j/k! Heh, I guess You didn't believed it anyway!
Of course not displacement mapping, this is just example of
hypertextures (basically it is raymarching over sphere volume in order
to calculate normal and density). Shader is adapted from ARM ("Advanced
RenderMan", shaders from this book are available at
http://www.bmrt.org/arman/materials.html ), rendering time was around 5
minutes on 350 Mhz Pentium (I tried to render it with self-shadowing,
but it took too long, perhaps later).
Of course such shader is quite expensive in terms of time and for now it
has no practical use (isosurfaces are probably faster), but perhaps in
future, when computing power of Cray is on each table, it will be more
used.

Post a reply to this message

Attachments:
Download 'hypertex.jpg' (26 KB)

Preview of image 'hypertex.jpg'

i'm liking your examples of POVman.  keep it up since i don't have time
to try it out myself for the next couple months.  try to find things pov
can't do or can't do easily/well.

Vahur Krouverk wrote:
> 
> j/k! Heh, I guess You didn't believed it anyway!
> Of course not displacement mapping, this is just example of
> hypertextures (basically it is raymarching over sphere volume in order
> to calculate normal and density). Shader is adapted from ARM ("Advanced
> RenderMan", shaders from this book are available at
> http://www.bmrt.org/arman/materials.html ), rendering time was around 5
> minutes on 350 Mhz Pentium (I tried to render it with self-shadowing,
> but it took too long, perhaps later).
> Of course such shader is quite expensive in terms of time and for now it
> has no practical use (isosurfaces are probably faster), but perhaps in
> future, when computing power of Cray is on each table, it will be more
> used.
> 
>   ------------------------------------------------------------------------
>  [Image]

Post a reply to this message

In article <39F86CEE.962DE9F4@aetec.ee>, Vahur Krouverk 
<vah### [at] aetecee> wrote:

> j/k! Heh, I guess You didn't believed it anyway!
> Of course not displacement mapping, this is just example of
> hypertextures (basically it is raymarching over sphere volume in order
> to calculate normal and density).

Sounds like a shader that mostly imitates the media feature. Maybe media 
can be extended to calculate "surface" attributes like this...

-- 
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/

<><

Post a reply to this message

Vahur Krouverk wrote:
> 
> j/k! Heh, I guess You didn't believed it anyway!
> Of course not displacement mapping, this is just example of
> hypertextures (basically it is raymarching over sphere volume in order
> to calculate normal and density). Shader is adapted from ARM ("Advanced
> RenderMan", shaders from this book are available at
> http://www.bmrt.org/arman/materials.html ), rendering time was around 5
> minutes on 350 Mhz Pentium (I tried to render it with self-shadowing,
> but it took too long, perhaps later).
> Of course such shader is quite expensive in terms of time and for now it
> has no practical use (isosurfaces are probably faster), but perhaps in
> future, when computing power of Cray is on each table, it will be more
> used.
> 

I have not much idea how the shape looks like, because it appears very 'cloudy',
but those shaders seem an interesting thing.  How do they work combined with
regular pov-stuff ?

Christoph

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

Post a reply to this message

Chris Huff wrote:
> 
> In article <39F86CEE.962DE9F4@aetec.ee>, Vahur Krouverk
> <vah### [at] aetecee> wrote:
> 
> > j/k! Heh, I guess You didn't believed it anyway!
> > Of course not displacement mapping, this is just example of
> > hypertextures (basically it is raymarching over sphere volume in order
> > to calculate normal and density).
> 
> Sounds like a shader that mostly imitates the media feature. Maybe media
> can be extended to calculate "surface" attributes like this...
> 
Sounds (or looks?) correctly. Actually in ARM it was used to make
clouds, but I can't get those knobs to right positions in order to get
something, which looks like cloud from this book, instead I got this
picture...

Post a reply to this message

Christoph Hormann wrote:
> 
> I have not much idea how the shape looks like, because it appears very 'cloudy',
> but those shaders seem an interesting thing.  How do they work combined with
> regular pov-stuff ?

Shader in POV-Man is object's pigment and modifies pigment's color and
opacity (transparency), so it should work with regular "stuff" more or
less as expected. At least I hope so: I haven't tested it yet with
advanced features (such as multiple layered textures, media, photons,
glows etc.) and perhaps there are problems, lurking around.
Or did You meant something else?

Post a reply to this message

Vahur Krouverk wrote:
> 
> Shader in POV-Man is object's pigment and modifies pigment's color and
> opacity (transparency), so it should work with regular "stuff" more or
> less as expected. At least I hope so: I haven't tested it yet with
> advanced features (such as multiple layered textures, media, photons,
> glows etc.) and perhaps there are problems, lurking around.
> Or did You meant something else?

That's somehow what i meant, but as in your sample, it also seems to be capable
of surface perturbation and therefore i wonder how things work with CSG,
'trace', etc.

Christoph

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

Post a reply to this message

Christoph Hormann wrote:
> 
> Vahur Krouverk wrote:
> >
> > Shader in POV-Man is object's pigment and modifies pigment's color and
> > opacity (transparency), so it should work with regular "stuff" more or
> > less as expected. At least I hope so: I haven't tested it yet with
> > advanced features (such as multiple layered textures, media, photons,
> > glows etc.) and perhaps there are problems, lurking around.
> > Or did You meant something else?
> 
> That's somehow what i meant, but as in your sample, it also seems to be capable
> of surface perturbation and therefore i wonder how things work with CSG,
> 'trace', etc.
Well, forgot to mention, that shader can modify shading normal (i.e.
change normal's direction).
But in this hypertexture example it is just play with transparency and
color, shading calculation (for diffuse, highlights etc.) in shader uses
own calculated normal, object itself remains still sphere. I guess that
trace will detect as sphere, same with CSG. But such example is probably
not the "usual" use of shader, mostly it just modifies object pigment's
color.

Post a reply to this message