Xplo Eristotle wrote:
> 
[...]
> 
> The 5 1/2 minute render was probably the result of having such a simple
> scene and such a low recursion limit. I tried a sphere with a granite
> normal and set normal on for the radiosity to see how it would come out.
> I suppose it rendered acceptably fast, but it certainly wasn't done in
> five minutes...
> 
> (BTW, it appears crooked because I stuck it on the leftmost high disc
> and didn't bother to re-aim the camera.)
> 

I made some comparision between normal and isosurface bumps in this scene and
the result seems interesting ( thanks to Xplo Eristotle for sharing the
code ).

Of course the rendering time is much longer with the isosurface version ( more
than 1 hour for the second pict)

In addition to the effects you would expect the version "sphere with normal
seems much smoother than the isosurface even though they both use the same
function.  

I'm not sure whether this has been discussed before, but the difference seems
quite important to me.

Christoph  

To the single picts:

1) isosurface with light pigment function
2) isosurface with stronger pigment function
3) same function as normal 1.0
4) same function as normal 2.5

--
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

Attachments:
Download 'iso_rad_01.jpg' (11 KB) Download 'iso_rad_02.jpg' (13 KB) Download 'iso_rad_03.jpg' (9 KB) Download 'iso_rad_04.jpg' (11 KB)

Preview of image 'iso_rad_01.jpg'

Preview of image 'iso_rad_02.jpg'

Preview of image 'iso_rad_03.jpg'

Preview of image 'iso_rad_04.jpg'