I have experimented a little with Ryoichi Suzuki's
"Density_File Extension Patch", available here:

http://staff.aist.go.jp/r-suzuki/e/povray/iso/df_body.htm

This POV-Ray patch has a feature for sampling functions
and storing the result in a density file or in a pattern.
(The patch will interpolate values when "reading" back
values at locations between the sample points.)

I have used this feature to sample complex functions for
isosurfaces and store the results in pattern functions.

When these sampled functions are used in isosurfaces they
render quite fast compared to the functions I started with.
(I have experienced 3 to 6 times faster renderings.)

The "previews" give a very good indication of how the final
isosurface shapes will be.

I have enclosed two images to illustrate the results.

The image marked FINAL rendered in 2 minutes and 33 seconds,
while the image marked "PREVIEW" rendered in 47 seconds.


Tor Olav

Post a reply to this message

Attachments:
Download 'MultiRotIsoModified02_09_ .jpg' (26 KB) Download 'MultiRotIsoModified02_10_ .jpg' (28 KB)

Preview of image 'MultiRotIsoModified02_09_ .jpg'

Preview of image 'MultiRotIsoModified02_10_ .jpg'

I forgot to mention that I have posted some source code
that shows how to do this. Here it is:

http://news.povray.org/povray.text.scene-files/31765/225568/#225568


Tor Olav

Post a reply to this message

On 26 May 2003 18:34:13 -0400, Tor Olav Kristensen wrote:

> 
> I have experimented a little with Ryoichi Suzuki's
> "Density_File Extension Patch", available here:
> 
> http://staff.aist.go.jp/r-suzuki/e/povray/iso/df_body.htm
> 
> This POV-Ray patch has a feature for sampling functions
> and storing the result in a density file or in a pattern.
> (The patch will interpolate values when "reading" back
> values at locations between the sample points.)
> 
> I have used this feature to sample complex functions for
> isosurfaces and store the results in pattern functions.
> 
> When these sampled functions are used in isosurfaces they
> render quite fast compared to the functions I started with.
> (I have experienced 3 to 6 times faster renderings.)
> 
> The "previews" give a very good indication of how the final
> isosurface shapes will be.
> 
> I have enclosed two images to illustrate the results.
> 
> The image marked FINAL rendered in 2 minutes and 33 seconds,
> while the image marked "PREVIEW" rendered in 47 seconds.
> 
> 
> Tor Olav

That looks like a good previewing technique for isosurfaces although I can 
see some voxel artifacts in the preview image particularly around the 
smooth surfaces. I'm still quite hopeful that tesselation could make for 
good isosurface approximation (recalling an older thread about that) but if 
not, this method looks quite promising. ;-)

-- 
light_source#macro G(E)sphere{z+E*y*5e-3.04rotate-z*E*6pigment{rgbt#end{
20*y-10#local n=162;1}#while(n)#local n=n-.3;G(n)x}}G(-n).7}}#end//GregE

Post a reply to this message