POV-Ray : Newsgroups : povray.general : GLP v0.9a : Re: GLP v0.9a Server Time
1 Aug 2024 20:14:38 EDT (-0400)
  Re: GLP v0.9a  
From: David Curtis
Date: 20 Sep 2005 09:08:06
Message: <433009b6$1@news.povray.org>
"Christoph Hormann" <chr### [at] gmxde> wrote in message 
news:dgore1$6fm$1@chho.imagico.de...
> David Curtis wrote:
>>
>> Here's some differrences to that patch.
>> GLP - has interactive view(orbit, pan, zoom, ...).
>>         - does wireframe and solid faces.
>>         - allows multiple camera definitions in scene file
>>         - has a keyword that controls density of generated mesh.
>>
>
> Let me see i understand this right - you parse the scene as usual but 
> instead of rendering you generate an OpenGL scene from the parsed scene 
> file, display it and allow to modify the camera?

While the scene is parsed an opengl window is displaying objects as they are 
parsed. After parsing the normal rendering takes place. A tab control lets 
you switch between render and opengl views, during rendering and after.

> First of all it would of course be important to what extent the OpenGL 
> view resembles the parsed scene.  What shapes, light sources and cameras 
> are aupported?  Does it support texturing of some sort?  How does it do 
> CSG?

Camera matching is currently limited to a 1.333 width-to-height aspect 
ratio. The most common that I use.

Currently supported objects are: boxes, spheres, cylinders, cones, mesh, 
mesh2, heightfield and isosurface. Point light sources and perspective 
camera only for the moment.

No texturing. OpenGL color is taken from pigment. In case of a patterned 
texture use of the quick_colour token is supported. With a toggle button you 
can switch all objects to user-definable color.

No CSG. (probably ever) Component objects are displayed fully, no trimming, 
etc.

> And the most important: How does the interactive view work?  Do you modify 
> the camera data and use it when rendering afterwards or do you modify the 
> scene file with the changed camera parameters?  Or do the changed camera 
> parameters have to be inserted into the scene manually?

Of course Frame.Camera is destroyed after rendering, so I keep a copy of 
camera *definitions* (currently stores up to 8 cameras from scene file), 
which can be modified in the opengl window. You can paste the opengl camera 
def into the scene file, or you can render opengl view by menu command to 
render whaterver is displayed in opengl, without modifying scene file. The 
render view code is not complete(actually broken) currently. It renders but 
the direction vector isn't correct yet.

Dave.


Post a reply to this message

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.