 |
|
|
|
|
|
| |
| |
|
|
From: Jörg 'Yadgar' Bleimann
Subject: Mesh2 to isosurface - how to do with IsoCSG?
Date: 18 Dec 2010 14:14:00
Message: <4d0d07f8$1@news.povray.org>
|
|
|
| |
| |
|
|
High!
Christoph Hormann's Earth renditions, apparently done with mesh2s
converted to isosurfaces (http://earth.imagico.de/technical.html) made
me drool instantly... and as he had done an include called "IsoCSG", I
wonder whether a mesh2 to isosurface conversion feature was contained
therein. But when I looked over the documentation of version 0.7, I
found no hint about how to do this. Could it be that only a version not
yet released contains this - or do I have to do it an entirely different
way! But there MUST be some way to convert mesh2s into isosurfaces...
See you in Khyberspace!
Yadgar
Post a reply to this message
|
|
| |
| |
|
|
From: Warp
Subject: Re: Mesh2 to isosurface - how to do with IsoCSG?
Date: 19 Dec 2010 08:10:30
Message: <4d0e0446@news.povray.org>
|
|
|
| |
| |
|
|
> Christoph Hormann's Earth renditions, apparently done with mesh2s
> converted to isosurfaces (http://earth.imagico.de/technical.html) made
> me drool instantly... and as he had done an include called "IsoCSG", I
> wonder whether a mesh2 to isosurface conversion feature was contained
> therein. But when I looked over the documentation of version 0.7, I
> found no hint about how to do this. Could it be that only a version not
> yet released contains this - or do I have to do it an entirely different
> way! But there MUST be some way to convert mesh2s into isosurfaces...
Sorry for asking, but for what possible reason would one want to
convert a mesh into an isosurface? A conversion to the other direction
would make sense and, in fact, be really useful (at least in terms of
rendering speed). However, converting a mesh to an isosurface seems to
be completely backwards. Why would one want to do that?
(One could perhaps argue that you can apply any kinds of transformations
to an isosurface. However, you can do that to a mesh too, although it
requires more indirect means. If you put your vertices and normal vectors
in arrays and then create the vertex and normal vectors of the mesh from
those arrays, you can apply any transformations you want to those vertices
in the arrays before creating the mesh, thus achieving free transformations.
While not absolutely trivial, it's not that hard either.)
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
From: Jörg 'Yadgar' Bleimann
Subject: Re: Mesh2 to isosurface - how to do with IsoCSG?
Date: 19 Dec 2010 16:51:22
Message: <4d0e7e5a@news.povray.org>
|
|
|
| |
| |
|
|
High!
On 19.12.2010 14:10, Warp wrote:
> Sorry for asking, but for what possible reason would one want to
> convert a mesh into an isosurface?
Recently, I started building mesh2s following Earth's curvature from
SRTM Earth elevation data (3 arcseconds per data point)... when
rendering those mesh2s, I found some strange artifacts (gaps between
triangles, see image on p.b.i.) which probably resulted from floating
point inaccuracy (the scale I used is 1 POV unit = 1 kilometre).
As Christoph Hormann, as he writes on his website, apparently uses a
kind of (self-written?) conversion tool, I thought that converting my
mesh2s to isosurfaces possible could solve the gaps problem...
See you in Khyberspace!
Yadgar
Now playing: Dis-Gorilla (Rick Dees & His Cast Of Idiots)
Post a reply to this message
|
|
| |
| |
|
|
From: Warp
Subject: Re: Mesh2 to isosurface - how to do with IsoCSG?
Date: 19 Dec 2010 16:58:00
Message: <4d0e7fe7@news.povray.org>
|
|
|
| |
| |
|
|
> Recently, I started building mesh2s following Earth's curvature from
> SRTM Earth elevation data (3 arcseconds per data point)... when
> rendering those mesh2s, I found some strange artifacts (gaps between
> triangles, see image on p.b.i.) which probably resulted from floating
> point inaccuracy (the scale I used is 1 POV unit = 1 kilometre).
I bet it's a bug in your conversion program instead (or, alternatively,
something which gets fixed if you apply double illumination).
> As Christoph Hormann, as he writes on his website, apparently uses a
> kind of (self-written?) conversion tool, I thought that converting my
> mesh2s to isosurfaces possible could solve the gaps problem...
Do you really want your earth model to render a hundred times slower?
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
From: Jörg 'Yadgar' Bleimann
Subject: Re: Mesh2 to isosurface - how to do with IsoCSG?
Date: 19 Dec 2010 17:20:17
Message: <4d0e8521$1@news.povray.org>
|
|
|
| |
| |
|
|
High!
On 19.12.2010 22:58, Warp wrote:
> Do you really want your earth model to render a hundred times slower?
When I want to render animations, really not... but high-quality still
images ("When the Internet came to Abdurrahmankhan Kalay") probably
might improve!
