On Mon, 07 Feb 2000 16:21:37 -0500, Chris Huff wrote:
>Using more samples would smooth it out, and there might be ways to use 
>the samples other than just calculating the percentage that hit that 
>would give smoother results.

How about the percentage that hit the object within a specified 
distance?  If the object is a statue of a man on horseback and 
the point is in the center of the base, between the horse's legs,
lots of rays will hit the horse's body but the area is still quite
easy to clean.

-- 
These are my opinions.  I do NOT speak for the POV-Team.
The superpatch: http://www2.fwi.com/~parkerr/superpatch/
My other stuff: http://www2.fwi.com/~parkerr/traces.html

Post a reply to this message

In article <slr### [at] rongwmicrocom>, 
ron### [at] povrayorg wrote:

> How about the percentage that hit the object within a specified 
> distance?  If the object is a statue of a man on horseback and 
> the point is in the center of the base, between the horse's legs,
> lots of rays will hit the horse's body but the area is still quite
> easy to clean.

That would be one way, and would help reduce the effects of just a few 
rays that hit and give a wrong result. Another way would be to have the 
contribution of each test ray vary depending on the hit distance. Or to 
shoot rays in the general direction of the evaluation point from a 
position slightly "above" it, and use their distance and normals to 
calculate the curvature...

The first two would be easy to implement, I will work on them when I 
take a break from my particle simulator.

-- 
Chris Huff
e-mail: chr### [at] yahoocom
Web page: http://chrishuff.dhs.org/

Post a reply to this message

In article <389ff3c1@news.povray.org>, "Bob Hughes" 
<per### [at] aolcom?subject=PoV-News:> wrote:

> I can't really ask the right questions since I don't know how you're 
> doing this, and trying to explain it to me would perhaps be futile.

Well, I will try anyway.


> All I'm wondering though is if this "texture" were more a pattern 
> type than a sampling of a surface that it might be more linear I 
> guess would be the term. I'm just dreaming this up in a way that 
> would mean the texture would follow the contours of any object like a 
> uv map or something and without the sampling sort of stuff which 
> obviously is hit and miss to deduce where it goes.

It is a pattern, just like bozo, crackle, or wrinkles. It is calculated 
a bit differently from most, but is used in the same way any other 
pattern is used. It depends on the object, like slope.
At each point on the surface of the object, several rays are shot in all 
directions that don't go "down" relative to the normal at that point. 
They spread out in a half-sphere with the flat portion centered over the 
point. The percentage of the rays that hit is returned.

Another way to think of it is as a pattern which returns the amount of 
visible "sky". If the point is on the side of a box sitting on a plane, 
and very close to the ground plane, about half of it's "sky" will be 
covered by the plane. On a color_map ranging from black at 0 to white at 
1, it would be 50% gray. If the point is on the surface of a sphere, the 
whole sky will be visible, and the result will be white. If it is close 
to the point of contact between two spheres, only a narrow strip of sky 
will be visible, and you will get a dark gray.

It doesn't have anything to do with UV mapping, and I don't see how UV 
mapping could be applied to this problem, unless you mean by using an 
image_map or material_map to specify the areas with "patina".


> Plain and simple I can't understand the dynamics of your work anyhow, 
> so excuse me if I go way off track trying to figure it out.  What it 
> causes me to think is that if a object surface can be found and 
> mimicked in a changed way like this does do then it just seems then 
> that the texture could follow that surface and be applied as though a 
> typical patterned texture is with the usual way.

Are you thinking of a possible way to use this kind of thing to control 
UV mapping? So that the map is adjusted over the contours of the surface 
of the object?


> I think I see how you might not be following what I say, it reads to 
> me like round in circles.

I think I am starting to understand a little, although I am still unsure 
what exactly you mean.

-- 
Chris Huff
e-mail: chr### [at] yahoocom
Web page: http://chrishuff.dhs.org/

Post a reply to this message

"Chris Huff" <chr### [at] yahoocom> wrote in message
news:chrishuff_99-338EF3.16045808022000@news.povray.org...
|
| It doesn't have anything to do with UV mapping, and I don't see how UV
| mapping could be applied to this problem, unless you mean by using an
| image_map or material_map to specify the areas with "patina".
|
| Are you thinking of a possible way to use this kind of thing to control
| UV mapping? So that the map is adjusted over the contours of the surface
| of the object?

I was using the term uv map loosely to describe what I thought might closely
resemble what your stuff was doing.  That is, following the surfaces of objects
and their normals.  Again I say I'm not knowledgeable enough about any of it I
was just trying to get the sense of what was going on and if there were any
potential of it being more pigment pattern-like than it appeared to be.
Instead of scattering points on a surface maybe spreading the texture on like a
second skin.  If that makes any sense.  It's of course impossible if there isn't
any way to follow the original object precisely and deform it from there I
suppose.

Nice try at explaining it, I got closer than ever at understanding you.

Bob

Post a reply to this message