Re: [PyOpenGL-Users] size limit of vertex array?

[Roland E. <r.e...@gm...>]

Ian,

Here is the code to create the list of vertices:

         self.vertices = []
         self.normals = []
         self.nbrindices = 0
         unit = 1.0
         y = 0.0
         rows = 8
         columns = 8
         z_center_delta = unit * (float(rows) / 2.0)
         x_center_delta = unit * (float(columns) / 2.0)
         for i in range(rows):
             delta_z = (unit * float(i)) - z_center_delta
             for j in range(columns):
                 delta_x = (unit * float(j)) - x_center_delta
                 self.vertices += [0.0 + delta_x, y, 0.0 + delta_z,
                                   0.0 + delta_x, y, unit + delta_z,
                                   unit + delta_x, y, unit + delta_z,
                                   unit + delta_x, y, 0.0 + delta_z]
                 self.normals += [0.0, +1.0, 0.0, 0.0, +1.0, 0.0, 0.0, 
+1.0, 0.0, 0.0, +1.0, 0.0]

         self.nbrindices = len(self.vertices) / 3
         self.indices = range(self.nbrindices)


In addition to creating the squares, the code try also to center the 
plane on the origin.

Next is the code used for the rendering:

         glPolygonMode(GL_FRONT_AND_BACK, GL_FILL)

         glEnableClientState(GL_VERTEX_ARRAY)
         glEnableClientState(GL_NORMAL_ARRAY)

         glVertexPointer(3, GL_FLOAT, 0, self.vertices) # Specify the 
vertex list to be used to draw the object
         glNormalPointer(GL_FLOAT, 0, self.normals) # Specify the 
normals list to be used to draw the object

         glDrawElements(GL_QUADS, self.nbrindices, GL_UNSIGNED_BYTE, 
self.indices)

         glDisableClientState(GL_VERTEX_ARRAY)
         glDisableClientState(GL_NORMAL_ARRAY)



Thanks for the help,


Roland.




Le 07/13/10 16:39, Ian Mallett a écrit :
> Hi,
>
> The bounds of a vertex array or VBO object should ultimately be 
> constrained by the amount of RAM and video memory you have.  I've 
> personally used vertex arrays of larger size than 64*4*3, so 
> something's wrong.
>
> What specifically goes wrong when you add more?  Can we see some code?
>
> Ian