See you in Khyberspace!
Yadgar
Now playing: Love Or Nothing (Uriah Heep)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> High!
>
> Christoph Hormann's Earth renditions, apparently done with mesh2s
> converted to isosurfaces
???
No, apparently not. He describes quite clearly that he uses the
elevation data to create an isosurface-function. I'm pretty sure to
remember that on his homepage is even a full blown tutorial including
POV code on how he did it - but right now I'm too lazy to search the
link for you.
> and as he had done an include called "IsoCSG", I
> wonder whether a mesh2 to isosurface conversion feature was contained
> therein. But when I looked over the documentation of version 0.7, I
> found no hint about how to do this. Could it be that only a version not
> yet released contains this - or do I have to do it an entirely different
> way!
IsoCSG is about doing CSG with isosurfaces (therefor the name, otherwise
I bet Christoph would have named it IsoMesh).
-Ive
Post a reply to this message
|
|
| |
| |
|
|
From: stbenge
Subject: Re: Mesh2 to isosurface - how to do with IsoCSG?
Date: 19 Dec 2010 20:40:03
Message: <4d0eb3f3@news.povray.org>
|
|
|
| |
| |
|
|
On 12/19/2010 5:10 AM, Warp wrote:
> Sorry for asking, but for what possible reason would one want to
> convert a mesh into an isosurface? A conversion to the other direction
> would make sense and, in fact, be really useful (at least in terms of
> rendering speed). However, converting a mesh to an isosurface seems to
> be completely backwards. Why would one want to do that?
Why wouldn't one want to do it? If it could be done efficiently, a
person could give erosion to a statue, or add rusty lumps to an object,
at will.
The big question is how to do it? You could make an object pattern from
the mesh, and convert that to an isosurface. But that always results in
a lot of artifacts. Or you could average jittered copies of the object
pattern together, but that renders very slowly and the artifacts are
still present (though somewhat reduced). Or perhaps... you could start
with a spherical function and push and pull space at each triangle thus
distributing it into the mesh's shape. It's only a theory, and might
only make sense for small meshes, but maybe it would work...
Sam
Post a reply to this message
|
|
| |
| |
|
|
From: Warp
Subject: Re: Mesh2 to isosurface - how to do with IsoCSG?
Date: 20 Dec 2010 02:10:33
Message: <4d0f0169@news.povray.org>
|
|
|
| |
| |
|
|
stbenge <myu### [at] hotmail com> wrote:
> On 12/19/2010 5:10 AM, Warp wrote:
> > Sorry for asking, but for what possible reason would one want to
> > convert a mesh into an isosurface? A conversion to the other direction
> > would make sense and, in fact, be really useful (at least in terms of
> > rendering speed). However, converting a mesh to an isosurface seems to
> > be completely backwards. Why would one want to do that?
> Why wouldn't one want to do it?
I don't know... maybe because rendering an isosurface takes about one
hundred or even one thousand times more time than rendering a mesh? The
more complicated the mesh, the slower an equivalent isosurface would be.
> If it could be done efficiently, a
> person could give erosion to a statue, or add rusty lumps to an object,
> at will.
Those things can be done to a mesh, via subdivision and vertex
transformations. Of course this would require some coding, but it's
nevertheless possible.
--
- Warp
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Recently, I started building mesh2s following Earth's curvature from
> SRTM Earth elevation data (3 arcseconds per data point)... when
> rendering those mesh2s, I found some strange artifacts (gaps between
> triangles, see image on p.b.i.) which probably resulted from floating
> point inaccuracy (the scale I used is 1 POV unit = 1 kilometre).
Are your adjacent triangles referencing the same vertex, or are you
defining the coordinates repeatedly for each triangle?
> As Christoph Hormann, as he writes on his website, apparently uses a
> kind of (self-written?) conversion tool, I thought that converting my
> mesh2s to isosurfaces possible could solve the gaps problem...
It seems unlikely the isosurface rendering code would provide better
accuracy than the triangle rendering code, if indeed you are hitting
some limitation with POV. IIRC the triangle intersection code is only a
few calculations, whereas the isosurface intersection code is much more
complex (which means more calculations and more possibility of floating
point errors building up).
Post a reply to this message
|
|
| |
| |
|
|
From: Jörg 'Yadgar' Bleimann
Subject: Re: Mesh2 to isosurface - how to do with IsoCSG?
Date: 20 Dec 2010 08:00:48
Message: <4d0f5380@news.povray.org>
|
|
|
| |
| |
|
|
High!
On 20.12.2010 11:29, scott wrote:
> Are your adjacent triangles referencing the same vertex, or are you
> defining the coordinates repeatedly for each triangle?
No, each vertex is defined only one - as I use mesh2 rather than
classical mesh!
See you in Khyberspace!
Yadgar
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
