -----
>From: Ben Abbott 
>To: Thomas D. Dean  
>Cc: address@hidden; Doug Stewart ; Tatsuro MATSUOKA 
>Date: 2015/7/11, Sat 11:30
>Subject: Re: 3d Plot
>
>
>On Jul 10, 2015, at 10:20 PM, Thomas D. Dean <address@hidden> wrote:
>>
>>On 07/10/15 18:57, Doug Stewart wrote:
>>
>
>>
>On Fri, Jul 10, 2015 at 9:25 PM, tmacchant <address@hidden
>>><mailto:address@hidden>> wrote:
>>>
>>>   xx and yy is generated by u and v so that they are not appropriate
>>>   for mesh
>>>   seeing gnuplot example. Questioner wants plot against X and Y mesh.
>>>
>>>   I am writing from my smart phone and I cannot use octave.
>>>
>>>   Tatsuro
>>>
>>>try #2
>>>
>>>[xx,yy]=meshgrid(real(X),real(Y));
>>>Z=z(xx,yy);
>>>mesh(xx,yy,real(Z))
>>>
>>Thanks, Doug.  This does not resemble the gnuplot demo.  This is the surface of a 3d solid.
>>
>>Try the example directly into gnuplot.
>>
>>Tom Dean
>
>
>I don’t think Octave is able to render a parametric surface.
>
>
>The plot rendered by gnuplot is attached for those interested.
>
>
>Ben

Tom seem to want transparent rendering of overlapped 3D surfaces.
AFAIK, octave does not have such rendering, am I right?

Tatsuro 

this is closer

1

function [p]=x(u,v)

  p=2 * (cos(u)+u .* sin(u)) .* sin(v) ./ (1+u .^ 2 .* (sin(v)) .^ 2);

endfunction;

function [p]=y(u,v)

  p=2 * (sin(u)-u .* cos(u)) .* sin(v) ./ (1+u .^ 2 .* (sin(v)) .^ 2);

endfunction;

function [p]=z(u,v)

  p=log(tan(v/2.)) + 2 * 1 .^ cos(v)     ./ (1+u .^ 2 .* (sin(v)) .^ 2);

endfunction;

u=[-4.5:9/99:4.5]; v=[0.05:(pi-0.10)/99:pi-0.05];

[uu, vv]=meshgrid(u,v);

X=x(uu,vv);

Y=y(uu,vv);

Z=z(uu,vv);

scatter3(X,Y,Z)

--

DAS

you have a different z function

function [p]=z(u,v)

 # p=log(tan(v/2.)) + 2 * 1 .^ cos(v)     ./ (1+u .^ 2 .* (sin(v)) .^ 2);

 p=log(tan(v/2.)) + 2 * cos(v)     ./ (1+u .^ 2 .* (sin(v)) .^ 2);

endfunction;

Now it looks to be the same shape. but not a nice as gnuplot.

--

DAS