Chrome die on a checkered plane, with varying surface roughness.

Renders were done using POV-Ray 3.7 with a blurred reflection patch; 
highlights are specular highlights carefully tuned in both intensity and 
roughness to match the reflection.

Roughly 2 minutes of parsing time for the die mesh, plus 20 minutes 
rendering time. Not to forget the one-time 8 minutes or so to generate 
the mesh from an isosurface.


Before you get over-excited: No, I do /not/ intend to push this patch 
into POV-Ray 3.7 proper. Instead, I plan to expand it into a full-blown 
stochastical rendering patch, akin to MC-Pov.

There - that's it, now you know. The die is cast. Alea iacta est.

Post a reply to this message

Attachments:
Download 'steeldie 2011-11-22 1854.png' (154 KB) Download 'steeldie 2011-11-22 1925.png' (167 KB) Download 'steeldie 2011-11-22 1955.png' (180 KB) Download 'steeldie 2011-11-22 2018.png' (181 KB)

Preview of image 'steeldie 2011-11-22 1854.png'

Preview of image 'steeldie 2011-11-22 1925.png'

Preview of image 'steeldie 2011-11-22 1955.png'

Preview of image 'steeldie 2011-11-22 2018.png'

On 22-11-2011 22:34, clipka wrote:
> Before you get over-excited: No, I do /not/ intend to push this patch
> into POV-Ray 3.7 proper. Instead, I plan to expand it into a full-blown
> stochastical rendering patch, akin to MC-Pov.
>
> There - that's it, now you know. The die is cast. Alea iacta est.

I must say that this is the best isosurface to mesh patch I have seen 
until now, and fast too. Congrats with that Christoph!

I can live with a full-blown rendering patch, but any reason you can 
give us why not in 3.7?

Thomas

Post a reply to this message

Am 23.11.2011 09:43, schrieb Thomas de Groot:

> I must say that this is the best isosurface to mesh patch I have seen
> until now, and fast too. Congrats with that Christoph!

Um... I think we have a misunderstanding here: The patch I wrote does 
the blurred reflections (and auto-adjustment of specular highlight 
parameters to match the reflections). The isosurface to mesh conversion 
was done with some classic SDL-based isosurface-to-mesh converter by... 
let me see... there: Kevin Loney, Jaap Frank & Tor Olav Kristensen 
(Kudos to them).

> I can live with a full-blown rendering patch, but any reason you can
> give us why not in 3.7?

As for the blurred reflections, they need serious oversampling, and may 
lead to bright speckles otherwise; Therefore some mechanisms need to be 
implemented for the feature to work stand-alone (currently I have to use 
focal blur to achieve the oversampling) which will make it a piece of 
cake to also add a stochastic implementation for what's currently done 
by radiosity.

Post a reply to this message

On 23-11-2011 14:39, clipka wrote:
> Um... I think we have a misunderstanding here: The patch I wrote does
> the blurred reflections (and auto-adjustment of specular highlight
> parameters to match the reflections). The isosurface to mesh conversion
> was done with some classic SDL-based isosurface-to-mesh converter by...
> let me see... there: Kevin Loney, Jaap Frank & Tor Olav Kristensen
> (Kudos to them).

Right! I thought it was a new one. I know the one you mention here 
although I have not used it for many geological eras.

> As for the blurred reflections, they need serious oversampling, and may
> lead to bright speckles otherwise; Therefore some mechanisms need to be
> implemented for the feature to work stand-alone (currently I have to use
> focal blur to achieve the oversampling) which will make it a piece of
> cake to also add a stochastic implementation for what's currently done
> by radiosity.

I see.

Thomas

Post a reply to this message

On 11/22/2011 11:34 PM, clipka wrote:
> Renders were done using POV-Ray 3.7 with a blurred reflection patch;
> highlights are specular highlights carefully tuned in both intensity and
> roughness to match the reflection.

   Is this different from the averaged textures trick?

Post a reply to this message