After a false start a month or more back, I recently resumed trying to 
use Thomas's peeling paint macro texture as the basis for an isosurface.

The attached image is a single isosurface where his macro was used to 
create two complementary black and white images each blurred/played with 
in Gimp. The blurred true image is used for where there is paint and the 
false blurred for displacement. There is a third povray image which 
serves as the wall behind the paint

It is a single isosurface because I had the idea to do another color of 
paint on the underside of the peel, but in the end didn't get that far. 
  Guess I'd say the isosurface technique sort of works though a bit hard 
to control. Expect other approaches to “real” peeling would be tricky 
too. About 15 minutes for a 1 light render at 1920x1200.

Bill P.

Post a reply to this message

Attachments:
Download 'peelingpaintwall.jpg' (493 KB)

Preview of image 'peelingpaintwall.jpg'

Goodness!

This is really looking good indeed. I think you have tackled the problem 
in a very elegant way, even if the isosurface is difficult to control. I 
am impatient to try it out!

Thomas

Post a reply to this message

Holy cow - that's an amazing effect!
(I like the graffiti)

You clearly have a deep understanding and excellent ability to utilize
isosurfaces in a way I haven't even begun to understand.

I have some questions -

Can you increase the "height" of the peel?
Does this method allow for isolated flakes to appear?
Is it possible to define the parameters used to make this sort of thing
infinitely tilable?
And lastly, do you have any examples or tutorials you can point to, for your
viewers to play along at home?   :)

Thanks!  Great work, and an impressive result.

Post a reply to this message

Am 16.08.2014 03:17, schrieb William F Pokorny:
> After a false start a month or more back, I recently resumed trying to
> use Thomas's peeling paint macro texture as the basis for an isosurface.

This looks incredibly credible!

Post a reply to this message

> Holy cow - that's an amazing effect!
> (I like the graffiti)
>
> You clearly have a deep understanding and excellent ability to utilize
> isosurfaces in a way I haven't even begun to understand.
>
> I have some questions -
>
> Can you increase the "height" of the peel?
> Does this method allow for isolated flakes to appear?
> Is it possible to define the parameters used to make this sort of thing
> infinitely tilable?
> And lastly, do you have any examples or tutorials you can point to, for your
> viewers to play along at home?   :)
>
> Thanks!  Great work, and an impressive result.
>
>
>

If the isosurface function looks like z-InagePattern(x,y,0)*1, changing 
the *1 to *2 will double the hight.

The pattern is controled by two images. If those images are tileable, 
then, the resulting isosurface can tile.

The same will apply to create isolated flakes: If the source image 
contains isolated areas, those will show as isolated flakes.


Alain

Post a reply to this message

On 16-8-2014 20:07, Alain wrote:
> If the isosurface function looks like z-InagePattern(x,y,0)*1, changing
> the *1 to *2 will double the hight.

However, it may be that /all/ parts of the isosurface double in height 
and that would not be desirable. Only the peels.

Thomas

Post a reply to this message

On 08/16/2014 02:07 PM, Alain wrote:
>
> If the isosurface function looks like z-InagePattern(x,y,0)*1, changing
> the *1 to *2 will double the hight.
>
> The pattern is controled by two images. If those images are tileable,
> then, the resulting isosurface can tile.
>
> The same will apply to create isolated flakes: If the source image
> contains isolated areas, those will show as isolated flakes.
>
> Alain

Thanks for answering Alain.

I'll add that the component shape functions for the wall and paint are 
thin - using scaled abs(z)-image. While the peel height can be increased 
or decreased, it is just a displacement in z and so the struggle is 
keeping a realistic peeling look as the displacement amount changes.

Related, it is possible to play with the gamma used while reading the 
image to change the blur profile to perhaps get more of a curl to the 
peel, but I've not yet played with that technique on this peeling paint 
isosurface.

I've posted a zip of code posted to p.b.scene-files.

Bill P.

Post a reply to this message

> On 16-8-2014 20:07, Alain wrote:
>> If the isosurface function looks like z-InagePattern(x,y,0)*1, changing
>> the *1 to *2 will double the hight.
>
> However, it may be that /all/ parts of the isosurface double in height
> and that would not be desirable. Only the peels.
>
> Thomas
>
>

In that case, it may be interesting to play with the contrast or the 
gamma curve used for the base image. If playing with the gamma, saving 
it in a format that don't keep that information would be advisable.

Post a reply to this message

William F Pokorny wrote on 16/08/2014 3.17:
> After a false start a month or more back, I recently resumed trying to
> use Thomas's peeling paint macro texture as the basis for an isosurface.

A really good work, thank you!
Paolo

Post a reply to this message