POV-Ray : Newsgroups : povray.general : Mesh Q's about Rendering & Scale : Re: Mesh Q's about Rendering & Scale Server Time
31 Jul 2024 14:32:44 EDT (-0400)
  Re: Mesh Q's about Rendering & Scale  
From: Warp
Date: 20 Dec 2006 11:27:35
Message: <45896476@news.povray.org>
Bryan Heit <bjh### [at] nospamucalgaryca> wrote:
> 1) In terms of rendering time, would it be quicker to use the mesh, or 
> to convert the mesh into a 16-bit height_field, like what is done in the 
> height_field tutorial of the documentation?

  A height_field should render faster than a mesh, but take into account
that height_field uses a different normal smoothing algorithm than meshes,
so it may end up looking different at places.

  Try both to see what happens.

> 2) If rendering the mesh is faster, how do I determine the size of the 
> mesh?  I know it's not coming in as a 1x1x1 sized object, like the 
> height_field would be.  In my test run, centering required that I 
> translate it by <0.05,0.1375,0.34>, and even that is my best guess.

  You can use min_extent() and max_extent() to get the opposite corners
of the bounding box of the mesh (which povray calculates automatically).
AFAIK with a mesh the bounding box should be a perfect fit.

> 3) According to the documentation, a Mesh2 object is better (and faster) 
> for povray.  If their any convenient way to convert my mesh into a mesh2?

  A mesh2 is only faster to parse compared to mesh. Once parsed, there's
no difference.

  Are you sure poseray doesn't support saving as a mesh2? That would be
strange, given that it's a rather new program.
  If not, perhaps try to find another converter. Or as a last resort,
try this:

http://warp.povusers.org/WinMeshcomp30_beta3.zip

-- 
                                                          - Warp


Post a reply to this message

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